leandrof/linux
1
0
Fork 0
forked from Minki/linux
Code Issues Pull Requests Packages Projects Releases Wiki Activity
5be6ada39f
linux/drivers/acpi/acpi_lpss.c

961 lines
24 KiB
C
Raw Normal View History

ACPI / scan: Add special handler for Intel Lynxpoint LPSS devices Devices on the Intel Lynxpoint Low Power Subsystem (LPSS) have some common features that aren't shared with any other platform devices, including the clock and LTR (Latency Tolerance Reporting) registers. It is better to handle those features in common code than to bother device drivers with doing that (I/O functionality-wise the LPSS devices are generally compatible with other devices that don't have those special registers and may be handled by the same drivers). The clock registers of the LPSS devices are now taken care of by the special clk-x86-lpss driver, but the MMIO mappings used for accessing those registers can also be used for accessing the LTR registers on those devices (LTR support for the Lynxpoint LPSS is going to be added by a subsequent patch). Thus it is convenient to add a special ACPI scan handler for the Lynxpoint LPSS devices that will create the MMIO mappings for accessing the clock (and LTR in the future) registers and will register the LPSS devices' clocks, so the clk-x86-lpss driver will only need to take care of the main Lynxpoint LPSS clock. Introduce a special ACPI scan handler for Intel Lynxpoint LPSS devices as described above. This also reduces overhead related to browsing the ACPI namespace in search of the LPSS devices before the registration of their clocks, removes some LPSS-specific (and somewhat ugly) code from acpi_platform.c and shrinks the overall code size slightly. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Acked-by: Mike Turquette <mturquette@linaro.org>
2013-03-06 22:46:20 +00:00
/*
* ACPI support for Intel Lynxpoint LPSS.
*
Revert "ACPI / LPSS: introduce a 'proxy' device to power on LPSS for DMA" Revert commit 6c17ee44d524 (ACPI / LPSS: introduce a 'proxy' device to power on LPSS for DMA), as it introduced registration and probe ordering problems between devices on the LPSS that may lead to full hard system hang on boot in some cases.
2015-02-03 13:29:43 +00:00
* Copyright (C) 2013, Intel Corporation
ACPI / scan: Add special handler for Intel Lynxpoint LPSS devices Devices on the Intel Lynxpoint Low Power Subsystem (LPSS) have some common features that aren't shared with any other platform devices, including the clock and LTR (Latency Tolerance Reporting) registers. It is better to handle those features in common code than to bother device drivers with doing that (I/O functionality-wise the LPSS devices are generally compatible with other devices that don't have those special registers and may be handled by the same drivers). The clock registers of the LPSS devices are now taken care of by the special clk-x86-lpss driver, but the MMIO mappings used for accessing those registers can also be used for accessing the LTR registers on those devices (LTR support for the Lynxpoint LPSS is going to be added by a subsequent patch). Thus it is convenient to add a special ACPI scan handler for the Lynxpoint LPSS devices that will create the MMIO mappings for accessing the clock (and LTR in the future) registers and will register the LPSS devices' clocks, so the clk-x86-lpss driver will only need to take care of the main Lynxpoint LPSS clock. Introduce a special ACPI scan handler for Intel Lynxpoint LPSS devices as described above. This also reduces overhead related to browsing the ACPI namespace in search of the LPSS devices before the registration of their clocks, removes some LPSS-specific (and somewhat ugly) code from acpi_platform.c and shrinks the overall code size slightly. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Acked-by: Mike Turquette <mturquette@linaro.org>
2013-03-06 22:46:20 +00:00
* Authors: Mika Westerberg <mika.westerberg@linux.intel.com>
* Rafael J. Wysocki <rafael.j.wysocki@intel.com>
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License version 2 as
* published by the Free Software Foundation.
*/
#include <linux/acpi.h>
#include <linux/clkdev.h>
#include <linux/clk-provider.h>
#include <linux/err.h>
#include <linux/io.h>
ACPI / LPSS: override power state for LPSS DMA device This is a third approach to workaround long standing issue with LPSS on BayTrail. First one [1] was reverted since it didn't resolve the issue comprehensively. Second one [2] was rejected by internal review. The LPSS DMA controller does not have neither _PS0 nor _PS3 method. Moreover it can be powered off automatically whenever the last LPSS device goes down. In case of no power any access to the DMA controller will hang the system. The behaviour is reproduced on some HP laptops based on Intel BayTrail [3,4] as well as on ASuS T100TA transformer. Power on the LPSS island through the registers accessible in a specific way. [1] http://www.spinics.net/lists/linux-acpi/msg53963.html [2] https://bugzilla.redhat.com/attachment.cgi?id=1066779&action=diff [3] https://bugzilla.redhat.com/show_bug.cgi?id=1184273 [4] http://www.spinics.net/lists/dmaengine/msg01514.html Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-12-12 01:45:06 +00:00
#include <linux/mutex.h>
ACPI / scan: Add special handler for Intel Lynxpoint LPSS devices Devices on the Intel Lynxpoint Low Power Subsystem (LPSS) have some common features that aren't shared with any other platform devices, including the clock and LTR (Latency Tolerance Reporting) registers. It is better to handle those features in common code than to bother device drivers with doing that (I/O functionality-wise the LPSS devices are generally compatible with other devices that don't have those special registers and may be handled by the same drivers). The clock registers of the LPSS devices are now taken care of by the special clk-x86-lpss driver, but the MMIO mappings used for accessing those registers can also be used for accessing the LTR registers on those devices (LTR support for the Lynxpoint LPSS is going to be added by a subsequent patch). Thus it is convenient to add a special ACPI scan handler for the Lynxpoint LPSS devices that will create the MMIO mappings for accessing the clock (and LTR in the future) registers and will register the LPSS devices' clocks, so the clk-x86-lpss driver will only need to take care of the main Lynxpoint LPSS clock. Introduce a special ACPI scan handler for Intel Lynxpoint LPSS devices as described above. This also reduces overhead related to browsing the ACPI namespace in search of the LPSS devices before the registration of their clocks, removes some LPSS-specific (and somewhat ugly) code from acpi_platform.c and shrinks the overall code size slightly. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Acked-by: Mike Turquette <mturquette@linaro.org>
2013-03-06 22:46:20 +00:00
#include <linux/platform_device.h>
#include <linux/platform_data/clk-lpss.h>
PM / Domains: add setter for dev.pm_domain Adds a function that sets the pointer to dev_pm_domain in struct device and that warns if the device has already finished probing. The reason why we want to enforce that is because in the general case that can cause problems and also that we can simplify code quite a bit if we can always assume that. This patch also changes all current code that directly sets the dev.pm_domain pointer. Signed-off-by: Tomeu Vizoso <tomeu.vizoso@collabora.com> Reviewed-by: Ulf Hansson <ulf.hansson@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2016-01-07 15:46:13 +00:00
#include <linux/pm_domain.h>
ACPI / LPSS: Add support for exposing LTR registers to user space Devices on the Intel Lynxpoint Low Power Subsystem (LPSS) have registers providing access to LTR (Latency Tolerance Reporting) functionality that allows software to monitor and possibly influence the aggressiveness of the platform's active-state power management. For each LPSS device, there are two modes of operation related to LTR, the auto mode and the software mode. In the auto mode the LTR is set up by the platform firmware and managed by hardware. Software can only read the LTR register values to monitor the platform's behavior. In the software mode it is possible to use LTR to control the extent to which the platform will use its built-in power management features. This changeset adds support for reading the LPSS devices' LTR registers and exposing their values to user space for monitoring and diagnostics purposes. It re-uses the MMIO mappings created to access the LPSS devices' clock registers for reading the values of the LTR registers and exposes them to user space through sysfs device attributes. Namely, a new atrribute group, lpss_ltr, is created for each LPSS device. It contains three new attributes: ltr_mode, auto_ltr, sw_ltr. The value of the ltr_mode attribute reflects the LTR mode being used at the moment (software vs auto) and the other two contain the actual register values (raw) whose meaning depends on the LTR mode. All of these attributes are read-only. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2013-03-06 22:46:28 +00:00
#include <linux/pm_runtime.h>
ACPI / LPSS: custom power domain for LPSS A power domain where we save the context of the additional LPSS registers. We need to do this or all LPSS devices are left in reset state when resuming from D3 on some Baytrails. The devices with the fractional clock divider also have zeros for N and M values after resuming unless they are reset. Li Aubrey found the root cause for the issue. The idea of using power domain for LPSS came from Mika Westerberg. Reported-by: Jin Yao <yao.jin@linux.intel.com> Suggested-by: Li Aubrey <aubrey.li@linux.intel.com> Suggested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Tested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> [rjw: Added the .complete() callback to the PM domain, fixed build warning on 32-bit.] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-05-23 13:15:09 +00:00
#include <linux/delay.h>
ACPI / scan: Add special handler for Intel Lynxpoint LPSS devices Devices on the Intel Lynxpoint Low Power Subsystem (LPSS) have some common features that aren't shared with any other platform devices, including the clock and LTR (Latency Tolerance Reporting) registers. It is better to handle those features in common code than to bother device drivers with doing that (I/O functionality-wise the LPSS devices are generally compatible with other devices that don't have those special registers and may be handled by the same drivers). The clock registers of the LPSS devices are now taken care of by the special clk-x86-lpss driver, but the MMIO mappings used for accessing those registers can also be used for accessing the LTR registers on those devices (LTR support for the Lynxpoint LPSS is going to be added by a subsequent patch). Thus it is convenient to add a special ACPI scan handler for the Lynxpoint LPSS devices that will create the MMIO mappings for accessing the clock (and LTR in the future) registers and will register the LPSS devices' clocks, so the clk-x86-lpss driver will only need to take care of the main Lynxpoint LPSS clock. Introduce a special ACPI scan handler for Intel Lynxpoint LPSS devices as described above. This also reduces overhead related to browsing the ACPI namespace in search of the LPSS devices before the registration of their clocks, removes some LPSS-specific (and somewhat ugly) code from acpi_platform.c and shrinks the overall code size slightly. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Acked-by: Mike Turquette <mturquette@linaro.org>
2013-03-06 22:46:20 +00:00
#include "internal.h"
ACPI_MODULE_NAME("acpi_lpss");
ACPI / scan: always register ACPI LPSS scan handler Prevent platform devices from being created for ACPI LPSS devices if CONFIG_X86_INTEL_LPSS is unset by compiling out the LPSS scan handler's callbacks only in that case and still compiling its device ID list in and registering the scan handler in either case. This change is based on a prototype from Zhang Rui. Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Reviewed-by: Mika Westerberg <mika.westerberg@linux.intel.com>
2014-05-30 12:34:05 +00:00
#ifdef CONFIG_X86_INTEL_LPSS
ACPI / LPSS: override power state for LPSS DMA device This is a third approach to workaround long standing issue with LPSS on BayTrail. First one [1] was reverted since it didn't resolve the issue comprehensively. Second one [2] was rejected by internal review. The LPSS DMA controller does not have neither _PS0 nor _PS3 method. Moreover it can be powered off automatically whenever the last LPSS device goes down. In case of no power any access to the DMA controller will hang the system. The behaviour is reproduced on some HP laptops based on Intel BayTrail [3,4] as well as on ASuS T100TA transformer. Power on the LPSS island through the registers accessible in a specific way. [1] http://www.spinics.net/lists/linux-acpi/msg53963.html [2] https://bugzilla.redhat.com/attachment.cgi?id=1066779&action=diff [3] https://bugzilla.redhat.com/show_bug.cgi?id=1184273 [4] http://www.spinics.net/lists/dmaengine/msg01514.html Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-12-12 01:45:06 +00:00
#include <asm/cpu_device_id.h>
#include <asm/iosf_mbi.h>
#include <asm/pmc_atom.h>
ACPI / scan: always register ACPI LPSS scan handler Prevent platform devices from being created for ACPI LPSS devices if CONFIG_X86_INTEL_LPSS is unset by compiling out the LPSS scan handler's callbacks only in that case and still compiling its device ID list in and registering the scan handler in either case. This change is based on a prototype from Zhang Rui. Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Reviewed-by: Mika Westerberg <mika.westerberg@linux.intel.com>
2014-05-30 12:34:05 +00:00
#define LPSS_ADDR(desc) ((unsigned long)&desc)
ACPI / scan: Add special handler for Intel Lynxpoint LPSS devices Devices on the Intel Lynxpoint Low Power Subsystem (LPSS) have some common features that aren't shared with any other platform devices, including the clock and LTR (Latency Tolerance Reporting) registers. It is better to handle those features in common code than to bother device drivers with doing that (I/O functionality-wise the LPSS devices are generally compatible with other devices that don't have those special registers and may be handled by the same drivers). The clock registers of the LPSS devices are now taken care of by the special clk-x86-lpss driver, but the MMIO mappings used for accessing those registers can also be used for accessing the LTR registers on those devices (LTR support for the Lynxpoint LPSS is going to be added by a subsequent patch). Thus it is convenient to add a special ACPI scan handler for the Lynxpoint LPSS devices that will create the MMIO mappings for accessing the clock (and LTR in the future) registers and will register the LPSS devices' clocks, so the clk-x86-lpss driver will only need to take care of the main Lynxpoint LPSS clock. Introduce a special ACPI scan handler for Intel Lynxpoint LPSS devices as described above. This also reduces overhead related to browsing the ACPI namespace in search of the LPSS devices before the registration of their clocks, removes some LPSS-specific (and somewhat ugly) code from acpi_platform.c and shrinks the overall code size slightly. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Acked-by: Mike Turquette <mturquette@linaro.org>
2013-03-06 22:46:20 +00:00
#define LPSS_CLK_SIZE 0x04
ACPI / LPSS: Add support for exposing LTR registers to user space Devices on the Intel Lynxpoint Low Power Subsystem (LPSS) have registers providing access to LTR (Latency Tolerance Reporting) functionality that allows software to monitor and possibly influence the aggressiveness of the platform's active-state power management. For each LPSS device, there are two modes of operation related to LTR, the auto mode and the software mode. In the auto mode the LTR is set up by the platform firmware and managed by hardware. Software can only read the LTR register values to monitor the platform's behavior. In the software mode it is possible to use LTR to control the extent to which the platform will use its built-in power management features. This changeset adds support for reading the LPSS devices' LTR registers and exposing their values to user space for monitoring and diagnostics purposes. It re-uses the MMIO mappings created to access the LPSS devices' clock registers for reading the values of the LTR registers and exposes them to user space through sysfs device attributes. Namely, a new atrribute group, lpss_ltr, is created for each LPSS device. It contains three new attributes: ltr_mode, auto_ltr, sw_ltr. The value of the ltr_mode attribute reflects the LTR mode being used at the moment (software vs auto) and the other two contain the actual register values (raw) whose meaning depends on the LTR mode. All of these attributes are read-only. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2013-03-06 22:46:28 +00:00
#define LPSS_LTR_SIZE 0x18
/* Offsets relative to LPSS_PRIVATE_OFFSET */
ACPI / LPSS: support for fractional divider clock This creates fractional divider type clock for the ones that have it. It is needed by the UART driver as the clock rate must accommodate to the requested baud rate. Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-05-19 11:42:07 +00:00
#define LPSS_CLK_DIVIDER_DEF_MASK (BIT(1) | BIT(16))
ACPI / LPSS: Take I2C host controllers out of reset On Intel Baytrail, some I2C host controllers are held in reset when the OS gets control. This causes the driver to fail to detect the hardware properly. Fix this so that we make sure that the I2C host controller is not in reset when the driver gets probe'd. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-06-17 11:33:39 +00:00
#define LPSS_RESETS 0x04
#define LPSS_RESETS_RESET_FUNC BIT(0)
#define LPSS_RESETS_RESET_APB BIT(1)
ACPI / LPSS: Add support for exposing LTR registers to user space Devices on the Intel Lynxpoint Low Power Subsystem (LPSS) have registers providing access to LTR (Latency Tolerance Reporting) functionality that allows software to monitor and possibly influence the aggressiveness of the platform's active-state power management. For each LPSS device, there are two modes of operation related to LTR, the auto mode and the software mode. In the auto mode the LTR is set up by the platform firmware and managed by hardware. Software can only read the LTR register values to monitor the platform's behavior. In the software mode it is possible to use LTR to control the extent to which the platform will use its built-in power management features. This changeset adds support for reading the LPSS devices' LTR registers and exposing their values to user space for monitoring and diagnostics purposes. It re-uses the MMIO mappings created to access the LPSS devices' clock registers for reading the values of the LTR registers and exposes them to user space through sysfs device attributes. Namely, a new atrribute group, lpss_ltr, is created for each LPSS device. It contains three new attributes: ltr_mode, auto_ltr, sw_ltr. The value of the ltr_mode attribute reflects the LTR mode being used at the moment (software vs auto) and the other two contain the actual register values (raw) whose meaning depends on the LTR mode. All of these attributes are read-only. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2013-03-06 22:46:28 +00:00
#define LPSS_GENERAL 0x08
#define LPSS_GENERAL_LTR_MODE_SW BIT(2)
ACPI / LPSS: fix UART Auto Flow Control There is an additional bit in the GENERAL register on newer silicon that needs to be set or UART's RTS pin fails to reflect the flow control settings in the Modem Control Register. This will fix an issue where the RTS pin of the UART stays always at 1.8V, regardless of the register settings. Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2013-10-09 06:49:20 +00:00
#define LPSS_GENERAL_UART_RTS_OVRD BIT(3)
ACPI / LPSS: Add support for exposing LTR registers to user space Devices on the Intel Lynxpoint Low Power Subsystem (LPSS) have registers providing access to LTR (Latency Tolerance Reporting) functionality that allows software to monitor and possibly influence the aggressiveness of the platform's active-state power management. For each LPSS device, there are two modes of operation related to LTR, the auto mode and the software mode. In the auto mode the LTR is set up by the platform firmware and managed by hardware. Software can only read the LTR register values to monitor the platform's behavior. In the software mode it is possible to use LTR to control the extent to which the platform will use its built-in power management features. This changeset adds support for reading the LPSS devices' LTR registers and exposing their values to user space for monitoring and diagnostics purposes. It re-uses the MMIO mappings created to access the LPSS devices' clock registers for reading the values of the LTR registers and exposes them to user space through sysfs device attributes. Namely, a new atrribute group, lpss_ltr, is created for each LPSS device. It contains three new attributes: ltr_mode, auto_ltr, sw_ltr. The value of the ltr_mode attribute reflects the LTR mode being used at the moment (software vs auto) and the other two contain the actual register values (raw) whose meaning depends on the LTR mode. All of these attributes are read-only. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2013-03-06 22:46:28 +00:00
#define LPSS_SW_LTR 0x10
#define LPSS_AUTO_LTR 0x14
ACPI / LPSS: Support for device latency tolerance PM QoS Add a new routine, acpi_lpss_set_ltr(), for setting latency tolerance values for LPSS devices having LTR (Latency Tolerance Reporting) registers. Add .bind()/.unbind() callbacks to lpss_handler to set the LPSS devices' power.set_latency_tolerance callback pointers to acpi_lpss_set_ltr() during device addition and to clear them on device removal, respectively. That will cause the device latency tolerance PM QoS to work for the devices in question as documented. This changeset includes a fix from Mika Westerberg. Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-02-10 23:35:53 +00:00
#define LPSS_LTR_SNOOP_REQ BIT(15)
#define LPSS_LTR_SNOOP_MASK 0x0000FFFF
#define LPSS_LTR_SNOOP_LAT_1US 0x800
#define LPSS_LTR_SNOOP_LAT_32US 0xC00
#define LPSS_LTR_SNOOP_LAT_SHIFT 5
#define LPSS_LTR_SNOOP_LAT_CUTOFF 3000
#define LPSS_LTR_MAX_VAL 0x3FF
ACPI / LPSS: mask the UART TX completion interrupt Intel LPSS provides an extra TX byte counter and an extra TX completion interrupt for some of its bus controllers. However, there is no use for the extra UART interrupt and it has to be masked out during initialization. Otherwise, if the firmware does not mask the interrupt and the driver does not clear it, it may cause an interrupt flood freezing the board to happen. Add code masking that problematic interrupt to the ACPI LPSS driver. [rjw: Changelog] Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2013-06-17 10:25:46 +00:00
#define LPSS_TX_INT 0x20
#define LPSS_TX_INT_MASK BIT(1)
ACPI / scan: Add special handler for Intel Lynxpoint LPSS devices Devices on the Intel Lynxpoint Low Power Subsystem (LPSS) have some common features that aren't shared with any other platform devices, including the clock and LTR (Latency Tolerance Reporting) registers. It is better to handle those features in common code than to bother device drivers with doing that (I/O functionality-wise the LPSS devices are generally compatible with other devices that don't have those special registers and may be handled by the same drivers). The clock registers of the LPSS devices are now taken care of by the special clk-x86-lpss driver, but the MMIO mappings used for accessing those registers can also be used for accessing the LTR registers on those devices (LTR support for the Lynxpoint LPSS is going to be added by a subsequent patch). Thus it is convenient to add a special ACPI scan handler for the Lynxpoint LPSS devices that will create the MMIO mappings for accessing the clock (and LTR in the future) registers and will register the LPSS devices' clocks, so the clk-x86-lpss driver will only need to take care of the main Lynxpoint LPSS clock. Introduce a special ACPI scan handler for Intel Lynxpoint LPSS devices as described above. This also reduces overhead related to browsing the ACPI namespace in search of the LPSS devices before the registration of their clocks, removes some LPSS-specific (and somewhat ugly) code from acpi_platform.c and shrinks the overall code size slightly. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Acked-by: Mike Turquette <mturquette@linaro.org>
2013-03-06 22:46:20 +00:00
ACPI / LPSS: custom power domain for LPSS A power domain where we save the context of the additional LPSS registers. We need to do this or all LPSS devices are left in reset state when resuming from D3 on some Baytrails. The devices with the fractional clock divider also have zeros for N and M values after resuming unless they are reset. Li Aubrey found the root cause for the issue. The idea of using power domain for LPSS came from Mika Westerberg. Reported-by: Jin Yao <yao.jin@linux.intel.com> Suggested-by: Li Aubrey <aubrey.li@linux.intel.com> Suggested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Tested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> [rjw: Added the .complete() callback to the PM domain, fixed build warning on 32-bit.] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-05-23 13:15:09 +00:00
#define LPSS_PRV_REG_COUNT 9
ACPI / LPSS: introduce flags Replace the booleans with a single flags member variable. Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Acked-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-09-02 07:55:07 +00:00
/* LPSS Flags */
#define LPSS_CLK BIT(0)
#define LPSS_CLK_GATE BIT(1)
#define LPSS_CLK_DIVIDER BIT(2)
#define LPSS_LTR BIT(3)
#define LPSS_SAVE_CTX BIT(4)
ACPI / LPSS: Ignore 10ms delay for Braswell LPSS devices in Braswell does not need the default 10ms d3_delay imposed by PCI specification. Removing this unnecessary delay significantly reduces the resume time approximately upto 200ms on this platform. Signed-off-by: Srinidhi Kasagar <srinidhi.kasagar@intel.com> Acked-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-08-27 16:00:55 +00:00
#define LPSS_NO_D3_DELAY BIT(5)
ACPI / LPSS: add support for Intel BayTrail Intel BayTrail has almost the same Low Power Subsystem than Lynxpoint with few differences. Peripherals are clocked with different speeds (typically lower) and the clock is not always gated. To support this we add possibility to share a common fixed rate clock and make clock gating optional. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Acked-by: Mike Turquette <mturquette@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2013-05-13 12:42:44 +00:00
ACPI / LPSS: mask the UART TX completion interrupt Intel LPSS provides an extra TX byte counter and an extra TX completion interrupt for some of its bus controllers. However, there is no use for the extra UART interrupt and it has to be masked out during initialization. Otherwise, if the firmware does not mask the interrupt and the driver does not clear it, it may cause an interrupt flood freezing the board to happen. Add code masking that problematic interrupt to the ACPI LPSS driver. [rjw: Changelog] Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2013-06-17 10:25:46 +00:00
struct lpss_private_data;
ACPI / scan: Add special handler for Intel Lynxpoint LPSS devices Devices on the Intel Lynxpoint Low Power Subsystem (LPSS) have some common features that aren't shared with any other platform devices, including the clock and LTR (Latency Tolerance Reporting) registers. It is better to handle those features in common code than to bother device drivers with doing that (I/O functionality-wise the LPSS devices are generally compatible with other devices that don't have those special registers and may be handled by the same drivers). The clock registers of the LPSS devices are now taken care of by the special clk-x86-lpss driver, but the MMIO mappings used for accessing those registers can also be used for accessing the LTR registers on those devices (LTR support for the Lynxpoint LPSS is going to be added by a subsequent patch). Thus it is convenient to add a special ACPI scan handler for the Lynxpoint LPSS devices that will create the MMIO mappings for accessing the clock (and LTR in the future) registers and will register the LPSS devices' clocks, so the clk-x86-lpss driver will only need to take care of the main Lynxpoint LPSS clock. Introduce a special ACPI scan handler for Intel Lynxpoint LPSS devices as described above. This also reduces overhead related to browsing the ACPI namespace in search of the LPSS devices before the registration of their clocks, removes some LPSS-specific (and somewhat ugly) code from acpi_platform.c and shrinks the overall code size slightly. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Acked-by: Mike Turquette <mturquette@linaro.org>
2013-03-06 22:46:20 +00:00
struct lpss_device_desc {
ACPI / LPSS: introduce flags Replace the booleans with a single flags member variable. Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Acked-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-09-02 07:55:07 +00:00
unsigned int flags;
ACPI / LPSS: provide con_id for the clkdev Commit 7d78cbefaa (serial: 8250_dw: add ability to handle the peripheral clock) introduces handling for a second clk to 8250_dw.c which is the driver also for LPSS UART. The second clk forces us to provide identifier (con_id) for the clkdev we create. This fixes an issue where 8250_dw.c is getting the same handler for both clocks. Fixes: 7d78cbefaa (serial: 8250_dw: add ability to handle the peripheral clock) Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Cc: 3.17+ <stable@vger.kernel.org> # 3.17+ Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-03-06 13:48:38 +00:00
const char *clk_con_id;
ACPI / LPSS: Add support for exposing LTR registers to user space Devices on the Intel Lynxpoint Low Power Subsystem (LPSS) have registers providing access to LTR (Latency Tolerance Reporting) functionality that allows software to monitor and possibly influence the aggressiveness of the platform's active-state power management. For each LPSS device, there are two modes of operation related to LTR, the auto mode and the software mode. In the auto mode the LTR is set up by the platform firmware and managed by hardware. Software can only read the LTR register values to monitor the platform's behavior. In the software mode it is possible to use LTR to control the extent to which the platform will use its built-in power management features. This changeset adds support for reading the LPSS devices' LTR registers and exposing their values to user space for monitoring and diagnostics purposes. It re-uses the MMIO mappings created to access the LPSS devices' clock registers for reading the values of the LTR registers and exposes them to user space through sysfs device attributes. Namely, a new atrribute group, lpss_ltr, is created for each LPSS device. It contains three new attributes: ltr_mode, auto_ltr, sw_ltr. The value of the ltr_mode attribute reflects the LTR mode being used at the moment (software vs auto) and the other two contain the actual register values (raw) whose meaning depends on the LTR mode. All of these attributes are read-only. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2013-03-06 22:46:28 +00:00
unsigned int prv_offset;
ACPI / LPSS: override SDIO private register space size from ACPI tables The SDIO device in Lynxpoint has its LTR registers reserved for a WiFi device (a child of the SDIO device) in the ACPI namespace even though those registers physically belong to the SDIO device itself. In order to be able to access the SDIO LTR registers from the ACPI LPSS driver for diagnostic purposes we need to use a size override for the SDIO private register space. Add a possibility to override the size of the private register space of an LPSS device provided by the ACPI tables in the ACPI LPSS driver and set the correct size for the SDIO device in there. [rjw: Changelog] Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2013-06-18 13:51:35 +00:00
size_t prv_size_override;
ACPI / LPSS: mask the UART TX completion interrupt Intel LPSS provides an extra TX byte counter and an extra TX completion interrupt for some of its bus controllers. However, there is no use for the extra UART interrupt and it has to be masked out during initialization. Otherwise, if the firmware does not mask the interrupt and the driver does not clear it, it may cause an interrupt flood freezing the board to happen. Add code masking that problematic interrupt to the ACPI LPSS driver. [rjw: Changelog] Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2013-06-17 10:25:46 +00:00
void (*setup)(struct lpss_private_data *pdata);
ACPI / scan: Add special handler for Intel Lynxpoint LPSS devices Devices on the Intel Lynxpoint Low Power Subsystem (LPSS) have some common features that aren't shared with any other platform devices, including the clock and LTR (Latency Tolerance Reporting) registers. It is better to handle those features in common code than to bother device drivers with doing that (I/O functionality-wise the LPSS devices are generally compatible with other devices that don't have those special registers and may be handled by the same drivers). The clock registers of the LPSS devices are now taken care of by the special clk-x86-lpss driver, but the MMIO mappings used for accessing those registers can also be used for accessing the LTR registers on those devices (LTR support for the Lynxpoint LPSS is going to be added by a subsequent patch). Thus it is convenient to add a special ACPI scan handler for the Lynxpoint LPSS devices that will create the MMIO mappings for accessing the clock (and LTR in the future) registers and will register the LPSS devices' clocks, so the clk-x86-lpss driver will only need to take care of the main Lynxpoint LPSS clock. Introduce a special ACPI scan handler for Intel Lynxpoint LPSS devices as described above. This also reduces overhead related to browsing the ACPI namespace in search of the LPSS devices before the registration of their clocks, removes some LPSS-specific (and somewhat ugly) code from acpi_platform.c and shrinks the overall code size slightly. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Acked-by: Mike Turquette <mturquette@linaro.org>
2013-03-06 22:46:20 +00:00
};
ACPI / LPSS: override power state for LPSS DMA device This is a third approach to workaround long standing issue with LPSS on BayTrail. First one [1] was reverted since it didn't resolve the issue comprehensively. Second one [2] was rejected by internal review. The LPSS DMA controller does not have neither _PS0 nor _PS3 method. Moreover it can be powered off automatically whenever the last LPSS device goes down. In case of no power any access to the DMA controller will hang the system. The behaviour is reproduced on some HP laptops based on Intel BayTrail [3,4] as well as on ASuS T100TA transformer. Power on the LPSS island through the registers accessible in a specific way. [1] http://www.spinics.net/lists/linux-acpi/msg53963.html [2] https://bugzilla.redhat.com/attachment.cgi?id=1066779&action=diff [3] https://bugzilla.redhat.com/show_bug.cgi?id=1184273 [4] http://www.spinics.net/lists/dmaengine/msg01514.html Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-12-12 01:45:06 +00:00
static const struct lpss_device_desc lpss_dma_desc = {
Revert "ACPI / LPSS: introduce a 'proxy' device to power on LPSS for DMA" Revert commit 6c17ee44d524 (ACPI / LPSS: introduce a 'proxy' device to power on LPSS for DMA), as it introduced registration and probe ordering problems between devices on the LPSS that may lead to full hard system hang on boot in some cases.
