OpenCloudOS-Kernel/drivers/pci/hotplug/pciehp_ctrl.c

434 lines
11 KiB
C
Raw Normal View History

// SPDX-License-Identifier: GPL-2.0+
/*
* PCI Express Hot Plug Controller Driver
*
* Copyright (C) 1995,2001 Compaq Computer Corporation
* Copyright (C) 2001 Greg Kroah-Hartman (greg@kroah.com)
* Copyright (C) 2001 IBM Corp.
* Copyright (C) 2003-2004 Intel Corporation
*
* All rights reserved.
*
* Send feedback to <greg@kroah.com>, <