2015-02-03 13:29:43 +00:00
.flags = LPSS_CLK,
ACPI / LPSS: register clock device for Lynxpoint DMA properly The DMA controller in Lynxpoint is enumerated as a regular ACPI device now. To work properly it is using the LPSS root clock as a functional clock. That's why we have to register the clock device accordingly to the ACPI ID of the DMA controller. The acpi_lpss.c module is responsible to do the job. This patch also removes hardcoded name of the DMA device in clk-lpt.c and the name of the root clock in acpi_lpss.c. Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Signed-off-by: Vinod Koul <vinod.koul@intel.com>
2013-05-08 08:55:49 +00:00
};
ACPI / scan: Add special handler for Intel Lynxpoint LPSS devices Devices on the Intel Lynxpoint Low Power Subsystem (LPSS) have some common features that aren't shared with any other platform devices, including the clock and LTR (Latency Tolerance Reporting) registers. It is better to handle those features in common code than to bother device drivers with doing that (I/O functionality-wise the LPSS devices are generally compatible with other devices that don't have those special registers and may be handled by the same drivers). The clock registers of the LPSS devices are now taken care of by the special clk-x86-lpss driver, but the MMIO mappings used for accessing those registers can also be used for accessing the LTR registers on those devices (LTR support for the Lynxpoint LPSS is going to be added by a subsequent patch). Thus it is convenient to add a special ACPI scan handler for the Lynxpoint LPSS devices that will create the MMIO mappings for accessing the clock (and LTR in the future) registers and will register the LPSS devices' clocks, so the clk-x86-lpss driver will only need to take care of the main Lynxpoint LPSS clock. Introduce a special ACPI scan handler for Intel Lynxpoint LPSS devices as described above. This also reduces overhead related to browsing the ACPI namespace in search of the LPSS devices before the registration of their clocks, removes some LPSS-specific (and somewhat ugly) code from acpi_platform.c and shrinks the overall code size slightly. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Acked-by: Mike Turquette <mturquette@linaro.org>
2013-03-06 22:46:20 +00:00
struct lpss_private_data {
void __iomem *mmio_base;
resource_size_t mmio_size;
ACPI / LPSS: support for 133MHz I2C source clock on Baytrail The I2C controllers on Baytrail can get the clock from 100MHz or 133MHz source clock. The first bits in the private clock parameter register indicates which one is being used. Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Acked-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-09-02 07:55:09 +00:00
unsigned int fixed_clk_rate;
ACPI / scan: Add special handler for Intel Lynxpoint LPSS devices Devices on the Intel Lynxpoint Low Power Subsystem (LPSS) have some common features that aren't shared with any other platform devices, including the clock and LTR (Latency Tolerance Reporting) registers. It is better to handle those features in common code than to bother device drivers with doing that (I/O functionality-wise the LPSS devices are generally compatible with other devices that don't have those special registers and may be handled by the same drivers). The clock registers of the LPSS devices are now taken care of by the special clk-x86-lpss driver, but the MMIO mappings used for accessing those registers can also be used for accessing the LTR registers on those devices (LTR support for the Lynxpoint LPSS is going to be added by a subsequent patch). Thus it is convenient to add a special ACPI scan handler for the Lynxpoint LPSS devices that will create the MMIO mappings for accessing the clock (and LTR in the future) registers and will register the LPSS devices' clocks, so the clk-x86-lpss driver will only need to take care of the main Lynxpoint LPSS clock. Introduce a special ACPI scan handler for Intel Lynxpoint LPSS devices as described above. This also reduces overhead related to browsing the ACPI namespace in search of the LPSS devices before the registration of their clocks, removes some LPSS-specific (and somewhat ugly) code from acpi_platform.c and shrinks the overall code size slightly. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Acked-by: Mike Turquette <mturquette@linaro.org>
2013-03-06 22:46:20 +00:00
struct clk *clk;
const struct lpss_device_desc *dev_desc;
ACPI / LPSS: custom power domain for LPSS A power domain where we save the context of the additional LPSS registers. We need to do this or all LPSS devices are left in reset state when resuming from D3 on some Baytrails. The devices with the fractional clock divider also have zeros for N and M values after resuming unless they are reset. Li Aubrey found the root cause for the issue. The idea of using power domain for LPSS came from Mika Westerberg. Reported-by: Jin Yao <yao.jin@linux.intel.com> Suggested-by: Li Aubrey <aubrey.li@linux.intel.com> Suggested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Tested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> [rjw: Added the .complete() callback to the PM domain, fixed build warning on 32-bit.] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-05-23 13:15:09 +00:00
u32 prv_reg_ctx[LPSS_PRV_REG_COUNT];
ACPI / scan: Add special handler for Intel Lynxpoint LPSS devices Devices on the Intel Lynxpoint Low Power Subsystem (LPSS) have some common features that aren't shared with any other platform devices, including the clock and LTR (Latency Tolerance Reporting) registers. It is better to handle those features in common code than to bother device drivers with doing that (I/O functionality-wise the LPSS devices are generally compatible with other devices that don't have those special registers and may be handled by the same drivers). The clock registers of the LPSS devices are now taken care of by the special clk-x86-lpss driver, but the MMIO mappings used for accessing those registers can also be used for accessing the LTR registers on those devices (LTR support for the Lynxpoint LPSS is going to be added by a subsequent patch). Thus it is convenient to add a special ACPI scan handler for the Lynxpoint LPSS devices that will create the MMIO mappings for accessing the clock (and LTR in the future) registers and will register the LPSS devices' clocks, so the clk-x86-lpss driver will only need to take care of the main Lynxpoint LPSS clock. Introduce a special ACPI scan handler for Intel Lynxpoint LPSS devices as described above. This also reduces overhead related to browsing the ACPI namespace in search of the LPSS devices before the registration of their clocks, removes some LPSS-specific (and somewhat ugly) code from acpi_platform.c and shrinks the overall code size slightly. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Acked-by: Mike Turquette <mturquette@linaro.org>
2013-03-06 22:46:20 +00:00
};
ACPI / LPSS: override power state for LPSS DMA device This is a third approach to workaround long standing issue with LPSS on BayTrail. First one [1] was reverted since it didn't resolve the issue comprehensively. Second one [2] was rejected by internal review. The LPSS DMA controller does not have neither _PS0 nor _PS3 method. Moreover it can be powered off automatically whenever the last LPSS device goes down. In case of no power any access to the DMA controller will hang the system. The behaviour is reproduced on some HP laptops based on Intel BayTrail [3,4] as well as on ASuS T100TA transformer. Power on the LPSS island through the registers accessible in a specific way. [1] http://www.spinics.net/lists/linux-acpi/msg53963.html [2] https://bugzilla.redhat.com/attachment.cgi?id=1066779&action=diff [3] https://bugzilla.redhat.com/show_bug.cgi?id=1184273 [4] http://www.spinics.net/lists/dmaengine/msg01514.html Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-12-12 01:45:06 +00:00
/* LPSS run time quirks */
static unsigned int lpss_quirks;
/*
* LPSS_QUIRK_ALWAYS_POWER_ON: override power state for LPSS DMA device.
*
ACPI / LPSS: change 'does not have' to 'has' in comment The LPSS DMA device has neither _PS0 nor _PS3 method. Fix the wording in the comment line. Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-12-21 20:31:09 +00:00
* The LPSS DMA controller has neither _PS0 nor _PS3 method. Moreover
ACPI / LPSS: override power state for LPSS DMA device This is a third approach to workaround long standing issue with LPSS on BayTrail. First one [1] was reverted since it didn't resolve the issue comprehensively. Second one [2] was rejected by internal review. The LPSS DMA controller does not have neither _PS0 nor _PS3 method. Moreover it can be powered off automatically whenever the last LPSS device goes down. In case of no power any access to the DMA controller will hang the system. The behaviour is reproduced on some HP laptops based on Intel BayTrail [3,4] as well as on ASuS T100TA transformer. Power on the LPSS island through the registers accessible in a specific way. [1] http://www.spinics.net/lists/linux-acpi/msg53963.html [2] https://bugzilla.redhat.com/attachment.cgi?id=1066779&action=diff [3] https://bugzilla.redhat.com/show_bug.cgi?id=1184273 [4] http://www.spinics.net/lists/dmaengine/msg01514.html Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-12-12 01:45:06 +00:00
* it can be powered off automatically whenever the last LPSS device goes down.
* In case of no power any access to the DMA controller will hang the system.
* The behaviour is reproduced on some HP laptops based on Intel BayTrail as
* well as on ASuS T100TA transformer.
*
* This quirk overrides power state of entire LPSS island to keep DMA powered
* on whenever we have at least one other device in use.
*/
#define LPSS_QUIRK_ALWAYS_POWER_ON BIT(0)
ACPI / LPSS: not using UART RTS override with Auto Flow Control Adding a check for UART Auto Flow Control feature and only enabling the RTS override when it's not supported. RTS override is not needed when Auto Flow Control is used and they shouldn't be used together. Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-09-11 12:19:33 +00:00
/* UART Component Parameter Register */
#define LPSS_UART_CPR 0xF4
#define LPSS_UART_CPR_AFCE BIT(4)
ACPI / LPSS: mask the UART TX completion interrupt Intel LPSS provides an extra TX byte counter and an extra TX completion interrupt for some of its bus controllers. However, there is no use for the extra UART interrupt and it has to be masked out during initialization. Otherwise, if the firmware does not mask the interrupt and the driver does not clear it, it may cause an interrupt flood freezing the board to happen. Add code masking that problematic interrupt to the ACPI LPSS driver. [rjw: Changelog] Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2013-06-17 10:25:46 +00:00
static void lpss_uart_setup(struct lpss_private_data *pdata)
{
ACPI / LPSS: fix UART Auto Flow Control There is an additional bit in the GENERAL register on newer silicon that needs to be set or UART's RTS pin fails to reflect the flow control settings in the Modem Control Register. This will fix an issue where the RTS pin of the UART stays always at 1.8V, regardless of the register settings. Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2013-10-09 06:49:20 +00:00
unsigned int offset;
ACPI / LPSS: not using UART RTS override with Auto Flow Control Adding a check for UART Auto Flow Control feature and only enabling the RTS override when it's not supported. RTS override is not needed when Auto Flow Control is used and they shouldn't be used together. Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-09-11 12:19:33 +00:00
u32 val;
ACPI / LPSS: mask the UART TX completion interrupt Intel LPSS provides an extra TX byte counter and an extra TX completion interrupt for some of its bus controllers. However, there is no use for the extra UART interrupt and it has to be masked out during initialization. Otherwise, if the firmware does not mask the interrupt and the driver does not clear it, it may cause an interrupt flood freezing the board to happen. Add code masking that problematic interrupt to the ACPI LPSS driver. [rjw: Changelog] Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2013-06-17 10:25:46 +00:00
ACPI / LPSS: fix UART Auto Flow Control There is an additional bit in the GENERAL register on newer silicon that needs to be set or UART's RTS pin fails to reflect the flow control settings in the Modem Control Register. This will fix an issue where the RTS pin of the UART stays always at 1.8V, regardless of the register settings. Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2013-10-09 06:49:20 +00:00
offset = pdata->dev_desc->prv_offset + LPSS_TX_INT;
ACPI / LPSS: not using UART RTS override with Auto Flow Control Adding a check for UART Auto Flow Control feature and only enabling the RTS override when it's not supported. RTS override is not needed when Auto Flow Control is used and they shouldn't be used together. Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-09-11 12:19:33 +00:00
val = readl(pdata->mmio_base + offset);
writel(val | LPSS_TX_INT_MASK, pdata->mmio_base + offset);
val = readl(pdata->mmio_base + LPSS_UART_CPR);
if (!(val & LPSS_UART_CPR_AFCE)) {
offset = pdata->dev_desc->prv_offset + LPSS_GENERAL;
val = readl(pdata->mmio_base + offset);
val |= LPSS_GENERAL_UART_RTS_OVRD;
writel(val, pdata->mmio_base + offset);
}
ACPI / LPSS: mask the UART TX completion interrupt Intel LPSS provides an extra TX byte counter and an extra TX completion interrupt for some of its bus controllers. However, there is no use for the extra UART interrupt and it has to be masked out during initialization. Otherwise, if the firmware does not mask the interrupt and the driver does not clear it, it may cause an interrupt flood freezing the board to happen. Add code masking that problematic interrupt to the ACPI LPSS driver. [rjw: Changelog] Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2013-06-17 10:25:46 +00:00
}
ACPI / LPSS: Deassert resets for SPI host controllers on Braswell On some Braswell systems BIOS leaves resets for SPI host controllers active. This prevents the SPI driver from transferring messages on wire. Fix this in similar way that we do for I2C already by deasserting resets for the SPI host controllers. Reported-by: Yang A Fang <yang.a.fang@intel.com> Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Cc: 3.17+ <stable@vger.kernel.org> # 3.17+ Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-02-18 11:50:17 +00:00
static void lpss_deassert_reset(struct lpss_private_data *pdata)
ACPI / LPSS: Take I2C host controllers out of reset On Intel Baytrail, some I2C host controllers are held in reset when the OS gets control. This causes the driver to fail to detect the hardware properly. Fix this so that we make sure that the I2C host controller is not in reset when the driver gets probe'd. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-06-17 11:33:39 +00:00
{
unsigned int offset;
u32 val;
offset = pdata->dev_desc->prv_offset + LPSS_RESETS;
val = readl(pdata->mmio_base + offset);
val |= LPSS_RESETS_RESET_APB | LPSS_RESETS_RESET_FUNC;
writel(val, pdata->mmio_base + offset);
ACPI / LPSS: Deassert resets for SPI host controllers on Braswell On some Braswell systems BIOS leaves resets for SPI host controllers active. This prevents the SPI driver from transferring messages on wire. Fix this in similar way that we do for I2C already by deasserting resets for the SPI host controllers. Reported-by: Yang A Fang <yang.a.fang@intel.com> Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Cc: 3.17+ <stable@vger.kernel.org> # 3.17+ Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-02-18 11:50:17 +00:00
}
#define LPSS_I2C_ENABLE 0x6c
static void byt_i2c_setup(struct lpss_private_data *pdata)
{
lpss_deassert_reset(pdata);
ACPI / LPSS: Take I2C host controllers out of reset On Intel Baytrail, some I2C host controllers are held in reset when the OS gets control. This causes the driver to fail to detect the hardware properly. Fix this so that we make sure that the I2C host controller is not in reset when the driver gets probe'd. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-06-17 11:33:39 +00:00
ACPI / LPSS: support for 133MHz I2C source clock on Baytrail The I2C controllers on Baytrail can get the clock from 100MHz or 133MHz source clock. The first bits in the private clock parameter register indicates which one is being used. Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Acked-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-09-02 07:55:09 +00:00
if (readl(pdata->mmio_base + pdata->dev_desc->prv_offset))
pdata->fixed_clk_rate = 133000000;
ACPI / LPSS: Always disable I2C host controllers On Baytrail and Braswell the BIOS might leave the I2C host controllers enabled, probably because it uses them for its own purposes. This is fine in normal cases because the I2C driver will disable the hardware when it is probed anyway. However, in case of suspend to disk it is different story. If the driver happens to be compiled as a module the boot kernel never loads the driver thus leaving host controllers enabled upon loading the hibernation image. The I2C host controller interrupt mask register has default value of 0x8ff, in other words it has most of the interrupts unmasked. When combined with the fact that the host controller is enabled, the driver immediately starts getting interrupts even before its resume hook is called (once IO-APIC is resumed). Since the driver is not prepared for this it will crash the kernel due to NULL pointer derefence because dev->msgs is NULL. Unfortunately we were not able to get full backtrace to from the console which could be reproduced here. In order to fix this even when the driver is compiled as module, we disable the I2C host controllers in byt_i2c_setup() before devices are created. Reported-by: Yu Chen <yu.c.chen@intel.com> Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Cc: 3.17+ <stable@vger.kernel.org> # 3.17+ Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-02-18 11:50:16 +00:00
writel(0, pdata->mmio_base + LPSS_I2C_ENABLE);
ACPI / LPSS: Take I2C host controllers out of reset On Intel Baytrail, some I2C host controllers are held in reset when the OS gets control. This causes the driver to fail to detect the hardware properly. Fix this so that we make sure that the I2C host controller is not in reset when the driver gets probe'd. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-06-17 11:33:39 +00:00
}
ACPI / LPSS: add LPSS device for Wildcat Point PCH INT3438 is the ADSP device on Wildcat Point platform with 2 DW DMA engines built In. The DMA engines are used for DSP FW loading and audio data transferring. These DMA engine probing need the clock, without it, probing may failed and can't go forward. Add LPSS device "INT3438" for Wildcat Point PCH, to provide clock for its ADSP DMA engine probing. Signed-off-by: Jie Yang <yang.jie@intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-08-01 01:06:35 +00:00
ACPI / LPSS: constify device descriptors The device descriptors are never written to -- even pointed to as 'const' from struct lpss_private_data. Make them r/o for real. Signed-off-by: Mathias Krause <minipli@googlemail.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-06-13 12:26:50 +00:00
static const struct lpss_device_desc lpt_dev_desc = {
ACPI / LPSS: introduce flags Replace the booleans with a single flags member variable. Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Acked-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-09-02 07:55:07 +00:00
.flags = LPSS_CLK | LPSS_CLK_GATE | LPSS_CLK_DIVIDER | LPSS_LTR,
ACPI / LPSS: support for fractional divider clock This creates fractional divider type clock for the ones that have it. It is needed by the UART driver as the clock rate must accommodate to the requested baud rate. Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-05-19 11:42:07 +00:00
.prv_offset = 0x800,
};
ACPI / LPSS: constify device descriptors The device descriptors are never written to -- even pointed to as 'const' from struct lpss_private_data. Make them r/o for real. Signed-off-by: Mathias Krause <minipli@googlemail.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-06-13 12:26:50 +00:00
static const struct lpss_device_desc lpt_i2c_dev_desc = {
ACPI / LPSS: introduce flags Replace the booleans with a single flags member variable. Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Acked-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-09-02 07:55:07 +00:00
.flags = LPSS_CLK | LPSS_CLK_GATE | LPSS_LTR,
ACPI / LPSS: Add support for exposing LTR registers to user space Devices on the Intel Lynxpoint Low Power Subsystem (LPSS) have registers providing access to LTR (Latency Tolerance Reporting) functionality that allows software to monitor and possibly influence the aggressiveness of the platform's active-state power management. For each LPSS device, there are two modes of operation related to LTR, the auto mode and the software mode. In the auto mode the LTR is set up by the platform firmware and managed by hardware. Software can only read the LTR register values to monitor the platform's behavior. In the software mode it is possible to use LTR to control the extent to which the platform will use its built-in power management features. This changeset adds support for reading the LPSS devices' LTR registers and exposing their values to user space for monitoring and diagnostics purposes. It re-uses the MMIO mappings created to access the LPSS devices' clock registers for reading the values of the LTR registers and exposes them to user space through sysfs device attributes. Namely, a new atrribute group, lpss_ltr, is created for each LPSS device. It contains three new attributes: ltr_mode, auto_ltr, sw_ltr. The value of the ltr_mode attribute reflects the LTR mode being used at the moment (software vs auto) and the other two contain the actual register values (raw) whose meaning depends on the LTR mode. All of these attributes are read-only. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2013-03-06 22:46:28 +00:00
.prv_offset = 0x800,
};
ACPI / LPSS: constify device descriptors The device descriptors are never written to -- even pointed to as 'const' from struct lpss_private_data. Make them r/o for real. Signed-off-by: Mathias Krause <minipli@googlemail.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-06-13 12:26:50 +00:00
static const struct lpss_device_desc lpt_uart_dev_desc = {
ACPI / LPSS: introduce flags Replace the booleans with a single flags member variable. Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Acked-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-09-02 07:55:07 +00:00
.flags = LPSS_CLK | LPSS_CLK_GATE | LPSS_CLK_DIVIDER | LPSS_LTR,
ACPI / LPSS: provide con_id for the clkdev Commit 7d78cbefaa (serial: 8250_dw: add ability to handle the peripheral clock) introduces handling for a second clk to 8250_dw.c which is the driver also for LPSS UART. The second clk forces us to provide identifier (con_id) for the clkdev we create. This fixes an issue where 8250_dw.c is getting the same handler for both clocks. Fixes: 7d78cbefaa (serial: 8250_dw: add ability to handle the peripheral clock) Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Cc: 3.17+ <stable@vger.kernel.org> # 3.17+ Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-03-06 13:48:38 +00:00
.clk_con_id = "baudclk",
ACPI / LPSS: mask the UART TX completion interrupt Intel LPSS provides an extra TX byte counter and an extra TX completion interrupt for some of its bus controllers. However, there is no use for the extra UART interrupt and it has to be masked out during initialization. Otherwise, if the firmware does not mask the interrupt and the driver does not clear it, it may cause an interrupt flood freezing the board to happen. Add code masking that problematic interrupt to the ACPI LPSS driver. [rjw: Changelog] Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2013-06-17 10:25:46 +00:00
.prv_offset = 0x800,
.setup = lpss_uart_setup,
ACPI / LPSS: Add support for exposing LTR registers to user space Devices on the Intel Lynxpoint Low Power Subsystem (LPSS) have registers providing access to LTR (Latency Tolerance Reporting) functionality that allows software to monitor and possibly influence the aggressiveness of the platform's active-state power management. For each LPSS device, there are two modes of operation related to LTR, the auto mode and the software mode. In the auto mode the LTR is set up by the platform firmware and managed by hardware. Software can only read the LTR register values to monitor the platform's behavior. In the software mode it is possible to use LTR to control the extent to which the platform will use its built-in power management features. This changeset adds support for reading the LPSS devices' LTR registers and exposing their values to user space for monitoring and diagnostics purposes. It re-uses the MMIO mappings created to access the LPSS devices' clock registers for reading the values of the LTR registers and exposes them to user space through sysfs device attributes. Namely, a new atrribute group, lpss_ltr, is created for each LPSS device. It contains three new attributes: ltr_mode, auto_ltr, sw_ltr. The value of the ltr_mode attribute reflects the LTR mode being used at the moment (software vs auto) and the other two contain the actual register values (raw) whose meaning depends on the LTR mode. All of these attributes are read-only. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2013-03-06 22:46:28 +00:00
};
ACPI / LPSS: constify device descriptors The device descriptors are never written to -- even pointed to as 'const' from struct lpss_private_data. Make them r/o for real. Signed-off-by: Mathias Krause <minipli@googlemail.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-06-13 12:26:50 +00:00
static const struct lpss_device_desc lpt_sdio_dev_desc = {
ACPI / LPSS: introduce flags Replace the booleans with a single flags member variable. Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Acked-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-09-02 07:55:07 +00:00
.flags = LPSS_LTR,
ACPI / LPSS: Add support for exposing LTR registers to user space Devices on the Intel Lynxpoint Low Power Subsystem (LPSS) have registers providing access to LTR (Latency Tolerance Reporting) functionality that allows software to monitor and possibly influence the aggressiveness of the platform's active-state power management. For each LPSS device, there are two modes of operation related to LTR, the auto mode and the software mode. In the auto mode the LTR is set up by the platform firmware and managed by hardware. Software can only read the LTR register values to monitor the platform's behavior. In the software mode it is possible to use LTR to control the extent to which the platform will use its built-in power management features. This changeset adds support for reading the LPSS devices' LTR registers and exposing their values to user space for monitoring and diagnostics purposes. It re-uses the MMIO mappings created to access the LPSS devices' clock registers for reading the values of the LTR registers and exposes them to user space through sysfs device attributes. Namely, a new atrribute group, lpss_ltr, is created for each LPSS device. It contains three new attributes: ltr_mode, auto_ltr, sw_ltr. The value of the ltr_mode attribute reflects the LTR mode being used at the moment (software vs auto) and the other two contain the actual register values (raw) whose meaning depends on the LTR mode. All of these attributes are read-only. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2013-03-06 22:46:28 +00:00
.prv_offset = 0x1000,
ACPI / LPSS: override SDIO private register space size from ACPI tables The SDIO device in Lynxpoint has its LTR registers reserved for a WiFi device (a child of the SDIO device) in the ACPI namespace even though those registers physically belong to the SDIO device itself. In order to be able to access the SDIO LTR registers from the ACPI LPSS driver for diagnostic purposes we need to use a size override for the SDIO private register space. Add a possibility to override the size of the private register space of an LPSS device provided by the ACPI tables in the ACPI LPSS driver and set the correct size for the SDIO device in there. [rjw: Changelog] Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2013-06-18 13:51:35 +00:00
.prv_size_override = 0x1018,
ACPI / LPSS: Add Intel BayTrail ACPI mode PWM Intel BayTrail LPSS consists of two PWM controllers which can be enumerated from ACPI namespace. This change will cause platform device objects to be created for Intel BayTrail PWM controllers which will allow the pwm-lpss driver to bind to them and handle those devices. Signed-off-by: Chew, Chiau Ee <chiau.ee.chew@intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-02-18 18:24:29 +00:00
};
ACPI / LPSS: constify device descriptors The device descriptors are never written to -- even pointed to as 'const' from struct lpss_private_data. Make them r/o for real. Signed-off-by: Mathias Krause <minipli@googlemail.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-06-13 12:26:50 +00:00
static const struct lpss_device_desc byt_pwm_dev_desc = {
ACPI / LPSS: remove struct lpss_shared_clock Nothing requires it anymore. The PWM driver no longer uses clk framework to get the rate. Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Acked-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-09-02 07:55:10 +00:00
.flags = LPSS_SAVE_CTX,
ACPI / LPSS: Add Intel BayTrail ACPI mode PWM Intel BayTrail LPSS consists of two PWM controllers which can be enumerated from ACPI namespace. This change will cause platform device objects to be created for Intel BayTrail PWM controllers which will allow the pwm-lpss driver to bind to them and handle those devices. Signed-off-by: Chew, Chiau Ee <chiau.ee.chew@intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-02-18 18:24:29 +00:00
};
ACPI / LPSS: Ignore 10ms delay for Braswell LPSS devices in Braswell does not need the default 10ms d3_delay imposed by PCI specification. Removing this unnecessary delay significantly reduces the resume time approximately upto 200ms on this platform. Signed-off-by: Srinidhi Kasagar <srinidhi.kasagar@intel.com> Acked-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-08-27 16:00:55 +00:00
static const struct lpss_device_desc bsw_pwm_dev_desc = {
.flags = LPSS_SAVE_CTX | LPSS_NO_D3_DELAY,
};
ACPI / LPSS: constify device descriptors The device descriptors are never written to -- even pointed to as 'const' from struct lpss_private_data. Make them r/o for real. Signed-off-by: Mathias Krause <minipli@googlemail.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-06-13 12:26:50 +00:00
static const struct lpss_device_desc byt_uart_dev_desc = {
Revert "ACPI / LPSS: introduce a 'proxy' device to power on LPSS for DMA" Revert commit 6c17ee44d524 (ACPI / LPSS: introduce a 'proxy' device to power on LPSS for DMA), as it introduced registration and probe ordering problems between devices on the LPSS that may lead to full hard system hang on boot in some cases.
2015-02-03 13:29:43 +00:00
.flags = LPSS_CLK | LPSS_CLK_GATE | LPSS_CLK_DIVIDER | LPSS_SAVE_CTX,
ACPI / LPSS: provide con_id for the clkdev Commit 7d78cbefaa (serial: 8250_dw: add ability to handle the peripheral clock) introduces handling for a second clk to 8250_dw.c which is the driver also for LPSS UART. The second clk forces us to provide identifier (con_id) for the clkdev we create. This fixes an issue where 8250_dw.c is getting the same handler for both clocks. Fixes: 7d78cbefaa (serial: 8250_dw: add ability to handle the peripheral clock) Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Cc: 3.17+ <stable@vger.kernel.org> # 3.17+ Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-03-06 13:48:38 +00:00
.clk_con_id = "baudclk",
ACPI / LPSS: add support for Intel BayTrail Intel BayTrail has almost the same Low Power Subsystem than Lynxpoint with few differences. Peripherals are clocked with different speeds (typically lower) and the clock is not always gated. To support this we add possibility to share a common fixed rate clock and make clock gating optional. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Acked-by: Mike Turquette <mturquette@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2013-05-13 12:42:44 +00:00
.prv_offset = 0x800,
ACPI / LPSS: mask the UART TX completion interrupt Intel LPSS provides an extra TX byte counter and an extra TX completion interrupt for some of its bus controllers. However, there is no use for the extra UART interrupt and it has to be masked out during initialization. Otherwise, if the firmware does not mask the interrupt and the driver does not clear it, it may cause an interrupt flood freezing the board to happen. Add code masking that problematic interrupt to the ACPI LPSS driver. [rjw: Changelog] Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2013-06-17 10:25:46 +00:00
.setup = lpss_uart_setup,
ACPI / LPSS: add support for Intel BayTrail Intel BayTrail has almost the same Low Power Subsystem than Lynxpoint with few differences. Peripherals are clocked with different speeds (typically lower) and the clock is not always gated. To support this we add possibility to share a common fixed rate clock and make clock gating optional. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Acked-by: Mike Turquette <mturquette@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2013-05-13 12:42:44 +00:00
};
ACPI / LPSS: Ignore 10ms delay for Braswell LPSS devices in Braswell does not need the default 10ms d3_delay imposed by PCI specification. Removing this unnecessary delay significantly reduces the resume time approximately upto 200ms on this platform. Signed-off-by: Srinidhi Kasagar <srinidhi.kasagar@intel.com> Acked-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-08-27 16:00:55 +00:00
static const struct lpss_device_desc bsw_uart_dev_desc = {
.flags = LPSS_CLK | LPSS_CLK_GATE | LPSS_CLK_DIVIDER | LPSS_SAVE_CTX
| LPSS_NO_D3_DELAY,
.clk_con_id = "baudclk",
.prv_offset = 0x800,
.setup = lpss_uart_setup,
};
ACPI / LPSS: constify device descriptors The device descriptors are never written to -- even pointed to as 'const' from struct lpss_private_data. Make them r/o for real. Signed-off-by: Mathias Krause <minipli@googlemail.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-06-13 12:26:50 +00:00
static const struct lpss_device_desc byt_spi_dev_desc = {
Revert "ACPI / LPSS: introduce a 'proxy' device to power on LPSS for DMA" Revert commit 6c17ee44d524 (ACPI / LPSS: introduce a 'proxy' device to power on LPSS for DMA), as it introduced registration and probe ordering problems between devices on the LPSS that may lead to full hard system hang on boot in some cases.
2015-02-03 13:29:43 +00:00
.flags = LPSS_CLK | LPSS_CLK_GATE | LPSS_CLK_DIVIDER | LPSS_SAVE_CTX,
ACPI / LPSS: add support for Intel BayTrail Intel BayTrail has almost the same Low Power Subsystem than Lynxpoint with few differences. Peripherals are clocked with different speeds (typically lower) and the clock is not always gated. To support this we add possibility to share a common fixed rate clock and make clock gating optional. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Acked-by: Mike Turquette <mturquette@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2013-05-13 12:42:44 +00:00
.prv_offset = 0x400,
};
ACPI / LPSS: constify device descriptors The device descriptors are never written to -- even pointed to as 'const' from struct lpss_private_data. Make them r/o for real. Signed-off-by: Mathias Krause <minipli@googlemail.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-06-13 12:26:50 +00:00
static const struct lpss_device_desc byt_sdio_dev_desc = {
Revert "ACPI / LPSS: introduce a 'proxy' device to power on LPSS for DMA" Revert commit 6c17ee44d524 (ACPI / LPSS: introduce a 'proxy' device to power on LPSS for DMA), as it introduced registration and probe ordering problems between devices on the LPSS that may lead to full hard system hang on boot in some cases.
2015-02-03 13:29:43 +00:00
.flags = LPSS_CLK,
ACPI / LPSS: add support for Intel BayTrail Intel BayTrail has almost the same Low Power Subsystem than Lynxpoint with few differences. Peripherals are clocked with different speeds (typically lower) and the clock is not always gated. To support this we add possibility to share a common fixed rate clock and make clock gating optional. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Acked-by: Mike Turquette <mturquette@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2013-05-13 12:42:44 +00:00
};
ACPI / LPSS: constify device descriptors The device descriptors are never written to -- even pointed to as 'const' from struct lpss_private_data. Make them r/o for real. Signed-off-by: Mathias Krause <minipli@googlemail.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-06-13 12:26:50 +00:00
static const struct lpss_device_desc byt_i2c_dev_desc = {
Revert "ACPI / LPSS: introduce a 'proxy' device to power on LPSS for DMA" Revert commit 6c17ee44d524 (ACPI / LPSS: introduce a 'proxy' device to power on LPSS for DMA), as it introduced registration and probe ordering problems between devices on the LPSS that may lead to full hard system hang on boot in some cases.
2015-02-03 13:29:43 +00:00
.flags = LPSS_CLK | LPSS_SAVE_CTX,
ACPI / LPSS: add support for Intel BayTrail Intel BayTrail has almost the same Low Power Subsystem than Lynxpoint with few differences. Peripherals are clocked with different speeds (typically lower) and the clock is not always gated. To support this we add possibility to share a common fixed rate clock and make clock gating optional. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Acked-by: Mike Turquette <mturquette@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2013-05-13 12:42:44 +00:00
.prv_offset = 0x800,
ACPI / LPSS: support for 133MHz I2C source clock on Baytrail The I2C controllers on Baytrail can get the clock from 100MHz or 133MHz source clock. The first bits in the private clock parameter register indicates which one is being used. Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Acked-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-09-02 07:55:09 +00:00
.setup = byt_i2c_setup,
ACPI / LPSS: Add ACPI IDs for Intel Braswell Enable more identifiers for the existing devices for Intel Braswell and Cherryview. Signed-off-by: Alan Cox <alan@linux.intel.com> Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-08-19 12:55:22 +00:00
};
ACPI / LPSS: Ignore 10ms delay for Braswell LPSS devices in Braswell does not need the default 10ms d3_delay imposed by PCI specification. Removing this unnecessary delay significantly reduces the resume time approximately upto 200ms on this platform. Signed-off-by: Srinidhi Kasagar <srinidhi.kasagar@intel.com> Acked-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-08-27 16:00:55 +00:00
static const struct lpss_device_desc bsw_i2c_dev_desc = {
.flags = LPSS_CLK | LPSS_SAVE_CTX | LPSS_NO_D3_DELAY,
.prv_offset = 0x800,
.setup = byt_i2c_setup,
};
ACPI / LPSS: override power state for LPSS DMA device This is a third approach to workaround long standing issue with LPSS on BayTrail. First one [1] was reverted since it didn't resolve the issue comprehensively. Second one [2] was rejected by internal review. The LPSS DMA controller does not have neither _PS0 nor _PS3 method. Moreover it can be powered off automatically whenever the last LPSS device goes down. In case of no power any access to the DMA controller will hang the system. The behaviour is reproduced on some HP laptops based on Intel BayTrail [3,4] as well as on ASuS T100TA transformer. Power on the LPSS island through the registers accessible in a specific way. [1] http://www.spinics.net/lists/linux-acpi/msg53963.html [2] https://bugzilla.redhat.com/attachment.cgi?id=1066779&action=diff [3] https://bugzilla.redhat.com/show_bug.cgi?id=1184273 [4] http://www.spinics.net/lists/dmaengine/msg01514.html Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-12-12 01:45:06 +00:00
static const struct lpss_device_desc bsw_spi_dev_desc = {
ACPI / LPSS: Ignore 10ms delay for Braswell LPSS devices in Braswell does not need the default 10ms d3_delay imposed by PCI specification. Removing this unnecessary delay significantly reduces the resume time approximately upto 200ms on this platform. Signed-off-by: Srinidhi Kasagar <srinidhi.kasagar@intel.com> Acked-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-08-27 16:00:55 +00:00
.flags = LPSS_CLK | LPSS_CLK_GATE | LPSS_CLK_DIVIDER | LPSS_SAVE_CTX
| LPSS_NO_D3_DELAY,
ACPI / LPSS: Deassert resets for SPI host controllers on Braswell On some Braswell systems BIOS leaves resets for SPI host controllers active. This prevents the SPI driver from transferring messages on wire. Fix this in similar way that we do for I2C already by deasserting resets for the SPI host controllers. Reported-by: Yang A Fang <yang.a.fang@intel.com> Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Cc: 3.17+ <stable@vger.kernel.org> # 3.17+ Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-02-18 11:50:17 +00:00
.prv_offset = 0x400,
.setup = lpss_deassert_reset,
};
ACPI / LPSS: override power state for LPSS DMA device This is a third approach to workaround long standing issue with LPSS on BayTrail. First one [1] was reverted since it didn't resolve the issue comprehensively. Second one [2] was rejected by internal review. The LPSS DMA controller does not have neither _PS0 nor _PS3 method. Moreover it can be powered off automatically whenever the last LPSS device goes down. In case of no power any access to the DMA controller will hang the system. The behaviour is reproduced on some HP laptops based on Intel BayTrail [3,4] as well as on ASuS T100TA transformer. Power on the LPSS island through the registers accessible in a specific way. [1] http://www.spinics.net/lists/linux-acpi/msg53963.html [2] https://bugzilla.redhat.com/attachment.cgi?id=1066779&action=diff [3] https://bugzilla.redhat.com/show_bug.cgi?id=1184273 [4] http://www.spinics.net/lists/dmaengine/msg01514.html Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-12-12 01:45:06 +00:00
#define ICPU(model) { X86_VENDOR_INTEL, 6, model, X86_FEATURE_ANY, }
static const struct x86_cpu_id lpss_cpu_ids[] = {
ICPU(0x37), /* Valleyview, Bay Trail */
ICPU(0x4c), /* Braswell, Cherry Trail */
{}
};
ACPI / scan: always register ACPI LPSS scan handler Prevent platform devices from being created for ACPI LPSS devices if CONFIG_X86_INTEL_LPSS is unset by compiling out the LPSS scan handler's callbacks only in that case and still compiling its device ID list in and registering the scan handler in either case. This change is based on a prototype from Zhang Rui. Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Reviewed-by: Mika Westerberg <mika.westerberg@linux.intel.com>
2014-05-30 12:34:05 +00:00
#else
#define LPSS_ADDR(desc) (0UL)
#endif /* CONFIG_X86_INTEL_LPSS */
ACPI / scan: Add special handler for Intel Lynxpoint LPSS devices Devices on the Intel Lynxpoint Low Power Subsystem (LPSS) have some common features that aren't shared with any other platform devices, including the clock and LTR (Latency Tolerance Reporting) registers. It is better to handle those features in common code than to bother device drivers with doing that (I/O functionality-wise the LPSS devices are generally compatible with other devices that don't have those special registers and may be handled by the same drivers). The clock registers of the LPSS devices are now taken care of by the special clk-x86-lpss driver, but the MMIO mappings used for accessing those registers can also be used for accessing the LTR registers on those devices (LTR support for the Lynxpoint LPSS is going to be added by a subsequent patch). Thus it is convenient to add a special ACPI scan handler for the Lynxpoint LPSS devices that will create the MMIO mappings for accessing the clock (and LTR in the future) registers and will register the LPSS devices' clocks, so the clk-x86-lpss driver will only need to take care of the main Lynxpoint LPSS clock. Introduce a special ACPI scan handler for Intel Lynxpoint LPSS devices as described above. This also reduces overhead related to browsing the ACPI namespace in search of the LPSS devices before the registration of their clocks, removes some LPSS-specific (and somewhat ugly) code from acpi_platform.c and shrinks the overall code size slightly. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Acked-by: Mike Turquette <mturquette@linaro.org>
2013-03-06 22:46:20 +00:00
static const struct acpi_device_id acpi_lpss_device_ids[] = {
ACPI / LPSS: register clock device for Lynxpoint DMA properly The DMA controller in Lynxpoint is enumerated as a regular ACPI device now. To work properly it is using the LPSS root clock as a functional clock. That's why we have to register the clock device accordingly to the ACPI ID of the DMA controller. The acpi_lpss.c module is responsible to do the job. This patch also removes hardcoded name of the DMA device in clk-lpt.c and the name of the root clock in acpi_lpss.c. Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Signed-off-by: Vinod Koul <vinod.koul@intel.com>
2013-05-08 08:55:49 +00:00
/* Generic LPSS devices */
ACPI / scan: always register ACPI LPSS scan handler Prevent platform devices from being created for ACPI LPSS devices if CONFIG_X86_INTEL_LPSS is unset by compiling out the LPSS scan handler's callbacks only in that case and still compiling its device ID list in and registering the scan handler in either case. This change is based on a prototype from Zhang Rui. Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Reviewed-by: Mika Westerberg <mika.westerberg@linux.intel.com>
2014-05-30 12:34:05 +00:00
{ "INTL9C60", LPSS_ADDR(lpss_dma_desc) },
ACPI / LPSS: register clock device for Lynxpoint DMA properly The DMA controller in Lynxpoint is enumerated as a regular ACPI device now. To work properly it is using the LPSS root clock as a functional clock. That's why we have to register the clock device accordingly to the ACPI ID of the DMA controller. The acpi_lpss.c module is responsible to do the job. This patch also removes hardcoded name of the DMA device in clk-lpt.c and the name of the root clock in acpi_lpss.c. Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Signed-off-by: Vinod Koul <vinod.koul@intel.com>
2013-05-08 08:55:49 +00:00
ACPI / scan: Add special handler for Intel Lynxpoint LPSS devices Devices on the Intel Lynxpoint Low Power Subsystem (LPSS) have some common features that aren't shared with any other platform devices, including the clock and LTR (Latency Tolerance Reporting) registers. It is better to handle those features in common code than to bother device drivers with doing that (I/O functionality-wise the LPSS devices are generally compatible with other devices that don't have those special registers and may be handled by the same drivers). The clock registers of the LPSS devices are now taken care of by the special clk-x86-lpss driver, but the MMIO mappings used for accessing those registers can also be used for accessing the LTR registers on those devices (LTR support for the Lynxpoint LPSS is going to be added by a subsequent patch). Thus it is convenient to add a special ACPI scan handler for the Lynxpoint LPSS devices that will create the MMIO mappings for accessing the clock (and LTR in the future) registers and will register the LPSS devices' clocks, so the clk-x86-lpss driver will only need to take care of the main Lynxpoint LPSS clock. Introduce a special ACPI scan handler for Intel Lynxpoint LPSS devices as described above. This also reduces overhead related to browsing the ACPI namespace in search of the LPSS devices before the registration of their clocks, removes some LPSS-specific (and somewhat ugly) code from acpi_platform.c and shrinks the overall code size slightly. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Acked-by: Mike Turquette <mturquette@linaro.org>
2013-03-06 22:46:20 +00:00
/* Lynxpoint LPSS devices */
ACPI / scan: always register ACPI LPSS scan handler Prevent platform devices from being created for ACPI LPSS devices if CONFIG_X86_INTEL_LPSS is unset by compiling out the LPSS scan handler's callbacks only in that case and still compiling its device ID list in and registering the scan handler in either case. This change is based on a prototype from Zhang Rui. Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Reviewed-by: Mika Westerberg <mika.westerberg@linux.intel.com>
2014-05-30 12:34:05 +00:00
{ "INT33C0", LPSS_ADDR(lpt_dev_desc) },
{ "INT33C1", LPSS_ADDR(lpt_dev_desc) },
{ "INT33C2", LPSS_ADDR(lpt_i2c_dev_desc) },
{ "INT33C3", LPSS_ADDR(lpt_i2c_dev_desc) },
{ "INT33C4", LPSS_ADDR(lpt_uart_dev_desc) },
{ "INT33C5", LPSS_ADDR(lpt_uart_dev_desc) },
{ "INT33C6", LPSS_ADDR(lpt_sdio_dev_desc) },
ACPI / scan: Add special handler for Intel Lynxpoint LPSS devices Devices on the Intel Lynxpoint Low Power Subsystem (LPSS) have some common features that aren't shared with any other platform devices, including the clock and LTR (Latency Tolerance Reporting) registers. It is better to handle those features in common code than to bother device drivers with doing that (I/O functionality-wise the LPSS devices are generally compatible with other devices that don't have those special registers and may be handled by the same drivers). The clock registers of the LPSS devices are now taken care of by the special clk-x86-lpss driver, but the MMIO mappings used for accessing those registers can also be used for accessing the LTR registers on those devices (LTR support for the Lynxpoint LPSS is going to be added by a subsequent patch). Thus it is convenient to add a special ACPI scan handler for the Lynxpoint LPSS devices that will create the MMIO mappings for accessing the clock (and LTR in the future) registers and will register the LPSS devices' clocks, so the clk-x86-lpss driver will only need to take care of the main Lynxpoint LPSS clock. Introduce a special ACPI scan handler for Intel Lynxpoint LPSS devices as described above. This also reduces overhead related to browsing the ACPI namespace in search of the LPSS devices before the registration of their clocks, removes some LPSS-specific (and somewhat ugly) code from acpi_platform.c and shrinks the overall code size slightly. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Acked-by: Mike Turquette <mturquette@linaro.org>
2013-03-06 22:46:20 +00:00
{ "INT33C7", },
ACPI / LPSS: add support for Intel BayTrail Intel BayTrail has almost the same Low Power Subsystem than Lynxpoint with few differences. Peripherals are clocked with different speeds (typically lower) and the clock is not always gated. To support this we add possibility to share a common fixed rate clock and make clock gating optional. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Acked-by: Mike Turquette <mturquette@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2013-05-13 12:42:44 +00:00
/* BayTrail LPSS devices */
ACPI / scan: always register ACPI LPSS scan handler Prevent platform devices from being created for ACPI LPSS devices if CONFIG_X86_INTEL_LPSS is unset by compiling out the LPSS scan handler's callbacks only in that case and still compiling its device ID list in and registering the scan handler in either case. This change is based on a prototype from Zhang Rui. Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Reviewed-by: Mika Westerberg <mika.westerberg@linux.intel.com>
2014-05-30 12:34:05 +00:00
{ "80860F09", LPSS_ADDR(byt_pwm_dev_desc) },
{ "80860F0A", LPSS_ADDR(byt_uart_dev_desc) },
{ "80860F0E", LPSS_ADDR(byt_spi_dev_desc) },
{ "80860F14", LPSS_ADDR(byt_sdio_dev_desc) },
{ "80860F41", LPSS_ADDR(byt_i2c_dev_desc) },
ACPI / LPSS: add support for Intel BayTrail Intel BayTrail has almost the same Low Power Subsystem than Lynxpoint with few differences. Peripherals are clocked with different speeds (typically lower) and the clock is not always gated. To support this we add possibility to share a common fixed rate clock and make clock gating optional. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Acked-by: Mike Turquette <mturquette@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2013-05-13 12:42:44 +00:00
{ "INT33B2", },
pinctrl: baytrail: Add back Baytrail-T ACPI ID Now that the x86 dynamic IRQ allocation problem has been resolved with commmit 62a08ae2a576 (genirq: x86: Ensure that dynamic irq allocation does not conflict), we can add back Baytrail-T ACPI ID to the pinctrl driver. This makes the driver to work on Asus T100 where it is needed for several things like ACPI GPIO events and SD card detection. References: https://bugzilla.kernel.org/show_bug.cgi?id=68291 Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Jin Yao <yao.jin@intel.com> Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2014-05-15 15:28:46 +00:00
{ "INT33FC", },
ACPI / LPSS: add support for Intel BayTrail Intel BayTrail has almost the same Low Power Subsystem than Lynxpoint with few differences. Peripherals are clocked with different speeds (typically lower) and the clock is not always gated. To support this we add possibility to share a common fixed rate clock and make clock gating optional. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Acked-by: Mike Turquette <mturquette@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2013-05-13 12:42:44 +00:00
ACPI / LPSS: Add ACPI IDs for Intel Braswell Enable more identifiers for the existing devices for Intel Braswell and Cherryview. Signed-off-by: Alan Cox <alan@linux.intel.com> Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-08-19 12:55:22 +00:00
/* Braswell LPSS devices */
ACPI / LPSS: Ignore 10ms delay for Braswell LPSS devices in Braswell does not need the default 10ms d3_delay imposed by PCI specification. Removing this unnecessary delay significantly reduces the resume time approximately upto 200ms on this platform. Signed-off-by: Srinidhi Kasagar <srinidhi.kasagar@intel.com> Acked-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-08-27 16:00:55 +00:00
{ "80862288", LPSS_ADDR(bsw_pwm_dev_desc) },
{ "8086228A", LPSS_ADDR(bsw_uart_dev_desc) },
ACPI / LPSS: Deassert resets for SPI host controllers on Braswell On some Braswell systems BIOS leaves resets for SPI host controllers active. This prevents the SPI driver from transferring messages on wire. Fix this in similar way that we do for I2C already by deasserting resets for the SPI host controllers. Reported-by: Yang A Fang <yang.a.fang@intel.com> Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Cc: 3.17+ <stable@vger.kernel.org> # 3.17+ Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-02-18 11:50:17 +00:00
{ "8086228E", LPSS_ADDR(bsw_spi_dev_desc) },
ACPI / LPSS: Ignore 10ms delay for Braswell LPSS devices in Braswell does not need the default 10ms d3_delay imposed by PCI specification. Removing this unnecessary delay significantly reduces the resume time approximately upto 200ms on this platform. Signed-off-by: Srinidhi Kasagar <srinidhi.kasagar@intel.com> Acked-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-08-27 16:00:55 +00:00
{ "808622C1", LPSS_ADDR(bsw_i2c_dev_desc) },
ACPI / LPSS: Add ACPI IDs for Intel Braswell Enable more identifiers for the existing devices for Intel Braswell and Cherryview. Signed-off-by: Alan Cox <alan@linux.intel.com> Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-08-19 12:55:22 +00:00
ACPI / LPSS: Ignore 10ms delay for Braswell LPSS devices in Braswell does not need the default 10ms d3_delay imposed by PCI specification. Removing this unnecessary delay significantly reduces the resume time approximately upto 200ms on this platform. Signed-off-by: Srinidhi Kasagar <srinidhi.kasagar@intel.com> Acked-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-08-27 16:00:55 +00:00
/* Broadwell LPSS devices */
ACPI / scan: always register ACPI LPSS scan handler Prevent platform devices from being created for ACPI LPSS devices if CONFIG_X86_INTEL_LPSS is unset by compiling out the LPSS scan handler's callbacks only in that case and still compiling its device ID list in and registering the scan handler in either case. This change is based on a prototype from Zhang Rui. Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Reviewed-by: Mika Westerberg <mika.westerberg@linux.intel.com>
2014-05-30 12:34:05 +00:00
{ "INT3430", LPSS_ADDR(lpt_dev_desc) },
{ "INT3431", LPSS_ADDR(lpt_dev_desc) },
{ "INT3432", LPSS_ADDR(lpt_i2c_dev_desc) },
{ "INT3433", LPSS_ADDR(lpt_i2c_dev_desc) },
{ "INT3434", LPSS_ADDR(lpt_uart_dev_desc) },
{ "INT3435", LPSS_ADDR(lpt_uart_dev_desc) },
{ "INT3436", LPSS_ADDR(lpt_sdio_dev_desc) },
ACPI / LPSS: add ACPI IDs for newer Intel PCHs Some recent Intel PCHs with LPSS have different ACPI IDs for the LPSS devices, so add these to the list as well. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2013-11-12 09:48:19 +00:00
{ "INT3437", },
ACPI / LPSS: introduce flags Replace the booleans with a single flags member variable. Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Acked-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-09-02 07:55:07 +00:00
/* Wildcat Point LPSS devices */
{ "INT3438", LPSS_ADDR(lpt_dev_desc) },
ACPI / LPSS: add LPSS device for Wildcat Point PCH INT3438 is the ADSP device on Wildcat Point platform with 2 DW DMA engines built In. The DMA engines are used for DSP FW loading and audio data transferring. These DMA engine probing need the clock, without it, probing may failed and can't go forward. Add LPSS device "INT3438" for Wildcat Point PCH, to provide clock for its ADSP DMA engine probing. Signed-off-by: Jie Yang <yang.jie@intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-08-01 01:06:35 +00:00
ACPI / scan: Add special handler for Intel Lynxpoint LPSS devices Devices on the Intel Lynxpoint Low Power Subsystem (LPSS) have some common features that aren't shared with any other platform devices, including the clock and LTR (Latency Tolerance Reporting) registers. It is better to handle those features in common code than to bother device drivers with doing that (I/O functionality-wise the LPSS devices are generally compatible with other devices that don't have those special registers and may be handled by the same drivers). The clock registers of the LPSS devices are now taken care of by the special clk-x86-lpss driver, but the MMIO mappings used for accessing those registers can also be used for accessing the LTR registers on those devices (LTR support for the Lynxpoint LPSS is going to be added by a subsequent patch). Thus it is convenient to add a special ACPI scan handler for the Lynxpoint LPSS devices that will create the MMIO mappings for accessing the clock (and LTR in the future) registers and will register the LPSS devices' clocks, so the clk-x86-lpss driver will only need to take care of the main Lynxpoint LPSS clock. Introduce a special ACPI scan handler for Intel Lynxpoint LPSS devices as described above. This also reduces overhead related to browsing the ACPI namespace in search of the LPSS devices before the registration of their clocks, removes some LPSS-specific (and somewhat ugly) code from acpi_platform.c and shrinks the overall code size slightly. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Acked-by: Mike Turquette <mturquette@linaro.org>
2013-03-06 22:46:20 +00:00
{ }
};
ACPI / scan: always register ACPI LPSS scan handler Prevent platform devices from being created for ACPI LPSS devices if CONFIG_X86_INTEL_LPSS is unset by compiling out the LPSS scan handler's callbacks only in that case and still compiling its device ID list in and registering the scan handler in either case. This change is based on a prototype from Zhang Rui. Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Reviewed-by: Mika Westerberg <mika.westerberg@linux.intel.com>
2014-05-30 12:34:05 +00:00
#ifdef CONFIG_X86_INTEL_LPSS
ACPI / scan: Add special handler for Intel Lynxpoint LPSS devices Devices on the Intel Lynxpoint Low Power Subsystem (LPSS) have some common features that aren't shared with any other platform devices, including the clock and LTR (Latency Tolerance Reporting) registers. It is better to handle those features in common code than to bother device drivers with doing that (I/O functionality-wise the LPSS devices are generally compatible with other devices that don't have those special registers and may be handled by the same drivers). The clock registers of the LPSS devices are now taken care of by the special clk-x86-lpss driver, but the MMIO mappings used for accessing those registers can also be used for accessing the LTR registers on those devices (LTR support for the Lynxpoint LPSS is going to be added by a subsequent patch). Thus it is convenient to add a special ACPI scan handler for the Lynxpoint LPSS devices that will create the MMIO mappings for accessing the clock (and LTR in the future) registers and will register the LPSS devices' clocks, so the clk-x86-lpss driver will only need to take care of the main Lynxpoint LPSS clock. Introduce a special ACPI scan handler for Intel Lynxpoint LPSS devices as described above. This also reduces overhead related to browsing the ACPI namespace in search of the LPSS devices before the registration of their clocks, removes some LPSS-specific (and somewhat ugly) code from acpi_platform.c and shrinks the overall code size slightly. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Acked-by: Mike Turquette <mturquette@linaro.org>
2013-03-06 22:46:20 +00:00
static int is_memory(struct acpi_resource *res, void *not_used)
{
struct resource r;
return !acpi_dev_resource_memory(res, &r);
}
/* LPSS main clock device. */
static struct platform_device *lpss_clk_dev;
static inline void lpt_register_clock_device(void)
{
lpss_clk_dev = platform_device_register_simple("clk-lpt", -1, NULL, 0);
}
static int register_device_clock(struct acpi_device *adev,
struct lpss_private_data *pdata)
{
const struct lpss_device_desc *dev_desc = pdata->dev_desc;
ACPI / LPSS: support for fractional divider clock This creates fractional divider type clock for the ones that have it. It is needed by the UART driver as the clock rate must accommodate to the requested baud rate. Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-05-19 11:42:07 +00:00
const char *devname = dev_name(&adev->dev);
ACPI / LPSS: add support for Intel BayTrail Intel BayTrail has almost the same Low Power Subsystem than Lynxpoint with few differences. Peripherals are clocked with different speeds (typically lower) and the clock is not always gated. To support this we add possibility to share a common fixed rate clock and make clock gating optional. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Acked-by: Mike Turquette <mturquette@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2013-05-13 12:42:44 +00:00
struct clk *clk = ERR_PTR(-ENODEV);
ACPI / LPSS: register clock device for Lynxpoint DMA properly The DMA controller in Lynxpoint is enumerated as a regular ACPI device now. To work properly it is using the LPSS root clock as a functional clock. That's why we have to register the clock device accordingly to the ACPI ID of the DMA controller. The acpi_lpss.c module is responsible to do the job. This patch also removes hardcoded name of the DMA device in clk-lpt.c and the name of the root clock in acpi_lpss.c. Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Signed-off-by: Vinod Koul <vinod.koul@intel.com>
2013-05-08 08:55:49 +00:00
struct lpss_clk_data *clk_data;
ACPI / LPSS: support for fractional divider clock This creates fractional divider type clock for the ones that have it. It is needed by the UART driver as the clock rate must accommodate to the requested baud rate. Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-05-19 11:42:07 +00:00
const char *parent, *clk_name;
void __iomem *prv_base;
ACPI / scan: Add special handler for Intel Lynxpoint LPSS devices Devices on the Intel Lynxpoint Low Power Subsystem (LPSS) have some common features that aren't shared with any other platform devices, including the clock and LTR (Latency Tolerance Reporting) registers. It is better to handle those features in common code than to bother device drivers with doing that (I/O functionality-wise the LPSS devices are generally compatible with other devices that don't have those special registers and may be handled by the same drivers). The clock registers of the LPSS devices are now taken care of by the special clk-x86-lpss driver, but the MMIO mappings used for accessing those registers can also be used for accessing the LTR registers on those devices (LTR support for the Lynxpoint LPSS is going to be added by a subsequent patch). Thus it is convenient to add a special ACPI scan handler for the Lynxpoint LPSS devices that will create the MMIO mappings for accessing the clock (and LTR in the future) registers and will register the LPSS devices' clocks, so the clk-x86-lpss driver will only need to take care of the main Lynxpoint LPSS clock. Introduce a special ACPI scan handler for Intel Lynxpoint LPSS devices as described above. This also reduces overhead related to browsing the ACPI namespace in search of the LPSS devices before the registration of their clocks, removes some LPSS-specific (and somewhat ugly) code from acpi_platform.c and shrinks the overall code size slightly. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Acked-by: Mike Turquette <mturquette@linaro.org>
2013-03-06 22:46:20 +00:00
if (!lpss_clk_dev)
lpt_register_clock_device();
ACPI / LPSS: register clock device for Lynxpoint DMA properly The DMA controller in Lynxpoint is enumerated as a regular ACPI device now. To work properly it is using the LPSS root clock as a functional clock. That's why we have to register the clock device accordingly to the ACPI ID of the DMA controller. The acpi_lpss.c module is responsible to do the job. This patch also removes hardcoded name of the DMA device in clk-lpt.c and the name of the root clock in acpi_lpss.c. Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Signed-off-by: Vinod Koul <vinod.koul@intel.com>
2013-05-08 08:55:49 +00:00
clk_data = platform_get_drvdata(lpss_clk_dev);
if (!clk_data)
return -ENODEV;
ACPI / LPSS: drop clkdev_name member from lpss_device_desc It was used to provide the correct con_id for the dma driver, but it's not needed. Even if the driver requests a clock with the con_id, it still gets the correct clock. The device name is enough to match a single clock. Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Acked-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Acked-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-09-02 07:55:08 +00:00
clk = clk_data->clk;
ACPI / LPSS: register clock device for Lynxpoint DMA properly The DMA controller in Lynxpoint is enumerated as a regular ACPI device now. To work properly it is using the LPSS root clock as a functional clock. That's why we have to register the clock device accordingly to the ACPI ID of the DMA controller. The acpi_lpss.c module is responsible to do the job. This patch also removes hardcoded name of the DMA device in clk-lpt.c and the name of the root clock in acpi_lpss.c. Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Signed-off-by: Vinod Koul <vinod.koul@intel.com>
2013-05-08 08:55:49 +00:00
if (!pdata->mmio_base
ACPI / LPSS: Add support for exposing LTR registers to user space Devices on the Intel Lynxpoint Low Power Subsystem (LPSS) have registers providing access to LTR (Latency Tolerance Reporting) functionality that allows software to monitor and possibly influence the aggressiveness of the platform's active-state power management. For each LPSS device, there are two modes of operation related to LTR, the auto mode and the software mode. In the auto mode the LTR is set up by the platform firmware and managed by hardware. Software can only read the LTR register values to monitor the platform's behavior. In the software mode it is possible to use LTR to control the extent to which the platform will use its built-in power management features. This changeset adds support for reading the LPSS devices' LTR registers and exposing their values to user space for monitoring and diagnostics purposes. It re-uses the MMIO mappings created to access the LPSS devices' clock registers for reading the values of the LTR registers and exposes them to user space through sysfs device attributes. Namely, a new atrribute group, lpss_ltr, is created for each LPSS device. It contains three new attributes: ltr_mode, auto_ltr, sw_ltr. The value of the ltr_mode attribute reflects the LTR mode being used at the moment (software vs auto) and the other two contain the actual register values (raw) whose meaning depends on the LTR mode. All of these attributes are read-only. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2013-03-06 22:46:28 +00:00
|| pdata->mmio_size < dev_desc->prv_offset + LPSS_CLK_SIZE)
ACPI / scan: Add special handler for Intel Lynxpoint LPSS devices Devices on the Intel Lynxpoint Low Power Subsystem (LPSS) have some common features that aren't shared with any other platform devices, including the clock and LTR (Latency Tolerance Reporting) registers. It is better to handle those features in common code than to bother device drivers with doing that (I/O functionality-wise the LPSS devices are generally compatible with other devices that don't have those special registers and may be handled by the same drivers). The clock registers of the LPSS devices are now taken care of by the special clk-x86-lpss driver, but the MMIO mappings used for accessing those registers can also be used for accessing the LTR registers on those devices (LTR support for the Lynxpoint LPSS is going to be added by a subsequent patch). Thus it is convenient to add a special ACPI scan handler for the Lynxpoint LPSS devices that will create the MMIO mappings for accessing the clock (and LTR in the future) registers and will register the LPSS devices' clocks, so the clk-x86-lpss driver will only need to take care of the main Lynxpoint LPSS clock. Introduce a special ACPI scan handler for Intel Lynxpoint LPSS devices as described above. This also reduces overhead related to browsing the ACPI namespace in search of the LPSS devices before the registration of their clocks, removes some LPSS-specific (and somewhat ugly) code from acpi_platform.c and shrinks the overall code size slightly. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Acked-by: Mike Turquette <mturquette@linaro.org>
2013-03-06 22:46:20 +00:00
return -ENODATA;
ACPI / LPSS: add support for Intel BayTrail Intel BayTrail has almost the same Low Power Subsystem than Lynxpoint with few differences. Peripherals are clocked with different speeds (typically lower) and the clock is not always gated. To support this we add possibility to share a common fixed rate clock and make clock gating optional. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Acked-by: Mike Turquette <mturquette@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2013-05-13 12:42:44 +00:00
parent = clk_data->name;
ACPI / LPSS: support for fractional divider clock This creates fractional divider type clock for the ones that have it. It is needed by the UART driver as the clock rate must accommodate to the requested baud rate. Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-05-19 11:42:07 +00:00
prv_base = pdata->mmio_base + dev_desc->prv_offset;
ACPI / LPSS: add support for Intel BayTrail Intel BayTrail has almost the same Low Power Subsystem than Lynxpoint with few differences. Peripherals are clocked with different speeds (typically lower) and the clock is not always gated. To support this we add possibility to share a common fixed rate clock and make clock gating optional. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Acked-by: Mike Turquette <mturquette@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2013-05-13 12:42:44 +00:00
ACPI / LPSS: support for 133MHz I2C source clock on Baytrail The I2C controllers on Baytrail can get the clock from 100MHz or 133MHz source clock. The first bits in the private clock parameter register indicates which one is being used. Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Acked-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-09-02 07:55:09 +00:00
if (pdata->fixed_clk_rate) {
clk = clk_register_fixed_rate(NULL, devname, parent, 0,
pdata->fixed_clk_rate);
goto out;
ACPI / LPSS: add support for Intel BayTrail Intel BayTrail has almost the same Low Power Subsystem than Lynxpoint with few differences. Peripherals are clocked with different speeds (typically lower) and the clock is not always gated. To support this we add possibility to share a common fixed rate clock and make clock gating optional. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Acked-by: Mike Turquette <mturquette@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2013-05-13 12:42:44 +00:00
}
ACPI / LPSS: introduce flags Replace the booleans with a single flags member variable. Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Acked-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-09-02 07:55:07 +00:00
if (dev_desc->flags & LPSS_CLK_GATE) {
ACPI / LPSS: support for fractional divider clock This creates fractional divider type clock for the ones that have it. It is needed by the UART driver as the clock rate must accommodate to the requested baud rate. Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-05-19 11:42:07 +00:00
clk = clk_register_gate(NULL, devname, parent, 0,
prv_base, 0, 0, NULL);
parent = devname;
}
ACPI / LPSS: introduce flags Replace the booleans with a single flags member variable. Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Acked-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-09-02 07:55:07 +00:00
if (dev_desc->flags & LPSS_CLK_DIVIDER) {
ACPI / LPSS: support for fractional divider clock This creates fractional divider type clock for the ones that have it. It is needed by the UART driver as the clock rate must accommodate to the requested baud rate. Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-05-19 11:42:07 +00:00
/* Prevent division by zero */
if (!readl(prv_base))
writel(LPSS_CLK_DIVIDER_DEF_MASK, prv_base);
clk_name = kasprintf(GFP_KERNEL, "%s-div", devname);
if (!clk_name)
return -ENOMEM;
clk = clk_register_fractional_divider(NULL, clk_name, parent,
0, prv_base,
1, 15, 16, 15, 0, NULL);
parent = clk_name;
clk_name = kasprintf(GFP_KERNEL, "%s-update", devname);
if (!clk_name) {
kfree(parent);
return -ENOMEM;
}
clk = clk_register_gate(NULL, clk_name, parent,
CLK_SET_RATE_PARENT | CLK_SET_RATE_GATE,
prv_base, 31, 0, NULL);
kfree(parent);
kfree(clk_name);
ACPI / LPSS: add support for Intel BayTrail Intel BayTrail has almost the same Low Power Subsystem than Lynxpoint with few differences. Peripherals are clocked with different speeds (typically lower) and the clock is not always gated. To support this we add possibility to share a common fixed rate clock and make clock gating optional. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Acked-by: Mike Turquette <mturquette@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2013-05-13 12:42:44 +00:00
}
ACPI / LPSS: support for 133MHz I2C source clock on Baytrail The I2C controllers on Baytrail can get the clock from 100MHz or 133MHz source clock. The first bits in the private clock parameter register indicates which one is being used. Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Acked-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-09-02 07:55:09 +00:00
out:
ACPI / LPSS: add support for Intel BayTrail Intel BayTrail has almost the same Low Power Subsystem than Lynxpoint with few differences. Peripherals are clocked with different speeds (typically lower) and the clock is not always gated. To support this we add possibility to share a common fixed rate clock and make clock gating optional. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Acked-by: Mike Turquette <mturquette@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2013-05-13 12:42:44 +00:00
if (IS_ERR(clk))
return PTR_ERR(clk);
ACPI / scan: Add special handler for Intel Lynxpoint LPSS devices Devices on the Intel Lynxpoint Low Power Subsystem (LPSS) have some common features that aren't shared with any other platform devices, including the clock and LTR (Latency Tolerance Reporting) registers. It is better to handle those features in common code than to bother device drivers with doing that (I/O functionality-wise the LPSS devices are generally compatible with other devices that don't have those special registers and may be handled by the same drivers). The clock registers of the LPSS devices are now taken care of by the special clk-x86-lpss driver, but the MMIO mappings used for accessing those registers can also be used for accessing the LTR registers on those devices (LTR support for the Lynxpoint LPSS is going to be added by a subsequent patch). Thus it is convenient to add a special ACPI scan handler for the Lynxpoint LPSS devices that will create the MMIO mappings for accessing the clock (and LTR in the future) registers and will register the LPSS devices' clocks, so the clk-x86-lpss driver will only need to take care of the main Lynxpoint LPSS clock. Introduce a special ACPI scan handler for Intel Lynxpoint LPSS devices as described above. This also reduces overhead related to browsing the ACPI namespace in search of the LPSS devices before the registration of their clocks, removes some LPSS-specific (and somewhat ugly) code from acpi_platform.c and shrinks the overall code size slightly. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Acked-by: Mike Turquette <mturquette@linaro.org>
2013-03-06 22:46:20 +00:00
ACPI / LPSS: support for fractional divider clock This creates fractional divider type clock for the ones that have it. It is needed by the UART driver as the clock rate must accommodate to the requested baud rate. Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-05-19 11:42:07 +00:00
pdata->clk = clk;
ACPI / LPSS: provide con_id for the clkdev Commit 7d78cbefaa (serial: 8250_dw: add ability to handle the peripheral clock) introduces handling for a second clk to 8250_dw.c which is the driver also for LPSS UART. The second clk forces us to provide identifier (con_id) for the clkdev we create. This fixes an issue where 8250_dw.c is getting the same handler for both clocks. Fixes: 7d78cbefaa (serial: 8250_dw: add ability to handle the peripheral clock) Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Cc: 3.17+ <stable@vger.kernel.org> # 3.17+ Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-03-06 13:48:38 +00:00
clk_register_clkdev(clk, dev_desc->clk_con_id, devname);
ACPI / scan: Add special handler for Intel Lynxpoint LPSS devices Devices on the Intel Lynxpoint Low Power Subsystem (LPSS) have some common features that aren't shared with any other platform devices, including the clock and LTR (Latency Tolerance Reporting) registers. It is better to handle those features in common code than to bother device drivers with doing that (I/O functionality-wise the LPSS devices are generally compatible with other devices that don't have those special registers and may be handled by the same drivers). The clock registers of the LPSS devices are now taken care of by the special clk-x86-lpss driver, but the MMIO mappings used for accessing those registers can also be used for accessing the LTR registers on those devices (LTR support for the Lynxpoint LPSS is going to be added by a subsequent patch). Thus it is convenient to add a special ACPI scan handler for the Lynxpoint LPSS devices that will create the MMIO mappings for accessing the clock (and LTR in the future) registers and will register the LPSS devices' clocks, so the clk-x86-lpss driver will only need to take care of the main Lynxpoint LPSS clock. Introduce a special ACPI scan handler for Intel Lynxpoint LPSS devices as described above. This also reduces overhead related to browsing the ACPI namespace in search of the LPSS devices before the registration of their clocks, removes some LPSS-specific (and somewhat ugly) code from acpi_platform.c and shrinks the overall code size slightly. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Acked-by: Mike Turquette <mturquette@linaro.org>
2013-03-06 22:46:20 +00:00
return 0;
}
static int acpi_lpss_create_device(struct acpi_device *adev,
const struct acpi_device_id *id)
{
ACPI / LPSS: constify device descriptors The device descriptors are never written to -- even pointed to as 'const' from struct lpss_private_data. Make them r/o for real. Signed-off-by: Mathias Krause <minipli@googlemail.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-06-13 12:26:50 +00:00
const struct lpss_device_desc *dev_desc;
ACPI / scan: Add special handler for Intel Lynxpoint LPSS devices Devices on the Intel Lynxpoint Low Power Subsystem (LPSS) have some common features that aren't shared with any other platform devices, including the clock and LTR (Latency Tolerance Reporting) registers. It is better to handle those features in common code than to bother device drivers with doing that (I/O functionality-wise the LPSS devices are generally compatible with other devices that don't have those special registers and may be handled by the same drivers). The clock registers of the LPSS devices are now taken care of by the special clk-x86-lpss driver, but the MMIO mappings used for accessing those registers can also be used for accessing the LTR registers on those devices (LTR support for the Lynxpoint LPSS is going to be added by a subsequent patch). Thus it is convenient to add a special ACPI scan handler for the Lynxpoint LPSS devices that will create the MMIO mappings for accessing the clock (and LTR in the future) registers and will register the LPSS devices' clocks, so the clk-x86-lpss driver will only need to take care of the main Lynxpoint LPSS clock. Introduce a special ACPI scan handler for Intel Lynxpoint LPSS devices as described above. This also reduces overhead related to browsing the ACPI namespace in search of the LPSS devices before the registration of their clocks, removes some LPSS-specific (and somewhat ugly) code from acpi_platform.c and shrinks the overall code size slightly. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Acked-by: Mike Turquette <mturquette@linaro.org>
2013-03-06 22:46:20 +00:00
struct lpss_private_data *pdata;
resources: Move struct resource_list_entry from ACPI into resource core Currently ACPI, PCI and pnp all implement the same resource list management with different data structure. We need to transfer from one data structure into another when passing resources from one subsystem into another subsystem. So move struct resource_list_entry from ACPI into resource core and rename it as resource_entry, then it could be reused by different subystems and avoid the data structure conversion. Introduce dedicated header file resource_ext.h instead of embedding it into ioport.h to avoid header file inclusion order issues. Signed-off-by: Jiang Liu <jiang.liu@linux.intel.com> Acked-by: Vinod Koul <vinod.koul@intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-02-05 05:44:43 +00:00
struct resource_entry *rentry;
ACPI / scan: Add special handler for Intel Lynxpoint LPSS devices Devices on the Intel Lynxpoint Low Power Subsystem (LPSS) have some common features that aren't shared with any other platform devices, including the clock and LTR (Latency Tolerance Reporting) registers. It is better to handle those features in common code than to bother device drivers with doing that (I/O functionality-wise the LPSS devices are generally compatible with other devices that don't have those special registers and may be handled by the same drivers). The clock registers of the LPSS devices are now taken care of by the special clk-x86-lpss driver, but the MMIO mappings used for accessing those registers can also be used for accessing the LTR registers on those devices (LTR support for the Lynxpoint LPSS is going to be added by a subsequent patch). Thus it is convenient to add a special ACPI scan handler for the Lynxpoint LPSS devices that will create the MMIO mappings for accessing the clock (and LTR in the future) registers and will register the LPSS devices' clocks, so the clk-x86-lpss driver will only need to take care of the main Lynxpoint LPSS clock. Introduce a special ACPI scan handler for Intel Lynxpoint LPSS devices as described above. This also reduces overhead related to browsing the ACPI namespace in search of the LPSS devices before the registration of their clocks, removes some LPSS-specific (and somewhat ugly) code from acpi_platform.c and shrinks the overall code size slightly. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Acked-by: Mike Turquette <mturquette@linaro.org>
2013-03-06 22:46:20 +00:00
struct list_head resource_list;
ACPI / platform / LPSS: Enable async suspend/resume of LPSS devices To seed up suspend and resume of devices included into Intel SoCs handled by the ACPI LPSS driver during system suspend, make acpi_lpss_create_device() call device_enable_async_suspend() for every device created by it. This requires acpi_create_platform_device() to be modified to return a pointer to struct platform_device instead of an int. As a result, acpi_create_platform_device() cannot be pointed to by the .attach pointer in platform_handler directly any more, so a simple wrapper around it is necessary for this purpose. That, in turn, allows the second unused argument of acpi_create_platform_device() to be dropped, which is an improvement. Tested-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-05-25 12:38:52 +00:00
struct platform_device *pdev;
ACPI / scan: Add special handler for Intel Lynxpoint LPSS devices Devices on the Intel Lynxpoint Low Power Subsystem (LPSS) have some common features that aren't shared with any other platform devices, including the clock and LTR (Latency Tolerance Reporting) registers. It is better to handle those features in common code than to bother device drivers with doing that (I/O functionality-wise the LPSS devices are generally compatible with other devices that don't have those special registers and may be handled by the same drivers). The clock registers of the LPSS devices are now taken care of by the special clk-x86-lpss driver, but the MMIO mappings used for accessing those registers can also be used for accessing the LTR registers on those devices (LTR support for the Lynxpoint LPSS is going to be added by a subsequent patch). Thus it is convenient to add a special ACPI scan handler for the Lynxpoint LPSS devices that will create the MMIO mappings for accessing the clock (and LTR in the future) registers and will register the LPSS devices' clocks, so the clk-x86-lpss driver will only need to take care of the main Lynxpoint LPSS clock. Introduce a special ACPI scan handler for Intel Lynxpoint LPSS devices as described above. This also reduces overhead related to browsing the ACPI namespace in search of the LPSS devices before the registration of their clocks, removes some LPSS-specific (and somewhat ugly) code from acpi_platform.c and shrinks the overall code size slightly. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Acked-by: Mike Turquette <mturquette@linaro.org>
2013-03-06 22:46:20 +00:00
int ret;
ACPI / LPSS: constify device descriptors The device descriptors are never written to -- even pointed to as 'const' from struct lpss_private_data. Make them r/o for real. Signed-off-by: Mathias Krause <minipli@googlemail.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-06-13 12:26:50 +00:00
dev_desc = (const struct lpss_device_desc *)id->driver_data;
ACPI / platform / LPSS: Enable async suspend/resume of LPSS devices To seed up suspend and resume of devices included into Intel SoCs handled by the ACPI LPSS driver during system suspend, make acpi_lpss_create_device() call device_enable_async_suspend() for every device created by it. This requires acpi_create_platform_device() to be modified to return a pointer to struct platform_device instead of an int. As a result, acpi_create_platform_device() cannot be pointed to by the .attach pointer in platform_handler directly any more, so a simple wrapper around it is necessary for this purpose. That, in turn, allows the second unused argument of acpi_create_platform_device() to be dropped, which is an improvement. Tested-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-05-25 12:38:52 +00:00
if (!dev_desc) {
pdev = acpi_create_platform_device(adev);
return IS_ERR_OR_NULL(pdev) ? PTR_ERR(pdev) : 1;
}
ACPI / scan: Add special handler for Intel Lynxpoint LPSS devices Devices on the Intel Lynxpoint Low Power Subsystem (LPSS) have some common features that aren't shared with any other platform devices, including the clock and LTR (Latency Tolerance Reporting) registers. It is better to handle those features in common code than to bother device drivers with doing that (I/O functionality-wise the LPSS devices are generally compatible with other devices that don't have those special registers and may be handled by the same drivers). The clock registers of the LPSS devices are now taken care of by the special clk-x86-lpss driver, but the MMIO mappings used for accessing those registers can also be used for accessing the LTR registers on those devices (LTR support for the Lynxpoint LPSS is going to be added by a subsequent patch). Thus it is convenient to add a special ACPI scan handler for the Lynxpoint LPSS devices that will create the MMIO mappings for accessing the clock (and LTR in the future) registers and will register the LPSS devices' clocks, so the clk-x86-lpss driver will only need to take care of the main Lynxpoint LPSS clock. Introduce a special ACPI scan handler for Intel Lynxpoint LPSS devices as described above. This also reduces overhead related to browsing the ACPI namespace in search of the LPSS devices before the registration of their clocks, removes some LPSS-specific (and somewhat ugly) code from acpi_platform.c and shrinks the overall code size slightly. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Acked-by: Mike Turquette <mturquette@linaro.org>
2013-03-06 22:46:20 +00:00
pdata = kzalloc(sizeof(*pdata), GFP_KERNEL);
if (!pdata)
return -ENOMEM;
INIT_LIST_HEAD(&resource_list);
ret = acpi_dev_get_resources(adev, &resource_list, is_memory, NULL);
if (ret < 0)
goto err_out;
list_for_each_entry(rentry, &resource_list, node)
resources: Move struct resource_list_entry from ACPI into resource core Currently ACPI, PCI and pnp all implement the same resource list management with different data structure. We need to transfer from one data structure into another when passing resources from one subsystem into another subsystem. So move struct resource_list_entry from ACPI into resource core and rename it as resource_entry, then it could be reused by different subystems and avoid the data structure conversion. Introduce dedicated header file resource_ext.h instead of embedding it into ioport.h to avoid header file inclusion order issues. Signed-off-by: Jiang Liu <jiang.liu@linux.intel.com> Acked-by: Vinod Koul <vinod.koul@intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-02-05 05:44:43 +00:00
if (resource_type(rentry->res) == IORESOURCE_MEM) {
ACPI / LPSS: override SDIO private register space size from ACPI tables The SDIO device in Lynxpoint has its LTR registers reserved for a WiFi device (a child of the SDIO device) in the ACPI namespace even though those registers physically belong to the SDIO device itself. In order to be able to access the SDIO LTR registers from the ACPI LPSS driver for diagnostic purposes we need to use a size override for the SDIO private register space. Add a possibility to override the size of the private register space of an LPSS device provided by the ACPI tables in the ACPI LPSS driver and set the correct size for the SDIO device in there. [rjw: Changelog] Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2013-06-18 13:51:35 +00:00
if (dev_desc->prv_size_override)
pdata->mmio_size = dev_desc->prv_size_override;
else
resources: Move struct resource_list_entry from ACPI into resource core Currently ACPI, PCI and pnp all implement the same resource list management with different data structure. We need to transfer from one data structure into another when passing resources from one subsystem into another subsystem. So move struct resource_list_entry from ACPI into resource core and rename it as resource_entry, then it could be reused by different subystems and avoid the data structure conversion. Introduce dedicated header file resource_ext.h instead of embedding it into ioport.h to avoid header file inclusion order issues. Signed-off-by: Jiang Liu <jiang.liu@linux.intel.com> Acked-by: Vinod Koul <vinod.koul@intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-02-05 05:44:43 +00:00
pdata->mmio_size = resource_size(rentry->res);
pdata->mmio_base = ioremap(rentry->res->start,
ACPI / scan: Add special handler for Intel Lynxpoint LPSS devices Devices on the Intel Lynxpoint Low Power Subsystem (LPSS) have some common features that aren't shared with any other platform devices, including the clock and LTR (Latency Tolerance Reporting) registers. It is better to handle those features in common code than to bother device drivers with doing that (I/O functionality-wise the LPSS devices are generally compatible with other devices that don't have those special registers and may be handled by the same drivers). The clock registers of the LPSS devices are now taken care of by the special clk-x86-lpss driver, but the MMIO mappings used for accessing those registers can also be used for accessing the LTR registers on those devices (LTR support for the Lynxpoint LPSS is going to be added by a subsequent patch). Thus it is convenient to add a special ACPI scan handler for the Lynxpoint LPSS devices that will create the MMIO mappings for accessing the clock (and LTR in the future) registers and will register the LPSS devices' clocks, so the clk-x86-lpss driver will only need to take care of the main Lynxpoint LPSS clock. Introduce a special ACPI scan handler for Intel Lynxpoint LPSS devices as described above. This also reduces overhead related to browsing the ACPI namespace in search of the LPSS devices before the registration of their clocks, removes some LPSS-specific (and somewhat ugly) code from acpi_platform.c and shrinks the overall code size slightly. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Acked-by: Mike Turquette <mturquette@linaro.org>
2013-03-06 22:46:20 +00:00
pdata->mmio_size);
break;
}
acpi_dev_free_resource_list(&resource_list);
ACPI / LPSS: Fix up acpi_lpss_create_device() Fix a return value (which should be a negative error code) and a memory leak (the list allocated by acpi_dev_get_resources() needs to be freed on ioremap() errors too) in acpi_lpss_create_device() introduced by commit 4483d59e29fe 'ACPI / LPSS: check the result of ioremap()'. Fixes: 4483d59e29fe 'ACPI / LPSS: check the result of ioremap()' Reported-by: Dan Carpenter <dan.carpenter@oracle.com> Cc: 4.0+ <stable@vger.kernel.org> # 4.0+ Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-07-06 22:31:47 +00:00
if (!pdata->mmio_base) {
ret = -ENOMEM;
goto err_out;
}
ACPI / LPSS: don't crash if a device has no MMIO resources Intel LPSS devices that are enumerated from ACPI have both MMIO and IRQ resources returned in their _CRS method. However, Apple Macbook Air with Haswell has LPSS devices enumerated from PCI bus instead and _CRS method returns only an interrupt number (but the device has _HID set that causes the scan handler to match it). The current ACPI / LPSS code sets pdata->dev_desc only when MMIO resource is found for the device and in case of Macbook Air it is never found. That leads to a NULL pointer dereference in register_device_clock(). Correct this by always setting the pdata->dev_desc. Reported-and-tested-by: Imre Kaloz <kaloz@openwrt.org> Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Cc: 3.10+ <stable@vger.kernel.org> # 3.10+ Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2013-09-02 10:30:25 +00:00
pdata->dev_desc = dev_desc;
ACPI / LPSS: support for 133MHz I2C source clock on Baytrail The I2C controllers on Baytrail can get the clock from 100MHz or 133MHz source clock. The first bits in the private clock parameter register indicates which one is being used. Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Acked-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-09-02 07:55:09 +00:00
if (dev_desc->setup)
dev_desc->setup(pdata);
ACPI / LPSS: introduce flags Replace the booleans with a single flags member variable. Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Acked-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-09-02 07:55:07 +00:00
if (dev_desc->flags & LPSS_CLK) {
ACPI / scan: Add special handler for Intel Lynxpoint LPSS devices Devices on the Intel Lynxpoint Low Power Subsystem (LPSS) have some common features that aren't shared with any other platform devices, including the clock and LTR (Latency Tolerance Reporting) registers. It is better to handle those features in common code than to bother device drivers with doing that (I/O functionality-wise the LPSS devices are generally compatible with other devices that don't have those special registers and may be handled by the same drivers). The clock registers of the LPSS devices are now taken care of by the special clk-x86-lpss driver, but the MMIO mappings used for accessing those registers can also be used for accessing the LTR registers on those devices (LTR support for the Lynxpoint LPSS is going to be added by a subsequent patch). Thus it is convenient to add a special ACPI scan handler for the Lynxpoint LPSS devices that will create the MMIO mappings for accessing the clock (and LTR in the future) registers and will register the LPSS devices' clocks, so the clk-x86-lpss driver will only need to take care of the main Lynxpoint LPSS clock. Introduce a special ACPI scan handler for Intel Lynxpoint LPSS devices as described above. This also reduces overhead related to browsing the ACPI namespace in search of the LPSS devices before the registration of their clocks, removes some LPSS-specific (and somewhat ugly) code from acpi_platform.c and shrinks the overall code size slightly. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Acked-by: Mike Turquette <mturquette@linaro.org>
2013-03-06 22:46:20 +00:00
ret = register_device_clock(adev, pdata);
if (ret) {
ACPI / LPSS: Power up LPSS devices during enumeration Commit 7cd8407 (ACPI / PM: Do not execute _PS0 for devices without _PSC during initialization) introduced a regression on some systems with Intel Lynxpoint Low-Power Subsystem (LPSS) where some devices need to be powered up during initialization, but their device objects in the ACPI namespace have _PS0 and _PS3 only (without _PSC or power resources). To work around this problem, make the ACPI LPSS driver power up devices it knows about by using a new helper function acpi_device_fix_up_power() that does all of the necessary sanity checks and calls acpi_dev_pm_explicit_set() to put the device into D0. Reported-and-tested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2013-06-18 22:45:34 +00:00
/* Skip the device, but continue the namespace scan. */
ret = 0;
goto err_out;
ACPI / scan: Add special handler for Intel Lynxpoint LPSS devices Devices on the Intel Lynxpoint Low Power Subsystem (LPSS) have some common features that aren't shared with any other platform devices, including the clock and LTR (Latency Tolerance Reporting) registers. It is better to handle those features in common code than to bother device drivers with doing that (I/O functionality-wise the LPSS devices are generally compatible with other devices that don't have those special registers and may be handled by the same drivers). The clock registers of the LPSS devices are now taken care of by the special clk-x86-lpss driver, but the MMIO mappings used for accessing those registers can also be used for accessing the LTR registers on those devices (LTR support for the Lynxpoint LPSS is going to be added by a subsequent patch). Thus it is convenient to add a special ACPI scan handler for the Lynxpoint LPSS devices that will create the MMIO mappings for accessing the clock (and LTR in the future) registers and will register the LPSS devices' clocks, so the clk-x86-lpss driver will only need to take care of the main Lynxpoint LPSS clock. Introduce a special ACPI scan handler for Intel Lynxpoint LPSS devices as described above. This also reduces overhead related to browsing the ACPI namespace in search of the LPSS devices before the registration of their clocks, removes some LPSS-specific (and somewhat ugly) code from acpi_platform.c and shrinks the overall code size slightly. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Acked-by: Mike Turquette <mturquette@linaro.org>
2013-03-06 22:46:20 +00:00
}
}
ACPI / LPSS: Power up LPSS devices during enumeration Commit 7cd8407 (ACPI / PM: Do not execute _PS0 for devices without _PSC during initialization) introduced a regression on some systems with Intel Lynxpoint Low-Power Subsystem (LPSS) where some devices need to be powered up during initialization, but their device objects in the ACPI namespace have _PS0 and _PS3 only (without _PSC or power resources). To work around this problem, make the ACPI LPSS driver power up devices it knows about by using a new helper function acpi_device_fix_up_power() that does all of the necessary sanity checks and calls acpi_dev_pm_explicit_set() to put the device into D0. Reported-and-tested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2013-06-18 22:45:34 +00:00
/*
* This works around a known issue in ACPI tables where LPSS devices
* have _PS0 and _PS3 without _PSC (and no power resources), so
* acpi_bus_init_power() will assume that the BIOS has put them into D0.
*/
ret = acpi_device_fix_up_power(adev);
if (ret) {
/* Skip the device, but continue the namespace scan. */
ret = 0;
goto err_out;
}
ACPI / scan: Add special handler for Intel Lynxpoint LPSS devices Devices on the Intel Lynxpoint Low Power Subsystem (LPSS) have some common features that aren't shared with any other platform devices, including the clock and LTR (Latency Tolerance Reporting) registers. It is better to handle those features in common code than to bother device drivers with doing that (I/O functionality-wise the LPSS devices are generally compatible with other devices that don't have those special registers and may be handled by the same drivers). The clock registers of the LPSS devices are now taken care of by the special clk-x86-lpss driver, but the MMIO mappings used for accessing those registers can also be used for accessing the LTR registers on those devices (LTR support for the Lynxpoint LPSS is going to be added by a subsequent patch). Thus it is convenient to add a special ACPI scan handler for the Lynxpoint LPSS devices that will create the MMIO mappings for accessing the clock (and LTR in the future) registers and will register the LPSS devices' clocks, so the clk-x86-lpss driver will only need to take care of the main Lynxpoint LPSS clock. Introduce a special ACPI scan handler for Intel Lynxpoint LPSS devices as described above. This also reduces overhead related to browsing the ACPI namespace in search of the LPSS devices before the registration of their clocks, removes some LPSS-specific (and somewhat ugly) code from acpi_platform.c and shrinks the overall code size slightly. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Acked-by: Mike Turquette <mturquette@linaro.org>
2013-03-06 22:46:20 +00:00
adev->driver_data = pdata;
ACPI / platform / LPSS: Enable async suspend/resume of LPSS devices To seed up suspend and resume of devices included into Intel SoCs handled by the ACPI LPSS driver during system suspend, make acpi_lpss_create_device() call device_enable_async_suspend() for every device created by it. This requires acpi_create_platform_device() to be modified to return a pointer to struct platform_device instead of an int. As a result, acpi_create_platform_device() cannot be pointed to by the .attach pointer in platform_handler directly any more, so a simple wrapper around it is necessary for this purpose. That, in turn, allows the second unused argument of acpi_create_platform_device() to be dropped, which is an improvement. Tested-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-05-25 12:38:52 +00:00
pdev = acpi_create_platform_device(adev);
if (!IS_ERR_OR_NULL(pdev)) {
return 1;
}
ACPI / scan: Add special handler for Intel Lynxpoint LPSS devices Devices on the Intel Lynxpoint Low Power Subsystem (LPSS) have some common features that aren't shared with any other platform devices, including the clock and LTR (Latency Tolerance Reporting) registers. It is better to handle those features in common code than to bother device drivers with doing that (I/O functionality-wise the LPSS devices are generally compatible with other devices that don't have those special registers and may be handled by the same drivers). The clock registers of the LPSS devices are now taken care of by the special clk-x86-lpss driver, but the MMIO mappings used for accessing those registers can also be used for accessing the LTR registers on those devices (LTR support for the Lynxpoint LPSS is going to be added by a subsequent patch). Thus it is convenient to add a special ACPI scan handler for the Lynxpoint LPSS devices that will create the MMIO mappings for accessing the clock (and LTR in the future) registers and will register the LPSS devices' clocks, so the clk-x86-lpss driver will only need to take care of the main Lynxpoint LPSS clock. Introduce a special ACPI scan handler for Intel Lynxpoint LPSS devices as described above. This also reduces overhead related to browsing the ACPI namespace in search of the LPSS devices before the registration of their clocks, removes some LPSS-specific (and somewhat ugly) code from acpi_platform.c and shrinks the overall code size slightly. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Acked-by: Mike Turquette <mturquette@linaro.org>
2013-03-06 22:46:20 +00:00
ACPI / platform / LPSS: Enable async suspend/resume of LPSS devices To seed up suspend and resume of devices included into Intel SoCs handled by the ACPI LPSS driver during system suspend, make acpi_lpss_create_device() call device_enable_async_suspend() for every device created by it. This requires acpi_create_platform_device() to be modified to return a pointer to struct platform_device instead of an int. As a result, acpi_create_platform_device() cannot be pointed to by the .attach pointer in platform_handler directly any more, so a simple wrapper around it is necessary for this purpose. That, in turn, allows the second unused argument of acpi_create_platform_device() to be dropped, which is an improvement. Tested-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-05-25 12:38:52 +00:00
ret = PTR_ERR(pdev);
ACPI / scan: Add special handler for Intel Lynxpoint LPSS devices Devices on the Intel Lynxpoint Low Power Subsystem (LPSS) have some common features that aren't shared with any other platform devices, including the clock and LTR (Latency Tolerance Reporting) registers. It is better to handle those features in common code than to bother device drivers with doing that (I/O functionality-wise the LPSS devices are generally compatible with other devices that don't have those special registers and may be handled by the same drivers). The clock registers of the LPSS devices are now taken care of by the special clk-x86-lpss driver, but the MMIO mappings used for accessing those registers can also be used for accessing the LTR registers on those devices (LTR support for the Lynxpoint LPSS is going to be added by a subsequent patch). Thus it is convenient to add a special ACPI scan handler for the Lynxpoint LPSS devices that will create the MMIO mappings for accessing the clock (and LTR in the future) registers and will register the LPSS devices' clocks, so the clk-x86-lpss driver will only need to take care of the main Lynxpoint LPSS clock. Introduce a special ACPI scan handler for Intel Lynxpoint LPSS devices as described above. This also reduces overhead related to browsing the ACPI namespace in search of the LPSS devices before the registration of their clocks, removes some LPSS-specific (and somewhat ugly) code from acpi_platform.c and shrinks the overall code size slightly. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Acked-by: Mike Turquette <mturquette@linaro.org>
2013-03-06 22:46:20 +00:00
adev->driver_data = NULL;
err_out:
kfree(pdata);
return ret;
}
ACPI / LPSS: Support for device latency tolerance PM QoS Add a new routine, acpi_lpss_set_ltr(), for setting latency tolerance values for LPSS devices having LTR (Latency Tolerance Reporting) registers. Add .bind()/.unbind() callbacks to lpss_handler to set the LPSS devices' power.set_latency_tolerance callback pointers to acpi_lpss_set_ltr() during device addition and to clear them on device removal, respectively. That will cause the device latency tolerance PM QoS to work for the devices in question as documented. This changeset includes a fix from Mika Westerberg. Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-02-10 23:35:53 +00:00
static u32 __lpss_reg_read(struct lpss_private_data *pdata, unsigned int reg)
{
return readl(pdata->mmio_base + pdata->dev_desc->prv_offset + reg);
}
static void __lpss_reg_write(u32 val, struct lpss_private_data *pdata,
unsigned int reg)
{
writel(val, pdata->mmio_base + pdata->dev_desc->prv_offset + reg);
}
ACPI / LPSS: Add support for exposing LTR registers to user space Devices on the Intel Lynxpoint Low Power Subsystem (LPSS) have registers providing access to LTR (Latency Tolerance Reporting) functionality that allows software to monitor and possibly influence the aggressiveness of the platform's active-state power management. For each LPSS device, there are two modes of operation related to LTR, the auto mode and the software mode. In the auto mode the LTR is set up by the platform firmware and managed by hardware. Software can only read the LTR register values to monitor the platform's behavior. In the software mode it is possible to use LTR to control the extent to which the platform will use its built-in power management features. This changeset adds support for reading the LPSS devices' LTR registers and exposing their values to user space for monitoring and diagnostics purposes. It re-uses the MMIO mappings created to access the LPSS devices' clock registers for reading the values of the LTR registers and exposes them to user space through sysfs device attributes. Namely, a new atrribute group, lpss_ltr, is created for each LPSS device. It contains three new attributes: ltr_mode, auto_ltr, sw_ltr. The value of the ltr_mode attribute reflects the LTR mode being used at the moment (software vs auto) and the other two contain the actual register values (raw) whose meaning depends on the LTR mode. All of these attributes are read-only. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2013-03-06 22:46:28 +00:00
static int lpss_reg_read(struct device *dev, unsigned int reg, u32 *val)
{
struct acpi_device *adev;
struct lpss_private_data *pdata;
unsigned long flags;
int ret;
ret = acpi_bus_get_device(ACPI_HANDLE(dev), &adev);
if (WARN_ON(ret))
return ret;
spin_lock_irqsave(&dev->power.lock, flags);
if (pm_runtime_suspended(dev)) {
ret = -EAGAIN;
goto out;
}
pdata = acpi_driver_data(adev);
if (WARN_ON(!pdata || !pdata->mmio_base)) {
ret = -ENODEV;
goto out;
}
ACPI / LPSS: Support for device latency tolerance PM QoS Add a new routine, acpi_lpss_set_ltr(), for setting latency tolerance values for LPSS devices having LTR (Latency Tolerance Reporting) registers. Add .bind()/.unbind() callbacks to lpss_handler to set the LPSS devices' power.set_latency_tolerance callback pointers to acpi_lpss_set_ltr() during device addition and to clear them on device removal, respectively. That will cause the device latency tolerance PM QoS to work for the devices in question as documented. This changeset includes a fix from Mika Westerberg. Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-02-10 23:35:53 +00:00
*val = __lpss_reg_read(pdata, reg);
ACPI / LPSS: Add support for exposing LTR registers to user space Devices on the Intel Lynxpoint Low Power Subsystem (LPSS) have registers providing access to LTR (Latency Tolerance Reporting) functionality that allows software to monitor and possibly influence the aggressiveness of the platform's active-state power management. For each LPSS device, there are two modes of operation related to LTR, the auto mode and the software mode. In the auto mode the LTR is set up by the platform firmware and managed by hardware. Software can only read the LTR register values to monitor the platform's behavior. In the software mode it is possible to use LTR to control the extent to which the platform will use its built-in power management features. This changeset adds support for reading the LPSS devices' LTR registers and exposing their values to user space for monitoring and diagnostics purposes. It re-uses the MMIO mappings created to access the LPSS devices' clock registers for reading the values of the LTR registers and exposes them to user space through sysfs device attributes. Namely, a new atrribute group, lpss_ltr, is created for each LPSS device. It contains three new attributes: ltr_mode, auto_ltr, sw_ltr. The value of the ltr_mode attribute reflects the LTR mode being used at the moment (software vs auto) and the other two contain the actual register values (raw) whose meaning depends on the LTR mode. All of these attributes are read-only. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2013-03-06 22:46:28 +00:00
out:
spin_unlock_irqrestore(&dev->power.lock, flags);
return ret;
}
static ssize_t lpss_ltr_show(struct device *dev, struct device_attribute *attr,
char *buf)
{
u32 ltr_value = 0;
unsigned int reg;
int ret;
reg = strcmp(attr->attr.name, "auto_ltr") ? LPSS_SW_LTR : LPSS_AUTO_LTR;
ret = lpss_reg_read(dev, reg, &ltr_value);
if (ret)
return ret;
return snprintf(buf, PAGE_SIZE, "%08x\n", ltr_value);
}
static ssize_t lpss_ltr_mode_show(struct device *dev,
struct device_attribute *attr, char *buf)
{
u32 ltr_mode = 0;
char *outstr;
int ret;
ret = lpss_reg_read(dev, LPSS_GENERAL, &ltr_mode);
if (ret)
return ret;
outstr = (ltr_mode & LPSS_GENERAL_LTR_MODE_SW) ? "sw" : "auto";
return sprintf(buf, "%s\n", outstr);
}
static DEVICE_ATTR(auto_ltr, S_IRUSR, lpss_ltr_show, NULL);
static DEVICE_ATTR(sw_ltr, S_IRUSR, lpss_ltr_show, NULL);
static DEVICE_ATTR(ltr_mode, S_IRUSR, lpss_ltr_mode_show, NULL);
static struct attribute *lpss_attrs[] = {
&dev_attr_auto_ltr.attr,
&dev_attr_sw_ltr.attr,
&dev_attr_ltr_mode.attr,
NULL,
};
static struct attribute_group lpss_attr_group = {
.attrs = lpss_attrs,
.name = "lpss_ltr",
};
ACPI / LPSS: Support for device latency tolerance PM QoS Add a new routine, acpi_lpss_set_ltr(), for setting latency tolerance values for LPSS devices having LTR (Latency Tolerance Reporting) registers. Add .bind()/.unbind() callbacks to lpss_handler to set the LPSS devices' power.set_latency_tolerance callback pointers to acpi_lpss_set_ltr() during device addition and to clear them on device removal, respectively. That will cause the device latency tolerance PM QoS to work for the devices in question as documented. This changeset includes a fix from Mika Westerberg. Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-02-10 23:35:53 +00:00
static void acpi_lpss_set_ltr(struct device *dev, s32 val)
{
struct lpss_private_data *pdata = acpi_driver_data(ACPI_COMPANION(dev));
u32 ltr_mode, ltr_val;
ltr_mode = __lpss_reg_read(pdata, LPSS_GENERAL);
if (val < 0) {
if (ltr_mode & LPSS_GENERAL_LTR_MODE_SW) {
ltr_mode &= ~LPSS_GENERAL_LTR_MODE_SW;
__lpss_reg_write(ltr_mode, pdata, LPSS_GENERAL);
}
return;
}
ltr_val = __lpss_reg_read(pdata, LPSS_SW_LTR) & ~LPSS_LTR_SNOOP_MASK;
if (val >= LPSS_LTR_SNOOP_LAT_CUTOFF) {
ltr_val |= LPSS_LTR_SNOOP_LAT_32US;
val = LPSS_LTR_MAX_VAL;
} else if (val > LPSS_LTR_MAX_VAL) {
ltr_val |= LPSS_LTR_SNOOP_LAT_32US | LPSS_LTR_SNOOP_REQ;
val >>= LPSS_LTR_SNOOP_LAT_SHIFT;
} else {
ltr_val |= LPSS_LTR_SNOOP_LAT_1US | LPSS_LTR_SNOOP_REQ;
}
ltr_val |= val;
__lpss_reg_write(ltr_val, pdata, LPSS_SW_LTR);
if (!(ltr_mode & LPSS_GENERAL_LTR_MODE_SW)) {
ltr_mode |= LPSS_GENERAL_LTR_MODE_SW;
__lpss_reg_write(ltr_mode, pdata, LPSS_GENERAL);
}
}
ACPI / LPSS: custom power domain for LPSS A power domain where we save the context of the additional LPSS registers. We need to do this or all LPSS devices are left in reset state when resuming from D3 on some Baytrails. The devices with the fractional clock divider also have zeros for N and M values after resuming unless they are reset. Li Aubrey found the root cause for the issue. The idea of using power domain for LPSS came from Mika Westerberg. Reported-by: Jin Yao <yao.jin@linux.intel.com> Suggested-by: Li Aubrey <aubrey.li@linux.intel.com> Suggested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Tested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> [rjw: Added the .complete() callback to the PM domain, fixed build warning on 32-bit.] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-05-23 13:15:09 +00:00
#ifdef CONFIG_PM
/**
* acpi_lpss_save_ctx() - Save the private registers of LPSS device
* @dev: LPSS device
ACPI / LPSS: add all LPSS devices to the specific power domain Currently the LPSS devices are located in the different power domains depends on LPSS_SAVE_CTX flag. We would like to use the specific power domain for all LPSS devices. The LPSS DMA controller has no knobs to control its power state. The specific power domain implementation will handle this case. The patch does a preparation for that. Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Tested-by: Scott Ashcroft <scott.ashcroft@talk21.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-11-05 16:34:45 +00:00
* @pdata: pointer to the private data of the LPSS device
ACPI / LPSS: custom power domain for LPSS A power domain where we save the context of the additional LPSS registers. We need to do this or all LPSS devices are left in reset state when resuming from D3 on some Baytrails. The devices with the fractional clock divider also have zeros for N and M values after resuming unless they are reset. Li Aubrey found the root cause for the issue. The idea of using power domain for LPSS came from Mika Westerberg. Reported-by: Jin Yao <yao.jin@linux.intel.com> Suggested-by: Li Aubrey <aubrey.li@linux.intel.com> Suggested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Tested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> [rjw: Added the .complete() callback to the PM domain, fixed build warning on 32-bit.] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-05-23 13:15:09 +00:00
*
* Most LPSS devices have private registers which may loose their context when
* the device is powered down. acpi_lpss_save_ctx() saves those registers into
* prv_reg_ctx array.
*/
ACPI / LPSS: add all LPSS devices to the specific power domain Currently the LPSS devices are located in the different power domains depends on LPSS_SAVE_CTX flag. We would like to use the specific power domain for all LPSS devices. The LPSS DMA controller has no knobs to control its power state. The specific power domain implementation will handle this case. The patch does a preparation for that. Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Tested-by: Scott Ashcroft <scott.ashcroft@talk21.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-11-05 16:34:45 +00:00
static void acpi_lpss_save_ctx(struct device *dev,
struct lpss_private_data *pdata)
ACPI / LPSS: custom power domain for LPSS A power domain where we save the context of the additional LPSS registers. We need to do this or all LPSS devices are left in reset state when resuming from D3 on some Baytrails. The devices with the fractional clock divider also have zeros for N and M values after resuming unless they are reset. Li Aubrey found the root cause for the issue. The idea of using power domain for LPSS came from Mika Westerberg. Reported-by: Jin Yao <yao.jin@linux.intel.com> Suggested-by: Li Aubrey <aubrey.li@linux.intel.com> Suggested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Tested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> [rjw: Added the .complete() callback to the PM domain, fixed build warning on 32-bit.] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-05-23 13:15:09 +00:00
{
unsigned int i;
for (i = 0; i < LPSS_PRV_REG_COUNT; i++) {
unsigned long offset = i * sizeof(u32);
pdata->prv_reg_ctx[i] = __lpss_reg_read(pdata, offset);
dev_dbg(dev, "saving 0x%08x from LPSS reg at offset 0x%02lx\n",
pdata->prv_reg_ctx[i], offset);
}
}
/**
* acpi_lpss_restore_ctx() - Restore the private registers of LPSS device
* @dev: LPSS device
ACPI / LPSS: add all LPSS devices to the specific power domain Currently the LPSS devices are located in the different power domains depends on LPSS_SAVE_CTX flag. We would like to use the specific power domain for all LPSS devices. The LPSS DMA controller has no knobs to control its power state. The specific power domain implementation will handle this case. The patch does a preparation for that. Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Tested-by: Scott Ashcroft <scott.ashcroft@talk21.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-11-05 16:34:45 +00:00
* @pdata: pointer to the private data of the LPSS device
ACPI / LPSS: custom power domain for LPSS A power domain where we save the context of the additional LPSS registers. We need to do this or all LPSS devices are left in reset state when resuming from D3 on some Baytrails. The devices with the fractional clock divider also have zeros for N and M values after resuming unless they are reset. Li Aubrey found the root cause for the issue. The idea of using power domain for LPSS came from Mika Westerberg. Reported-by: Jin Yao <yao.jin@linux.intel.com> Suggested-by: Li Aubrey <aubrey.li@linux.intel.com> Suggested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Tested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> [rjw: Added the .complete() callback to the PM domain, fixed build warning on 32-bit.] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-05-23 13:15:09 +00:00
*
* Restores the registers that were previously stored with acpi_lpss_save_ctx().
*/
ACPI / LPSS: add all LPSS devices to the specific power domain Currently the LPSS devices are located in the different power domains depends on LPSS_SAVE_CTX flag. We would like to use the specific power domain for all LPSS devices. The LPSS DMA controller has no knobs to control its power state. The specific power domain implementation will handle this case. The patch does a preparation for that. Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Tested-by: Scott Ashcroft <scott.ashcroft@talk21.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-11-05 16:34:45 +00:00
static void acpi_lpss_restore_ctx(struct device *dev,
struct lpss_private_data *pdata)
ACPI / LPSS: custom power domain for LPSS A power domain where we save the context of the additional LPSS registers. We need to do this or all LPSS devices are left in reset state when resuming from D3 on some Baytrails. The devices with the fractional clock divider also have zeros for N and M values after resuming unless they are reset. Li Aubrey found the root cause for the issue. The idea of using power domain for LPSS came from Mika Westerberg. Reported-by: Jin Yao <yao.jin@linux.intel.com> Suggested-by: Li Aubrey <aubrey.li@linux.intel.com> Suggested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Tested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> [rjw: Added the .complete() callback to the PM domain, fixed build warning on 32-bit.] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-05-23 13:15:09 +00:00
{
unsigned int i;
ACPI / LPSS: do delay for all LPSS devices when D3->D0 The LPSS DMA device has no context to save, though it requires the same delay like the rest of LPSS devices when power state is changed from D3 to D0. Do delay for the DMA device as well. Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-12-04 21:49:21 +00:00
for (i = 0; i < LPSS_PRV_REG_COUNT; i++) {
unsigned long offset = i * sizeof(u32);
__lpss_reg_write(pdata->prv_reg_ctx[i], pdata, offset);
dev_dbg(dev, "restoring 0x%08x to LPSS reg at offset 0x%02lx\n",
pdata->prv_reg_ctx[i], offset);
}
}
static void acpi_lpss_d3_to_d0_delay(struct lpss_private_data *pdata)
{
ACPI / LPSS: custom power domain for LPSS A power domain where we save the context of the additional LPSS registers. We need to do this or all LPSS devices are left in reset state when resuming from D3 on some Baytrails. The devices with the fractional clock divider also have zeros for N and M values after resuming unless they are reset. Li Aubrey found the root cause for the issue. The idea of using power domain for LPSS came from Mika Westerberg. Reported-by: Jin Yao <yao.jin@linux.intel.com> Suggested-by: Li Aubrey <aubrey.li@linux.intel.com> Suggested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Tested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> [rjw: Added the .complete() callback to the PM domain, fixed build warning on 32-bit.] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-05-23 13:15:09 +00:00
/*
* The following delay is needed or the subsequent write operations may
* fail. The LPSS devices are actually PCI devices and the PCI spec
* expects 10ms delay before the device can be accessed after D3 to D0
ACPI / LPSS: Ignore 10ms delay for Braswell LPSS devices in Braswell does not need the default 10ms d3_delay imposed by PCI specification. Removing this unnecessary delay significantly reduces the resume time approximately upto 200ms on this platform. Signed-off-by: Srinidhi Kasagar <srinidhi.kasagar@intel.com> Acked-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-08-27 16:00:55 +00:00
* transition. However some platforms like BSW does not need this delay.
ACPI / LPSS: custom power domain for LPSS A power domain where we save the context of the additional LPSS registers. We need to do this or all LPSS devices are left in reset state when resuming from D3 on some Baytrails. The devices with the fractional clock divider also have zeros for N and M values after resuming unless they are reset. Li Aubrey found the root cause for the issue. The idea of using power domain for LPSS came from Mika Westerberg. Reported-by: Jin Yao <yao.jin@linux.intel.com> Suggested-by: Li Aubrey <aubrey.li@linux.intel.com> Suggested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Tested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> [rjw: Added the .complete() callback to the PM domain, fixed build warning on 32-bit.] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-05-23 13:15:09 +00:00
*/
ACPI / LPSS: Ignore 10ms delay for Braswell LPSS devices in Braswell does not need the default 10ms d3_delay imposed by PCI specification. Removing this unnecessary delay significantly reduces the resume time approximately upto 200ms on this platform. Signed-off-by: Srinidhi Kasagar <srinidhi.kasagar@intel.com> Acked-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-08-27 16:00:55 +00:00
unsigned int delay = 10; /* default 10ms delay */
if (pdata->dev_desc->flags & LPSS_NO_D3_DELAY)
delay = 0;
msleep(delay);
ACPI / LPSS: custom power domain for LPSS A power domain where we save the context of the additional LPSS registers. We need to do this or all LPSS devices are left in reset state when resuming from D3 on some Baytrails. The devices with the fractional clock divider also have zeros for N and M values after resuming unless they are reset. Li Aubrey found the root cause for the issue. The idea of using power domain for LPSS came from Mika Westerberg. Reported-by: Jin Yao <yao.jin@linux.intel.com> Suggested-by: Li Aubrey <aubrey.li@linux.intel.com> Suggested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Tested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> [rjw: Added the .complete() callback to the PM domain, fixed build warning on 32-bit.] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-05-23 13:15:09 +00:00
}
ACPI / LPSS: power on when probe() and otherwise when remove() When LPSS drivers are compiled as a module, which is usually the case, the second probe of that driver may fail because the driver is written in an assumption that device is powered on. That is not the case for all drivers. Moreover we would like not drain power in vain. Implement ->activate() and ->dismiss() callbacks in the ACPI LPSS custom power domain. -------- 8< -------- 8< -------- 8< -------- 8< -------- 8< -------- Case 1: The I2C probe() repeat. /sys/bus/platform/devices/808622C1:00 \_SB_.PCI0.I2C1 [D3hot] /sys/bus/platform/devices/808622C1:01 \_SB_.PCI0.I2C2 [D3hot] /sys/bus/platform/devices/808622C1:02 \_SB_.PCI0.I2C3 [D3hot] /sys/bus/platform/devices/808622C1:03 \_SB_.PCI0.I2C4 [D3hot] /sys/bus/platform/devices/808622C1:05 \_SB_.PCI0.I2C6 [D3hot] /sys/bus/platform/devices/808622C1:06 \_SB_.PCI0.I2C7 [D3hot] % modprobe i2c-designware-platform i2c_designware 808622C1:00: Unknown Synopsys component type: 0xffffffff i2c_designware 808622C1:01: Unknown Synopsys component type: 0xffffffff i2c_designware 808622C1:02: Unknown Synopsys component type: 0xffffffff i2c_designware 808622C1:03: Unknown Synopsys component type: 0xffffffff i2c_designware 808622C1:05: Unknown Synopsys component type: 0xffffffff i2c_designware 808622C1:06: Unknown Synopsys component type: 0xffffffff Case 2: The power drain in case of SDHCI. /sys/bus/platform/devices/80860F14:00 \_SB_.PCI0.SDHA [D3hot] /sys/bus/platform/devices/80860F14:01 \_SB_.PCI0.SDHC [D3hot] % modprobe -r sdhci-acpi mmc0: card 0001 removed /sys/bus/platform/devices/80860F14:00 \_SB_.PCI0.SDHA [D0] /sys/bus/platform/devices/80860F14:01 \_SB_.PCI0.SDHC [D0] -------- 8< -------- 8< -------- 8< -------- 8< -------- 8< -------- Patch fixes above problems. Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-12-04 21:49:20 +00:00
static int acpi_lpss_activate(struct device *dev)
{
struct lpss_private_data *pdata = acpi_driver_data(ACPI_COMPANION(dev));
int ret;
ret = acpi_dev_runtime_resume(dev);
if (ret)
return ret;
acpi_lpss_d3_to_d0_delay(pdata);
/*
* This is called only on ->probe() stage where a device is either in
* known state defined by BIOS or most likely powered off. Due to this
* we have to deassert reset line to be sure that ->probe() will
* recognize the device.
*/
if (pdata->dev_desc->flags & LPSS_SAVE_CTX)
lpss_deassert_reset(pdata);
return 0;
}
static void acpi_lpss_dismiss(struct device *dev)
{
acpi_dev_runtime_suspend(dev);
}
ACPI / LPSS: custom power domain for LPSS A power domain where we save the context of the additional LPSS registers. We need to do this or all LPSS devices are left in reset state when resuming from D3 on some Baytrails. The devices with the fractional clock divider also have zeros for N and M values after resuming unless they are reset. Li Aubrey found the root cause for the issue. The idea of using power domain for LPSS came from Mika Westerberg. Reported-by: Jin Yao <yao.jin@linux.intel.com> Suggested-by: Li Aubrey <aubrey.li@linux.intel.com> Suggested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Tested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> [rjw: Added the .complete() callback to the PM domain, fixed build warning on 32-bit.] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-05-23 13:15:09 +00:00
#ifdef CONFIG_PM_SLEEP
static int acpi_lpss_suspend_late(struct device *dev)
{
ACPI / LPSS: add all LPSS devices to the specific power domain Currently the LPSS devices are located in the different power domains depends on LPSS_SAVE_CTX flag. We would like to use the specific power domain for all LPSS devices. The LPSS DMA controller has no knobs to control its power state. The specific power domain implementation will handle this case. The patch does a preparation for that. Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Tested-by: Scott Ashcroft <scott.ashcroft@talk21.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-11-05 16:34:45 +00:00
struct lpss_private_data *pdata = acpi_driver_data(ACPI_COMPANION(dev));
int ret;
ACPI / LPSS: custom power domain for LPSS A power domain where we save the context of the additional LPSS registers. We need to do this or all LPSS devices are left in reset state when resuming from D3 on some Baytrails. The devices with the fractional clock divider also have zeros for N and M values after resuming unless they are reset. Li Aubrey found the root cause for the issue. The idea of using power domain for LPSS came from Mika Westerberg. Reported-by: Jin Yao <yao.jin@linux.intel.com> Suggested-by: Li Aubrey <aubrey.li@linux.intel.com> Suggested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Tested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> [rjw: Added the .complete() callback to the PM domain, fixed build warning on 32-bit.] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-05-23 13:15:09 +00:00
ACPI / LPSS: add all LPSS devices to the specific power domain Currently the LPSS devices are located in the different power domains depends on LPSS_SAVE_CTX flag. We would like to use the specific power domain for all LPSS devices. The LPSS DMA controller has no knobs to control its power state. The specific power domain implementation will handle this case. The patch does a preparation for that. Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Tested-by: Scott Ashcroft <scott.ashcroft@talk21.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-11-05 16:34:45 +00:00
ret = pm_generic_suspend_late(dev);
ACPI / LPSS: custom power domain for LPSS A power domain where we save the context of the additional LPSS registers. We need to do this or all LPSS devices are left in reset state when resuming from D3 on some Baytrails. The devices with the fractional clock divider also have zeros for N and M values after resuming unless they are reset. Li Aubrey found the root cause for the issue. The idea of using power domain for LPSS came from Mika Westerberg. Reported-by: Jin Yao <yao.jin@linux.intel.com> Suggested-by: Li Aubrey <aubrey.li@linux.intel.com> Suggested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Tested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> [rjw: Added the .complete() callback to the PM domain, fixed build warning on 32-bit.] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-05-23 13:15:09 +00:00
if (ret)
return ret;
ACPI / LPSS: add all LPSS devices to the specific power domain Currently the LPSS devices are located in the different power domains depends on LPSS_SAVE_CTX flag. We would like to use the specific power domain for all LPSS devices. The LPSS DMA controller has no knobs to control its power state. The specific power domain implementation will handle this case. The patch does a preparation for that. Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Tested-by: Scott Ashcroft <scott.ashcroft@talk21.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-11-05 16:34:45 +00:00
if (pdata->dev_desc->flags & LPSS_SAVE_CTX)
acpi_lpss_save_ctx(dev, pdata);
ACPI / LPSS: custom power domain for LPSS A power domain where we save the context of the additional LPSS registers. We need to do this or all LPSS devices are left in reset state when resuming from D3 on some Baytrails. The devices with the fractional clock divider also have zeros for N and M values after resuming unless they are reset. Li Aubrey found the root cause for the issue. The idea of using power domain for LPSS came from Mika Westerberg. Reported-by: Jin Yao <yao.jin@linux.intel.com> Suggested-by: Li Aubrey <aubrey.li@linux.intel.com> Suggested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Tested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> [rjw: Added the .complete() callback to the PM domain, fixed build warning on 32-bit.] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-05-23 13:15:09 +00:00
return acpi_dev_suspend_late(dev);
}
ACPI / LPSS: complete PM entries for LPSS power domain PM entries of LPSS power domain were not implemented correctly in commit c78b0830667a "ACPI / LPSS: custom power domain for LPSS". This patch fixes and completes these PM entries. Fixes: c78b0830667a (ACPI / LPSS: custom power domain for LPSS) Signed-off-by: Li Aubrey <aubrey.li@linux.intel.com> Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Fu Zhonghui <zhonghui.fu@linux.intel.com> Cc: 3.16+ <stable@vger.kernel.org> # 3.16+ Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-09-09 14:30:06 +00:00
static int acpi_lpss_resume_early(struct device *dev)
ACPI / LPSS: custom power domain for LPSS A power domain where we save the context of the additional LPSS registers. We need to do this or all LPSS devices are left in reset state when resuming from D3 on some Baytrails. The devices with the fractional clock divider also have zeros for N and M values after resuming unless they are reset. Li Aubrey found the root cause for the issue. The idea of using power domain for LPSS came from Mika Westerberg. Reported-by: Jin Yao <yao.jin@linux.intel.com> Suggested-by: Li Aubrey <aubrey.li@linux.intel.com> Suggested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Tested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> [rjw: Added the .complete() callback to the PM domain, fixed build warning on 32-bit.] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-05-23 13:15:09 +00:00
{
ACPI / LPSS: add all LPSS devices to the specific power domain Currently the LPSS devices are located in the different power domains depends on LPSS_SAVE_CTX flag. We would like to use the specific power domain for all LPSS devices. The LPSS DMA controller has no knobs to control its power state. The specific power domain implementation will handle this case. The patch does a preparation for that. Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Tested-by: Scott Ashcroft <scott.ashcroft@talk21.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-11-05 16:34:45 +00:00
struct lpss_private_data *pdata = acpi_driver_data(ACPI_COMPANION(dev));
int ret;
ACPI / LPSS: custom power domain for LPSS A power domain where we save the context of the additional LPSS registers. We need to do this or all LPSS devices are left in reset state when resuming from D3 on some Baytrails. The devices with the fractional clock divider also have zeros for N and M values after resuming unless they are reset. Li Aubrey found the root cause for the issue. The idea of using power domain for LPSS came from Mika Westerberg. Reported-by: Jin Yao <yao.jin@linux.intel.com> Suggested-by: Li Aubrey <aubrey.li@linux.intel.com> Suggested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Tested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> [rjw: Added the .complete() callback to the PM domain, fixed build warning on 32-bit.] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-05-23 13:15:09 +00:00
ACPI / LPSS: add all LPSS devices to the specific power domain Currently the LPSS devices are located in the different power domains depends on LPSS_SAVE_CTX flag. We would like to use the specific power domain for all LPSS devices. The LPSS DMA controller has no knobs to control its power state. The specific power domain implementation will handle this case. The patch does a preparation for that. Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Tested-by: Scott Ashcroft <scott.ashcroft@talk21.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-11-05 16:34:45 +00:00
ret = acpi_dev_resume_early(dev);
ACPI / LPSS: custom power domain for LPSS A power domain where we save the context of the additional LPSS registers. We need to do this or all LPSS devices are left in reset state when resuming from D3 on some Baytrails. The devices with the fractional clock divider also have zeros for N and M values after resuming unless they are reset. Li Aubrey found the root cause for the issue. The idea of using power domain for LPSS came from Mika Westerberg. Reported-by: Jin Yao <yao.jin@linux.intel.com> Suggested-by: Li Aubrey <aubrey.li@linux.intel.com> Suggested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Tested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> [rjw: Added the .complete() callback to the PM domain, fixed build warning on 32-bit.] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-05-23 13:15:09 +00:00
if (ret)
return ret;
ACPI / LPSS: do delay for all LPSS devices when D3->D0 The LPSS DMA device has no context to save, though it requires the same delay like the rest of LPSS devices when power state is changed from D3 to D0. Do delay for the DMA device as well. Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-12-04 21:49:21 +00:00
acpi_lpss_d3_to_d0_delay(pdata);
ACPI / LPSS: add all LPSS devices to the specific power domain Currently the LPSS devices are located in the different power domains depends on LPSS_SAVE_CTX flag. We would like to use the specific power domain for all LPSS devices. The LPSS DMA controller has no knobs to control its power state. The specific power domain implementation will handle this case. The patch does a preparation for that. Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Tested-by: Scott Ashcroft <scott.ashcroft@talk21.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-11-05 16:34:45 +00:00
if (pdata->dev_desc->flags & LPSS_SAVE_CTX)
acpi_lpss_restore_ctx(dev, pdata);
ACPI / LPSS: custom power domain for LPSS A power domain where we save the context of the additional LPSS registers. We need to do this or all LPSS devices are left in reset state when resuming from D3 on some Baytrails. The devices with the fractional clock divider also have zeros for N and M values after resuming unless they are reset. Li Aubrey found the root cause for the issue. The idea of using power domain for LPSS came from Mika Westerberg. Reported-by: Jin Yao <yao.jin@linux.intel.com> Suggested-by: Li Aubrey <aubrey.li@linux.intel.com> Suggested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Tested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> [rjw: Added the .complete() callback to the PM domain, fixed build warning on 32-bit.] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-05-23 13:15:09 +00:00
return pm_generic_resume_early(dev);
}
#endif /* CONFIG_PM_SLEEP */
ACPI / LPSS: override power state for LPSS DMA device This is a third approach to workaround long standing issue with LPSS on BayTrail. First one [1] was reverted since it didn't resolve the issue comprehensively. Second one [2] was rejected by internal review. The LPSS DMA controller does not have neither _PS0 nor _PS3 method. Moreover it can be powered off automatically whenever the last LPSS device goes down. In case of no power any access to the DMA controller will hang the system. The behaviour is reproduced on some HP laptops based on Intel BayTrail [3,4] as well as on ASuS T100TA transformer. Power on the LPSS island through the registers accessible in a specific way. [1] http://www.spinics.net/lists/linux-acpi/msg53963.html [2] https://bugzilla.redhat.com/attachment.cgi?id=1066779&action=diff [3] https://bugzilla.redhat.com/show_bug.cgi?id=1184273 [4] http://www.spinics.net/lists/dmaengine/msg01514.html Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-12-12 01:45:06 +00:00
/* IOSF SB for LPSS island */
#define LPSS_IOSF_UNIT_LPIOEP 0xA0
#define LPSS_IOSF_UNIT_LPIO1 0xAB
#define LPSS_IOSF_UNIT_LPIO2 0xAC
#define LPSS_IOSF_PMCSR 0x84
#define LPSS_PMCSR_D0 0
#define LPSS_PMCSR_D3hot 3
#define LPSS_PMCSR_Dx_MASK GENMASK(1, 0)
#define LPSS_IOSF_GPIODEF0 0x154
#define LPSS_GPIODEF0_DMA1_D3 BIT(2)
#define LPSS_GPIODEF0_DMA2_D3 BIT(3)
#define LPSS_GPIODEF0_DMA_D3_MASK GENMASK(3, 2)
static DEFINE_MUTEX(lpss_iosf_mutex);
static void lpss_iosf_enter_d3_state(void)
{
u32 value1 = 0;
u32 mask1 = LPSS_GPIODEF0_DMA_D3_MASK;
u32 value2 = LPSS_PMCSR_D3hot;
u32 mask2 = LPSS_PMCSR_Dx_MASK;
/*
* PMC provides an information about actual status of the LPSS devices.
* Here we read the values related to LPSS power island, i.e. LPSS
* devices, excluding both LPSS DMA controllers, along with SCC domain.
*/
u32 func_dis, d3_sts_0, pmc_status, pmc_mask = 0xfe000ffe;
int ret;
ret = pmc_atom_read(PMC_FUNC_DIS, &func_dis);
if (ret)
return;
mutex_lock(&lpss_iosf_mutex);
ret = pmc_atom_read(PMC_D3_STS_0, &d3_sts_0);
if (ret)
goto exit;
/*
* Get the status of entire LPSS power island per device basis.
* Shutdown both LPSS DMA controllers if and only if all other devices
* are already in D3hot.
*/
pmc_status = (~(d3_sts_0 | func_dis)) & pmc_mask;
if (pmc_status)
goto exit;
iosf_mbi_modify(LPSS_IOSF_UNIT_LPIO1, MBI_CFG_WRITE,
LPSS_IOSF_PMCSR, value2, mask2);
iosf_mbi_modify(LPSS_IOSF_UNIT_LPIO2, MBI_CFG_WRITE,
LPSS_IOSF_PMCSR, value2, mask2);
iosf_mbi_modify(LPSS_IOSF_UNIT_LPIOEP, MBI_CR_WRITE,
LPSS_IOSF_GPIODEF0, value1, mask1);
exit:
mutex_unlock(&lpss_iosf_mutex);
}
static void lpss_iosf_exit_d3_state(void)
{
u32 value1 = LPSS_GPIODEF0_DMA1_D3 | LPSS_GPIODEF0_DMA2_D3;
u32 mask1 = LPSS_GPIODEF0_DMA_D3_MASK;
u32 value2 = LPSS_PMCSR_D0;
u32 mask2 = LPSS_PMCSR_Dx_MASK;
mutex_lock(&lpss_iosf_mutex);
iosf_mbi_modify(LPSS_IOSF_UNIT_LPIOEP, MBI_CR_WRITE,
LPSS_IOSF_GPIODEF0, value1, mask1);
iosf_mbi_modify(LPSS_IOSF_UNIT_LPIO2, MBI_CFG_WRITE,
LPSS_IOSF_PMCSR, value2, mask2);
iosf_mbi_modify(LPSS_IOSF_UNIT_LPIO1, MBI_CFG_WRITE,
LPSS_IOSF_PMCSR, value2, mask2);
mutex_unlock(&lpss_iosf_mutex);
}
ACPI / LPSS: custom power domain for LPSS A power domain where we save the context of the additional LPSS registers. We need to do this or all LPSS devices are left in reset state when resuming from D3 on some Baytrails. The devices with the fractional clock divider also have zeros for N and M values after resuming unless they are reset. Li Aubrey found the root cause for the issue. The idea of using power domain for LPSS came from Mika Westerberg. Reported-by: Jin Yao <yao.jin@linux.intel.com> Suggested-by: Li Aubrey <aubrey.li@linux.intel.com> Suggested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Tested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> [rjw: Added the .complete() callback to the PM domain, fixed build warning on 32-bit.] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-05-23 13:15:09 +00:00
static int acpi_lpss_runtime_suspend(struct device *dev)
{
ACPI / LPSS: add all LPSS devices to the specific power domain Currently the LPSS devices are located in the different power domains depends on LPSS_SAVE_CTX flag. We would like to use the specific power domain for all LPSS devices. The LPSS DMA controller has no knobs to control its power state. The specific power domain implementation will handle this case. The patch does a preparation for that. Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Tested-by: Scott Ashcroft <scott.ashcroft@talk21.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-11-05 16:34:45 +00:00
struct lpss_private_data *pdata = acpi_driver_data(ACPI_COMPANION(dev));
int ret;
ACPI / LPSS: custom power domain for LPSS A power domain where we save the context of the additional LPSS registers. We need to do this or all LPSS devices are left in reset state when resuming from D3 on some Baytrails. The devices with the fractional clock divider also have zeros for N and M values after resuming unless they are reset. Li Aubrey found the root cause for the issue. The idea of using power domain for LPSS came from Mika Westerberg. Reported-by: Jin Yao <yao.jin@linux.intel.com> Suggested-by: Li Aubrey <aubrey.li@linux.intel.com> Suggested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Tested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> [rjw: Added the .complete() callback to the PM domain, fixed build warning on 32-bit.] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-05-23 13:15:09 +00:00
ACPI / LPSS: add all LPSS devices to the specific power domain Currently the LPSS devices are located in the different power domains depends on LPSS_SAVE_CTX flag. We would like to use the specific power domain for all LPSS devices. The LPSS DMA controller has no knobs to control its power state. The specific power domain implementation will handle this case. The patch does a preparation for that. Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Tested-by: Scott Ashcroft <scott.ashcroft@talk21.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-11-05 16:34:45 +00:00
ret = pm_generic_runtime_suspend(dev);
ACPI / LPSS: custom power domain for LPSS A power domain where we save the context of the additional LPSS registers. We need to do this or all LPSS devices are left in reset state when resuming from D3 on some Baytrails. The devices with the fractional clock divider also have zeros for N and M values after resuming unless they are reset. Li Aubrey found the root cause for the issue. The idea of using power domain for LPSS came from Mika Westerberg. Reported-by: Jin Yao <yao.jin@linux.intel.com> Suggested-by: Li Aubrey <aubrey.li@linux.intel.com> Suggested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Tested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> [rjw: Added the .complete() callback to the PM domain, fixed build warning on 32-bit.] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-05-23 13:15:09 +00:00
if (ret)
return ret;
ACPI / LPSS: add all LPSS devices to the specific power domain Currently the LPSS devices are located in the different power domains depends on LPSS_SAVE_CTX flag. We would like to use the specific power domain for all LPSS devices. The LPSS DMA controller has no knobs to control its power state. The specific power domain implementation will handle this case. The patch does a preparation for that. Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Tested-by: Scott Ashcroft <scott.ashcroft@talk21.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-11-05 16:34:45 +00:00
if (pdata->dev_desc->flags & LPSS_SAVE_CTX)
acpi_lpss_save_ctx(dev, pdata);
ACPI / LPSS: override power state for LPSS DMA device This is a third approach to workaround long standing issue with LPSS on BayTrail. First one [1] was reverted since it didn't resolve the issue comprehensively. Second one [2] was rejected by internal review. The LPSS DMA controller does not have neither _PS0 nor _PS3 method. Moreover it can be powered off automatically whenever the last LPSS device goes down. In case of no power any access to the DMA controller will hang the system. The behaviour is reproduced on some HP laptops based on Intel BayTrail [3,4] as well as on ASuS T100TA transformer. Power on the LPSS island through the registers accessible in a specific way. [1] http://www.spinics.net/lists/linux-acpi/msg53963.html [2] https://bugzilla.redhat.com/attachment.cgi?id=1066779&action=diff [3] https://bugzilla.redhat.com/show_bug.cgi?id=1184273 [4] http://www.spinics.net/lists/dmaengine/msg01514.html Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-12-12 01:45:06 +00:00
ret = acpi_dev_runtime_suspend(dev);
/*
* This call must be last in the sequence, otherwise PMC will return
* wrong status for devices being about to be powered off. See
* lpss_iosf_enter_d3_state() for further information.
*/
if (lpss_quirks & LPSS_QUIRK_ALWAYS_POWER_ON && iosf_mbi_available())
lpss_iosf_enter_d3_state();
return ret;
ACPI / LPSS: custom power domain for LPSS A power domain where we save the context of the additional LPSS registers. We need to do this or all LPSS devices are left in reset state when resuming from D3 on some Baytrails. The devices with the fractional clock divider also have zeros for N and M values after resuming unless they are reset. Li Aubrey found the root cause for the issue. The idea of using power domain for LPSS came from Mika Westerberg. Reported-by: Jin Yao <yao.jin@linux.intel.com> Suggested-by: Li Aubrey <aubrey.li@linux.intel.com> Suggested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Tested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> [rjw: Added the .complete() callback to the PM domain, fixed build warning on 32-bit.] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-05-23 13:15:09 +00:00
}
static int acpi_lpss_runtime_resume(struct device *dev)
{
ACPI / LPSS: add all LPSS devices to the specific power domain Currently the LPSS devices are located in the different power domains depends on LPSS_SAVE_CTX flag. We would like to use the specific power domain for all LPSS devices. The LPSS DMA controller has no knobs to control its power state. The specific power domain implementation will handle this case. The patch does a preparation for that. Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Tested-by: Scott Ashcroft <scott.ashcroft@talk21.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-11-05 16:34:45 +00:00
struct lpss_private_data *pdata = acpi_driver_data(ACPI_COMPANION(dev));
int ret;
ACPI / LPSS: custom power domain for LPSS A power domain where we save the context of the additional LPSS registers. We need to do this or all LPSS devices are left in reset state when resuming from D3 on some Baytrails. The devices with the fractional clock divider also have zeros for N and M values after resuming unless they are reset. Li Aubrey found the root cause for the issue. The idea of using power domain for LPSS came from Mika Westerberg. Reported-by: Jin Yao <yao.jin@linux.intel.com> Suggested-by: Li Aubrey <aubrey.li@linux.intel.com> Suggested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Tested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> [rjw: Added the .complete() callback to the PM domain, fixed build warning on 32-bit.] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-05-23 13:15:09 +00:00
ACPI / LPSS: override power state for LPSS DMA device This is a third approach to workaround long standing issue with LPSS on BayTrail. First one [1] was reverted since it didn't resolve the issue comprehensively. Second one [2] was rejected by internal review. The LPSS DMA controller does not have neither _PS0 nor _PS3 method. Moreover it can be powered off automatically whenever the last LPSS device goes down. In case of no power any access to the DMA controller will hang the system. The behaviour is reproduced on some HP laptops based on Intel BayTrail [3,4] as well as on ASuS T100TA transformer. Power on the LPSS island through the registers accessible in a specific way. [1] http://www.spinics.net/lists/linux-acpi/msg53963.html [2] https://bugzilla.redhat.com/attachment.cgi?id=1066779&action=diff [3] https://bugzilla.redhat.com/show_bug.cgi?id=1184273 [4] http://www.spinics.net/lists/dmaengine/msg01514.html Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-12-12 01:45:06 +00:00
/*
* This call is kept first to be in symmetry with
* acpi_lpss_runtime_suspend() one.
*/
if (lpss_quirks & LPSS_QUIRK_ALWAYS_POWER_ON && iosf_mbi_available())
lpss_iosf_exit_d3_state();
ACPI / LPSS: add all LPSS devices to the specific power domain Currently the LPSS devices are located in the different power domains depends on LPSS_SAVE_CTX flag. We would like to use the specific power domain for all LPSS devices. The LPSS DMA controller has no knobs to control its power state. The specific power domain implementation will handle this case. The patch does a preparation for that. Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Tested-by: Scott Ashcroft <scott.ashcroft@talk21.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-11-05 16:34:45 +00:00
ret = acpi_dev_runtime_resume(dev);
ACPI / LPSS: custom power domain for LPSS A power domain where we save the context of the additional LPSS registers. We need to do this or all LPSS devices are left in reset state when resuming from D3 on some Baytrails. The devices with the fractional clock divider also have zeros for N and M values after resuming unless they are reset. Li Aubrey found the root cause for the issue. The idea of using power domain for LPSS came from Mika Westerberg. Reported-by: Jin Yao <yao.jin@linux.intel.com> Suggested-by: Li Aubrey <aubrey.li@linux.intel.com> Suggested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Tested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> [rjw: Added the .complete() callback to the PM domain, fixed build warning on 32-bit.] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-05-23 13:15:09 +00:00
if (ret)
return ret;
ACPI / LPSS: do delay for all LPSS devices when D3->D0 The LPSS DMA device has no context to save, though it requires the same delay like the rest of LPSS devices when power state is changed from D3 to D0. Do delay for the DMA device as well. Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-12-04 21:49:21 +00:00
acpi_lpss_d3_to_d0_delay(pdata);
ACPI / LPSS: add all LPSS devices to the specific power domain Currently the LPSS devices are located in the different power domains depends on LPSS_SAVE_CTX flag. We would like to use the specific power domain for all LPSS devices. The LPSS DMA controller has no knobs to control its power state. The specific power domain implementation will handle this case. The patch does a preparation for that. Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Tested-by: Scott Ashcroft <scott.ashcroft@talk21.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-11-05 16:34:45 +00:00
if (pdata->dev_desc->flags & LPSS_SAVE_CTX)
acpi_lpss_restore_ctx(dev, pdata);
ACPI / LPSS: custom power domain for LPSS A power domain where we save the context of the additional LPSS registers. We need to do this or all LPSS devices are left in reset state when resuming from D3 on some Baytrails. The devices with the fractional clock divider also have zeros for N and M values after resuming unless they are reset. Li Aubrey found the root cause for the issue. The idea of using power domain for LPSS came from Mika Westerberg. Reported-by: Jin Yao <yao.jin@linux.intel.com> Suggested-by: Li Aubrey <aubrey.li@linux.intel.com> Suggested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Tested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> [rjw: Added the .complete() callback to the PM domain, fixed build warning on 32-bit.] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-05-23 13:15:09 +00:00
return pm_generic_runtime_resume(dev);
}
#endif /* CONFIG_PM */
static struct dev_pm_domain acpi_lpss_pm_domain = {
ACPI / LPSS: power on when probe() and otherwise when remove() When LPSS drivers are compiled as a module, which is usually the case, the second probe of that driver may fail because the driver is written in an assumption that device is powered on. That is not the case for all drivers. Moreover we would like not drain power in vain. Implement ->activate() and ->dismiss() callbacks in the ACPI LPSS custom power domain. -------- 8< -------- 8< -------- 8< -------- 8< -------- 8< -------- Case 1: The I2C probe() repeat. /sys/bus/platform/devices/808622C1:00 \_SB_.PCI0.I2C1 [D3hot] /sys/bus/platform/devices/808622C1:01 \_SB_.PCI0.I2C2 [D3hot] /sys/bus/platform/devices/808622C1:02 \_SB_.PCI0.I2C3 [D3hot] /sys/bus/platform/devices/808622C1:03 \_SB_.PCI0.I2C4 [D3hot] /sys/bus/platform/devices/808622C1:05 \_SB_.PCI0.I2C6 [D3hot] /sys/bus/platform/devices/808622C1:06 \_SB_.PCI0.I2C7 [D3hot] % modprobe i2c-designware-platform i2c_designware 808622C1:00: Unknown Synopsys component type: 0xffffffff i2c_designware 808622C1:01: Unknown Synopsys component type: 0xffffffff i2c_designware 808622C1:02: Unknown Synopsys component type: 0xffffffff i2c_designware 808622C1:03: Unknown Synopsys component type: 0xffffffff i2c_designware 808622C1:05: Unknown Synopsys component type: 0xffffffff i2c_designware 808622C1:06: Unknown Synopsys component type: 0xffffffff Case 2: The power drain in case of SDHCI. /sys/bus/platform/devices/80860F14:00 \_SB_.PCI0.SDHA [D3hot] /sys/bus/platform/devices/80860F14:01 \_SB_.PCI0.SDHC [D3hot] % modprobe -r sdhci-acpi mmc0: card 0001 removed /sys/bus/platform/devices/80860F14:00 \_SB_.PCI0.SDHA [D0] /sys/bus/platform/devices/80860F14:01 \_SB_.PCI0.SDHC [D0] -------- 8< -------- 8< -------- 8< -------- 8< -------- 8< -------- Patch fixes above problems. Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-12-04 21:49:20 +00:00
#ifdef CONFIG_PM
.activate = acpi_lpss_activate,
.dismiss = acpi_lpss_dismiss,
#endif
ACPI / LPSS: custom power domain for LPSS A power domain where we save the context of the additional LPSS registers. We need to do this or all LPSS devices are left in reset state when resuming from D3 on some Baytrails. The devices with the fractional clock divider also have zeros for N and M values after resuming unless they are reset. Li Aubrey found the root cause for the issue. The idea of using power domain for LPSS came from Mika Westerberg. Reported-by: Jin Yao <yao.jin@linux.intel.com> Suggested-by: Li Aubrey <aubrey.li@linux.intel.com> Suggested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Tested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> [rjw: Added the .complete() callback to the PM domain, fixed build warning on 32-bit.] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-05-23 13:15:09 +00:00
.ops = {
ACPI / PM: Drop CONFIG_PM_RUNTIME from the ACPI core After commit b2b49ccbdd54 (PM: Kconfig: Set PM_RUNTIME if PM_SLEEP is selected) PM_RUNTIME is always set if PM is set, so quite a few depend on CONFIG_PM. Replace CONFIG_PM_RUNTIME with CONFIG_PM in the ACPI core code. Reviewed-by: Ulf Hansson <ulf.hansson@linaro.org> Acked-by: Kevin Hilman <khilman@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-11-27 21:38:23 +00:00
#ifdef CONFIG_PM
ACPI / LPSS: custom power domain for LPSS A power domain where we save the context of the additional LPSS registers. We need to do this or all LPSS devices are left in reset state when resuming from D3 on some Baytrails. The devices with the fractional clock divider also have zeros for N and M values after resuming unless they are reset. Li Aubrey found the root cause for the issue. The idea of using power domain for LPSS came from Mika Westerberg. Reported-by: Jin Yao <yao.jin@linux.intel.com> Suggested-by: Li Aubrey <aubrey.li@linux.intel.com> Suggested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Tested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> [rjw: Added the .complete() callback to the PM domain, fixed build warning on 32-bit.] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-05-23 13:15:09 +00:00
#ifdef CONFIG_PM_SLEEP
.prepare = acpi_subsys_prepare,
PM / PCI / ACPI: Kick devices that might have been reset by firmware There is a concern that if the platform firmware was involved in the system resume that's being completed, some devices might have been reset by it and if those devices had the power.direct_complete flag set during the preceding suspend transition, they may stay in a reset-power-on state indefinitely (until they are runtime-resumed and then suspended again). That may not be a big deal from the individual device's perspective, but if the system is an SoC, it may be prevented from entering deep SoC-wide low-power states on idle because of that. The devices that are most likely to be affected by this issue are PCI devices and ACPI-enumerated devices using the general ACPI PM domain, so to prevent it from happening for those devices, force a runtime resume for them if they have their power.direct_complete flags set and the platform firmware was involved in the resume transition currently in progress. Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-10-06 22:50:24 +00:00
.complete = pm_complete_with_resume_check,
ACPI / LPSS: custom power domain for LPSS A power domain where we save the context of the additional LPSS registers. We need to do this or all LPSS devices are left in reset state when resuming from D3 on some Baytrails. The devices with the fractional clock divider also have zeros for N and M values after resuming unless they are reset. Li Aubrey found the root cause for the issue. The idea of using power domain for LPSS came from Mika Westerberg. Reported-by: Jin Yao <yao.jin@linux.intel.com> Suggested-by: Li Aubrey <aubrey.li@linux.intel.com> Suggested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Tested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> [rjw: Added the .complete() callback to the PM domain, fixed build warning on 32-bit.] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-05-23 13:15:09 +00:00
.suspend = acpi_subsys_suspend,
ACPI / LPSS: complete PM entries for LPSS power domain PM entries of LPSS power domain were not implemented correctly in commit c78b0830667a "ACPI / LPSS: custom power domain for LPSS". This patch fixes and completes these PM entries. Fixes: c78b0830667a (ACPI / LPSS: custom power domain for LPSS) Signed-off-by: Li Aubrey <aubrey.li@linux.intel.com> Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Fu Zhonghui <zhonghui.fu@linux.intel.com> Cc: 3.16+ <stable@vger.kernel.org> # 3.16+ Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-09-09 14:30:06 +00:00
.suspend_late = acpi_lpss_suspend_late,
.resume_early = acpi_lpss_resume_early,
ACPI / LPSS: custom power domain for LPSS A power domain where we save the context of the additional LPSS registers. We need to do this or all LPSS devices are left in reset state when resuming from D3 on some Baytrails. The devices with the fractional clock divider also have zeros for N and M values after resuming unless they are reset. Li Aubrey found the root cause for the issue. The idea of using power domain for LPSS came from Mika Westerberg. Reported-by: Jin Yao <yao.jin@linux.intel.com> Suggested-by: Li Aubrey <aubrey.li@linux.intel.com> Suggested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Tested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> [rjw: Added the .complete() callback to the PM domain, fixed build warning on 32-bit.] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-05-23 13:15:09 +00:00
.freeze = acpi_subsys_freeze,
.poweroff = acpi_subsys_suspend,
ACPI / LPSS: complete PM entries for LPSS power domain PM entries of LPSS power domain were not implemented correctly in commit c78b0830667a "ACPI / LPSS: custom power domain for LPSS". This patch fixes and completes these PM entries. Fixes: c78b0830667a (ACPI / LPSS: custom power domain for LPSS) Signed-off-by: Li Aubrey <aubrey.li@linux.intel.com> Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Fu Zhonghui <zhonghui.fu@linux.intel.com> Cc: 3.16+ <stable@vger.kernel.org> # 3.16+ Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-09-09 14:30:06 +00:00
.poweroff_late = acpi_lpss_suspend_late,
.restore_early = acpi_lpss_resume_early,
ACPI / LPSS: custom power domain for LPSS A power domain where we save the context of the additional LPSS registers. We need to do this or all LPSS devices are left in reset state when resuming from D3 on some Baytrails. The devices with the fractional clock divider also have zeros for N and M values after resuming unless they are reset. Li Aubrey found the root cause for the issue. The idea of using power domain for LPSS came from Mika Westerberg. Reported-by: Jin Yao <yao.jin@linux.intel.com> Suggested-by: Li Aubrey <aubrey.li@linux.intel.com> Suggested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Tested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> [rjw: Added the .complete() callback to the PM domain, fixed build warning on 32-bit.] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-05-23 13:15:09 +00:00
#endif
.runtime_suspend = acpi_lpss_runtime_suspend,
.runtime_resume = acpi_lpss_runtime_resume,
#endif
},
};
ACPI / LPSS: Add support for exposing LTR registers to user space Devices on the Intel Lynxpoint Low Power Subsystem (LPSS) have registers providing access to LTR (Latency Tolerance Reporting) functionality that allows software to monitor and possibly influence the aggressiveness of the platform's active-state power management. For each LPSS device, there are two modes of operation related to LTR, the auto mode and the software mode. In the auto mode the LTR is set up by the platform firmware and managed by hardware. Software can only read the LTR register values to monitor the platform's behavior. In the software mode it is possible to use LTR to control the extent to which the platform will use its built-in power management features. This changeset adds support for reading the LPSS devices' LTR registers and exposing their values to user space for monitoring and diagnostics purposes. It re-uses the MMIO mappings created to access the LPSS devices' clock registers for reading the values of the LTR registers and exposes them to user space through sysfs device attributes. Namely, a new atrribute group, lpss_ltr, is created for each LPSS device. It contains three new attributes: ltr_mode, auto_ltr, sw_ltr. The value of the ltr_mode attribute reflects the LTR mode being used at the moment (software vs auto) and the other two contain the actual register values (raw) whose meaning depends on the LTR mode. All of these attributes are read-only. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2013-03-06 22:46:28 +00:00
static int acpi_lpss_platform_notify(struct notifier_block *nb,
unsigned long action, void *data)
{
struct platform_device *pdev = to_platform_device(data);
struct lpss_private_data *pdata;
struct acpi_device *adev;
const struct acpi_device_id *id;
id = acpi_match_device(acpi_lpss_device_ids, &pdev->dev);
if (!id || !id->driver_data)
return 0;
if (acpi_bus_get_device(ACPI_HANDLE(&pdev->dev), &adev))
return 0;
pdata = acpi_driver_data(adev);
ACPI / LPSS: add all LPSS devices to the specific power domain Currently the LPSS devices are located in the different power domains depends on LPSS_SAVE_CTX flag. We would like to use the specific power domain for all LPSS devices. The LPSS DMA controller has no knobs to control its power state. The specific power domain implementation will handle this case. The patch does a preparation for that. Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Tested-by: Scott Ashcroft <scott.ashcroft@talk21.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-11-05 16:34:45 +00:00
if (!pdata)
ACPI / LPSS: Add support for exposing LTR registers to user space Devices on the Intel Lynxpoint Low Power Subsystem (LPSS) have registers providing access to LTR (Latency Tolerance Reporting) functionality that allows software to monitor and possibly influence the aggressiveness of the platform's active-state power management. For each LPSS device, there are two modes of operation related to LTR, the auto mode and the software mode. In the auto mode the LTR is set up by the platform firmware and managed by hardware. Software can only read the LTR register values to monitor the platform's behavior. In the software mode it is possible to use LTR to control the extent to which the platform will use its built-in power management features. This changeset adds support for reading the LPSS devices' LTR registers and exposing their values to user space for monitoring and diagnostics purposes. It re-uses the MMIO mappings created to access the LPSS devices' clock registers for reading the values of the LTR registers and exposes them to user space through sysfs device attributes. Namely, a new atrribute group, lpss_ltr, is created for each LPSS device. It contains three new attributes: ltr_mode, auto_ltr, sw_ltr. The value of the ltr_mode attribute reflects the LTR mode being used at the moment (software vs auto) and the other two contain the actual register values (raw) whose meaning depends on the LTR mode. All of these attributes are read-only. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2013-03-06 22:46:28 +00:00
return 0;
ACPI / LPSS: add all LPSS devices to the specific power domain Currently the LPSS devices are located in the different power domains depends on LPSS_SAVE_CTX flag. We would like to use the specific power domain for all LPSS devices. The LPSS DMA controller has no knobs to control its power state. The specific power domain implementation will handle this case. The patch does a preparation for that. Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Tested-by: Scott Ashcroft <scott.ashcroft@talk21.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-11-05 16:34:45 +00:00
if (pdata->mmio_base &&
pdata->mmio_size < pdata->dev_desc->prv_offset + LPSS_LTR_SIZE) {
ACPI / LPSS: Add support for exposing LTR registers to user space Devices on the Intel Lynxpoint Low Power Subsystem (LPSS) have registers providing access to LTR (Latency Tolerance Reporting) functionality that allows software to monitor and possibly influence the aggressiveness of the platform's active-state power management. For each LPSS device, there are two modes of operation related to LTR, the auto mode and the software mode. In the auto mode the LTR is set up by the platform firmware and managed by hardware. Software can only read the LTR register values to monitor the platform's behavior. In the software mode it is possible to use LTR to control the extent to which the platform will use its built-in power management features. This changeset adds support for reading the LPSS devices' LTR registers and exposing their values to user space for monitoring and diagnostics purposes. It re-uses the MMIO mappings created to access the LPSS devices' clock registers for reading the values of the LTR registers and exposes them to user space through sysfs device attributes. Namely, a new atrribute group, lpss_ltr, is created for each LPSS device. It contains three new attributes: ltr_mode, auto_ltr, sw_ltr. The value of the ltr_mode attribute reflects the LTR mode being used at the moment (software vs auto) and the other two contain the actual register values (raw) whose meaning depends on the LTR mode. All of these attributes are read-only. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2013-03-06 22:46:28 +00:00
dev_err(&pdev->dev, "MMIO size insufficient to access LTR\n");
return 0;
}
ACPI / LPSS: custom power domain for LPSS A power domain where we save the context of the additional LPSS registers. We need to do this or all LPSS devices are left in reset state when resuming from D3 on some Baytrails. The devices with the fractional clock divider also have zeros for N and M values after resuming unless they are reset. Li Aubrey found the root cause for the issue. The idea of using power domain for LPSS came from Mika Westerberg. Reported-by: Jin Yao <yao.jin@linux.intel.com> Suggested-by: Li Aubrey <aubrey.li@linux.intel.com> Suggested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Tested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> [rjw: Added the .complete() callback to the PM domain, fixed build warning on 32-bit.] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-05-23 13:15:09 +00:00
switch (action) {
ACPI / LPSS: allow to use specific PM domain during ->probe() This is an amendment to previously pushed commit 01ac170ba29a (ACPI / LPSS: allow to use specific PM domain during ->probe()). We can't assign anything to the platform device on ADD_DEVICE stage since it might be changed during unbound / bind cycle. Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-12-04 21:49:19 +00:00
case BUS_NOTIFY_BIND_DRIVER:
PM / Domains: add setter for dev.pm_domain Adds a function that sets the pointer to dev_pm_domain in struct device and that warns if the device has already finished probing. The reason why we want to enforce that is because in the general case that can cause problems and also that we can simplify code quite a bit if we can always assume that. This patch also changes all current code that directly sets the dev.pm_domain pointer. Signed-off-by: Tomeu Vizoso <tomeu.vizoso@collabora.com> Reviewed-by: Ulf Hansson <ulf.hansson@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2016-01-07 15:46:13 +00:00
dev_pm_domain_set(&pdev->dev, &acpi_lpss_pm_domain);
Revert "ACPI / LPSS: allow to use specific PM domain during ->probe()" The specific power domain can't be used in a way provided by the commit 01ac170ba29a, i.e. pointer to platform device is a subject to change during unbound / bind cycle. This reverts commit 01ac170ba29a9903ee590e1ef2d8e6b27b49a16c. Fixes: 3df2da968744 (Revert "ACPI / LPSS: introduce a 'proxy' device to power on LPSS for DMA") Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-12-04 21:49:18 +00:00
break;
ACPI / LPSS: allow to use specific PM domain during ->probe() This is an amendment to previously pushed commit 01ac170ba29a (ACPI / LPSS: allow to use specific PM domain during ->probe()). We can't assign anything to the platform device on ADD_DEVICE stage since it might be changed during unbound / bind cycle. Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-12-04 21:49:19 +00:00
case BUS_NOTIFY_DRIVER_NOT_BOUND:
Revert "ACPI / LPSS: allow to use specific PM domain during ->probe()" The specific power domain can't be used in a way provided by the commit 01ac170ba29a, i.e. pointer to platform device is a subject to change during unbound / bind cycle. This reverts commit 01ac170ba29a9903ee590e1ef2d8e6b27b49a16c. Fixes: 3df2da968744 (Revert "ACPI / LPSS: introduce a 'proxy' device to power on LPSS for DMA") Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-12-04 21:49:18 +00:00
case BUS_NOTIFY_UNBOUND_DRIVER:
ACPI / LPSS: set PM domain via helper setter The commit 989561de9b51 ("PM / Domains: add setter for dev.pm_domain") changed acpi_lpss.c module to use PM domain setter, though it missed one assignment. Add it here. Fixes: 989561de9b51 (PM / Domains: add setter for dev.pm_domain) Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2016-02-01 14:17:38 +00:00
dev_pm_domain_set(&pdev->dev, NULL);
Revert "ACPI / LPSS: allow to use specific PM domain during ->probe()" The specific power domain can't be used in a way provided by the commit 01ac170ba29a, i.e. pointer to platform device is a subject to change during unbound / bind cycle. This reverts commit 01ac170ba29a9903ee590e1ef2d8e6b27b49a16c. Fixes: 3df2da968744 (Revert "ACPI / LPSS: introduce a 'proxy' device to power on LPSS for DMA") Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-12-04 21:49:18 +00:00
break;
case BUS_NOTIFY_ADD_DEVICE:
PM / Domains: add setter for dev.pm_domain Adds a function that sets the pointer to dev_pm_domain in struct device and that warns if the device has already finished probing. The reason why we want to enforce that is because in the general case that can cause problems and also that we can simplify code quite a bit if we can always assume that. This patch also changes all current code that directly sets the dev.pm_domain pointer. Signed-off-by: Tomeu Vizoso <tomeu.vizoso@collabora.com> Reviewed-by: Ulf Hansson <ulf.hansson@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2016-01-07 15:46:13 +00:00
dev_pm_domain_set(&pdev->dev, &acpi_lpss_pm_domain);
ACPI / LPSS: introduce flags Replace the booleans with a single flags member variable. Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Acked-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-09-02 07:55:07 +00:00
if (pdata->dev_desc->flags & LPSS_LTR)
ACPI / LPSS: custom power domain for LPSS A power domain where we save the context of the additional LPSS registers. We need to do this or all LPSS devices are left in reset state when resuming from D3 on some Baytrails. The devices with the fractional clock divider also have zeros for N and M values after resuming unless they are reset. Li Aubrey found the root cause for the issue. The idea of using power domain for LPSS came from Mika Westerberg. Reported-by: Jin Yao <yao.jin@linux.intel.com> Suggested-by: Li Aubrey <aubrey.li@linux.intel.com> Suggested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Tested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> [rjw: Added the .complete() callback to the PM domain, fixed build warning on 32-bit.] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-05-23 13:15:09 +00:00
return sysfs_create_group(&pdev->dev.kobj,
&lpss_attr_group);
ACPI / LPSS: allow to use specific PM domain during ->probe() The LPSS DMA controller would like to use the specific PM domain callbacks during early stage, namely in ->probe(). This patch moves the specific PM domain assignment early to be accessible during a whole life time of the device in the system. Suggested-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Tested-by: Scott Ashcroft <scott.ashcroft@talk21.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-11-05 16:34:46 +00:00
break;
ACPI / LPSS: custom power domain for LPSS A power domain where we save the context of the additional LPSS registers. We need to do this or all LPSS devices are left in reset state when resuming from D3 on some Baytrails. The devices with the fractional clock divider also have zeros for N and M values after resuming unless they are reset. Li Aubrey found the root cause for the issue. The idea of using power domain for LPSS came from Mika Westerberg. Reported-by: Jin Yao <yao.jin@linux.intel.com> Suggested-by: Li Aubrey <aubrey.li@linux.intel.com> Suggested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Tested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> [rjw: Added the .complete() callback to the PM domain, fixed build warning on 32-bit.] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-05-23 13:15:09 +00:00
case BUS_NOTIFY_DEL_DEVICE:
ACPI / LPSS: introduce flags Replace the booleans with a single flags member variable. Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Acked-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-09-02 07:55:07 +00:00
if (pdata->dev_desc->flags & LPSS_LTR)
ACPI / LPSS: custom power domain for LPSS A power domain where we save the context of the additional LPSS registers. We need to do this or all LPSS devices are left in reset state when resuming from D3 on some Baytrails. The devices with the fractional clock divider also have zeros for N and M values after resuming unless they are reset. Li Aubrey found the root cause for the issue. The idea of using power domain for LPSS came from Mika Westerberg. Reported-by: Jin Yao <yao.jin@linux.intel.com> Suggested-by: Li Aubrey <aubrey.li@linux.intel.com> Suggested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Tested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> [rjw: Added the .complete() callback to the PM domain, fixed build warning on 32-bit.] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-05-23 13:15:09 +00:00
sysfs_remove_group(&pdev->dev.kobj, &lpss_attr_group);
PM / Domains: add setter for dev.pm_domain Adds a function that sets the pointer to dev_pm_domain in struct device and that warns if the device has already finished probing. The reason why we want to enforce that is because in the general case that can cause problems and also that we can simplify code quite a bit if we can always assume that. This patch also changes all current code that directly sets the dev.pm_domain pointer. Signed-off-by: Tomeu Vizoso <tomeu.vizoso@collabora.com> Reviewed-by: Ulf Hansson <ulf.hansson@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2016-01-07 15:46:13 +00:00
dev_pm_domain_set(&pdev->dev, NULL);
ACPI / LPSS: allow to use specific PM domain during ->probe() The LPSS DMA controller would like to use the specific PM domain callbacks during early stage, namely in ->probe(). This patch moves the specific PM domain assignment early to be accessible during a whole life time of the device in the system. Suggested-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Tested-by: Scott Ashcroft <scott.ashcroft@talk21.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-11-05 16:34:46 +00:00
break;
ACPI / LPSS: custom power domain for LPSS A power domain where we save the context of the additional LPSS registers. We need to do this or all LPSS devices are left in reset state when resuming from D3 on some Baytrails. The devices with the fractional clock divider also have zeros for N and M values after resuming unless they are reset. Li Aubrey found the root cause for the issue. The idea of using power domain for LPSS came from Mika Westerberg. Reported-by: Jin Yao <yao.jin@linux.intel.com> Suggested-by: Li Aubrey <aubrey.li@linux.intel.com> Suggested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Tested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> [rjw: Added the .complete() callback to the PM domain, fixed build warning on 32-bit.] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-05-23 13:15:09 +00:00
default:
break;
}
ACPI / LPSS: Add support for exposing LTR registers to user space Devices on the Intel Lynxpoint Low Power Subsystem (LPSS) have registers providing access to LTR (Latency Tolerance Reporting) functionality that allows software to monitor and possibly influence the aggressiveness of the platform's active-state power management. For each LPSS device, there are two modes of operation related to LTR, the auto mode and the software mode. In the auto mode the LTR is set up by the platform firmware and managed by hardware. Software can only read the LTR register values to monitor the platform's behavior. In the software mode it is possible to use LTR to control the extent to which the platform will use its built-in power management features. This changeset adds support for reading the LPSS devices' LTR registers and exposing their values to user space for monitoring and diagnostics purposes. It re-uses the MMIO mappings created to access the LPSS devices' clock registers for reading the values of the LTR registers and exposes them to user space through sysfs device attributes. Namely, a new atrribute group, lpss_ltr, is created for each LPSS device. It contains three new attributes: ltr_mode, auto_ltr, sw_ltr. The value of the ltr_mode attribute reflects the LTR mode being used at the moment (software vs auto) and the other two contain the actual register values (raw) whose meaning depends on the LTR mode. All of these attributes are read-only. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2013-03-06 22:46:28 +00:00
ACPI / LPSS: custom power domain for LPSS A power domain where we save the context of the additional LPSS registers. We need to do this or all LPSS devices are left in reset state when resuming from D3 on some Baytrails. The devices with the fractional clock divider also have zeros for N and M values after resuming unless they are reset. Li Aubrey found the root cause for the issue. The idea of using power domain for LPSS came from Mika Westerberg. Reported-by: Jin Yao <yao.jin@linux.intel.com> Suggested-by: Li Aubrey <aubrey.li@linux.intel.com> Suggested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Tested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> [rjw: Added the .complete() callback to the PM domain, fixed build warning on 32-bit.] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-05-23 13:15:09 +00:00
return 0;
ACPI / LPSS: Add support for exposing LTR registers to user space Devices on the Intel Lynxpoint Low Power Subsystem (LPSS) have registers providing access to LTR (Latency Tolerance Reporting) functionality that allows software to monitor and possibly influence the aggressiveness of the platform's active-state power management. For each LPSS device, there are two modes of operation related to LTR, the auto mode and the software mode. In the auto mode the LTR is set up by the platform firmware and managed by hardware. Software can only read the LTR register values to monitor the platform's behavior. In the software mode it is possible to use LTR to control the extent to which the platform will use its built-in power management features. This changeset adds support for reading the LPSS devices' LTR registers and exposing their values to user space for monitoring and diagnostics purposes. It re-uses the MMIO mappings created to access the LPSS devices' clock registers for reading the values of the LTR registers and exposes them to user space through sysfs device attributes. Namely, a new atrribute group, lpss_ltr, is created for each LPSS device. It contains three new attributes: ltr_mode, auto_ltr, sw_ltr. The value of the ltr_mode attribute reflects the LTR mode being used at the moment (software vs auto) and the other two contain the actual register values (raw) whose meaning depends on the LTR mode. All of these attributes are read-only. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2013-03-06 22:46:28 +00:00
}
static struct notifier_block acpi_lpss_nb = {
.notifier_call = acpi_lpss_platform_notify,
};
ACPI / LPSS: Support for device latency tolerance PM QoS Add a new routine, acpi_lpss_set_ltr(), for setting latency tolerance values for LPSS devices having LTR (Latency Tolerance Reporting) registers. Add .bind()/.unbind() callbacks to lpss_handler to set the LPSS devices' power.set_latency_tolerance callback pointers to acpi_lpss_set_ltr() during device addition and to clear them on device removal, respectively. That will cause the device latency tolerance PM QoS to work for the devices in question as documented. This changeset includes a fix from Mika Westerberg. Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-02-10 23:35:53 +00:00
static void acpi_lpss_bind(struct device *dev)
{
struct lpss_private_data *pdata = acpi_driver_data(ACPI_COMPANION(dev));
ACPI / LPSS: introduce flags Replace the booleans with a single flags member variable. Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Acked-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-09-02 07:55:07 +00:00
if (!pdata || !pdata->mmio_base || !(pdata->dev_desc->flags & LPSS_LTR))
ACPI / LPSS: Support for device latency tolerance PM QoS Add a new routine, acpi_lpss_set_ltr(), for setting latency tolerance values for LPSS devices having LTR (Latency Tolerance Reporting) registers. Add .bind()/.unbind() callbacks to lpss_handler to set the LPSS devices' power.set_latency_tolerance callback pointers to acpi_lpss_set_ltr() during device addition and to clear them on device removal, respectively. That will cause the device latency tolerance PM QoS to work for the devices in question as documented. This changeset includes a fix from Mika Westerberg. Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-02-10 23:35:53 +00:00
return;
if (pdata->mmio_size >= pdata->dev_desc->prv_offset + LPSS_LTR_SIZE)
dev->power.set_latency_tolerance = acpi_lpss_set_ltr;
else
dev_err(dev, "MMIO size insufficient to access LTR\n");
}
static void acpi_lpss_unbind(struct device *dev)
{
dev->power.set_latency_tolerance = NULL;
}
ACPI / scan: Add special handler for Intel Lynxpoint LPSS devices Devices on the Intel Lynxpoint Low Power Subsystem (LPSS) have some common features that aren't shared with any other platform devices, including the clock and LTR (Latency Tolerance Reporting) registers. It is better to handle those features in common code than to bother device drivers with doing that (I/O functionality-wise the LPSS devices are generally compatible with other devices that don't have those special registers and may be handled by the same drivers). The clock registers of the LPSS devices are now taken care of by the special clk-x86-lpss driver, but the MMIO mappings used for accessing those registers can also be used for accessing the LTR registers on those devices (LTR support for the Lynxpoint LPSS is going to be added by a subsequent patch). Thus it is convenient to add a special ACPI scan handler for the Lynxpoint LPSS devices that will create the MMIO mappings for accessing the clock (and LTR in the future) registers and will register the LPSS devices' clocks, so the clk-x86-lpss driver will only need to take care of the main Lynxpoint LPSS clock. Introduce a special ACPI scan handler for Intel Lynxpoint LPSS devices as described above. This also reduces overhead related to browsing the ACPI namespace in search of the LPSS devices before the registration of their clocks, removes some LPSS-specific (and somewhat ugly) code from acpi_platform.c and shrinks the overall code size slightly. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Acked-by: Mike Turquette <mturquette@linaro.org>
2013-03-06 22:46:20 +00:00
static struct acpi_scan_handler lpss_handler = {
.ids = acpi_lpss_device_ids,
.attach = acpi_lpss_create_device,
ACPI / LPSS: Support for device latency tolerance PM QoS Add a new routine, acpi_lpss_set_ltr(), for setting latency tolerance values for LPSS devices having LTR (Latency Tolerance Reporting) registers. Add .bind()/.unbind() callbacks to lpss_handler to set the LPSS devices' power.set_latency_tolerance callback pointers to acpi_lpss_set_ltr() during device addition and to clear them on device removal, respectively. That will cause the device latency tolerance PM QoS to work for the devices in question as documented. This changeset includes a fix from Mika Westerberg. Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-02-10 23:35:53 +00:00
.bind = acpi_lpss_bind,
.unbind = acpi_lpss_unbind,
ACPI / scan: Add special handler for Intel Lynxpoint LPSS devices Devices on the Intel Lynxpoint Low Power Subsystem (LPSS) have some common features that aren't shared with any other platform devices, including the clock and LTR (Latency Tolerance Reporting) registers. It is better to handle those features in common code than to bother device drivers with doing that (I/O functionality-wise the LPSS devices are generally compatible with other devices that don't have those special registers and may be handled by the same drivers). The clock registers of the LPSS devices are now taken care of by the special clk-x86-lpss driver, but the MMIO mappings used for accessing those registers can also be used for accessing the LTR registers on those devices (LTR support for the Lynxpoint LPSS is going to be added by a subsequent patch). Thus it is convenient to add a special ACPI scan handler for the Lynxpoint LPSS devices that will create the MMIO mappings for accessing the clock (and LTR in the future) registers and will register the LPSS devices' clocks, so the clk-x86-lpss driver will only need to take care of the main Lynxpoint LPSS clock. Introduce a special ACPI scan handler for Intel Lynxpoint LPSS devices as described above. This also reduces overhead related to browsing the ACPI namespace in search of the LPSS devices before the registration of their clocks, removes some LPSS-specific (and somewhat ugly) code from acpi_platform.c and shrinks the overall code size slightly. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Acked-by: Mike Turquette <mturquette@linaro.org>
2013-03-06 22:46:20 +00:00
};
void __init acpi_lpss_init(void)
{
ACPI / LPSS: override power state for LPSS DMA device This is a third approach to workaround long standing issue with LPSS on BayTrail. First one [1] was reverted since it didn't resolve the issue comprehensively. Second one [2] was rejected by internal review. The LPSS DMA controller does not have neither _PS0 nor _PS3 method. Moreover it can be powered off automatically whenever the last LPSS device goes down. In case of no power any access to the DMA controller will hang the system. The behaviour is reproduced on some HP laptops based on Intel BayTrail [3,4] as well as on ASuS T100TA transformer. Power on the LPSS island through the registers accessible in a specific way. [1] http://www.spinics.net/lists/linux-acpi/msg53963.html [2] https://bugzilla.redhat.com/attachment.cgi?id=1066779&action=diff [3] https://bugzilla.redhat.com/show_bug.cgi?id=1184273 [4] http://www.spinics.net/lists/dmaengine/msg01514.html Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-12-12 01:45:06 +00:00
const struct x86_cpu_id *id;
int ret;
ret = lpt_clk_init();
if (ret)
return;
id = x86_match_cpu(lpss_cpu_ids);
if (id)
lpss_quirks |= LPSS_QUIRK_ALWAYS_POWER_ON;
bus_register_notifier(&platform_bus_type, &acpi_lpss_nb);
acpi_scan_add_handler(&lpss_handler);
ACPI / scan: Add special handler for Intel Lynxpoint LPSS devices Devices on the Intel Lynxpoint Low Power Subsystem (LPSS) have some common features that aren't shared with any other platform devices, including the clock and LTR (Latency Tolerance Reporting) registers. It is better to handle those features in common code than to bother device drivers with doing that (I/O functionality-wise the LPSS devices are generally compatible with other devices that don't have those special registers and may be handled by the same drivers). The clock registers of the LPSS devices are now taken care of by the special clk-x86-lpss driver, but the MMIO mappings used for accessing those registers can also be used for accessing the LTR registers on those devices (LTR support for the Lynxpoint LPSS is going to be added by a subsequent patch). Thus it is convenient to add a special ACPI scan handler for the Lynxpoint LPSS devices that will create the MMIO mappings for accessing the clock (and LTR in the future) registers and will register the LPSS devices' clocks, so the clk-x86-lpss driver will only need to take care of the main Lynxpoint LPSS clock. Introduce a special ACPI scan handler for Intel Lynxpoint LPSS devices as described above. This also reduces overhead related to browsing the ACPI namespace in search of the LPSS devices before the registration of their clocks, removes some LPSS-specific (and somewhat ugly) code from acpi_platform.c and shrinks the overall code size slightly. Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Acked-by: Mike Turquette <mturquette@linaro.org>
2013-03-06 22:46:20 +00:00
}
ACPI / scan: always register ACPI LPSS scan handler Prevent platform devices from being created for ACPI LPSS devices if CONFIG_X86_INTEL_LPSS is unset by compiling out the LPSS scan handler's callbacks only in that case and still compiling its device ID list in and registering the scan handler in either case. This change is based on a prototype from Zhang Rui. Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Reviewed-by: Mika Westerberg <mika.westerberg@linux.intel.com>
2014-05-30 12:34:05 +00:00
#else
static struct acpi_scan_handler lpss_handler = {
.ids = acpi_lpss_device_ids,
};
void __init acpi_lpss_init(void)
{
acpi_scan_add_handler(&lpss_handler);
}
#endif /* CONFIG_X86_INTEL_LPSS */
Reference in New Issue Copy Permalink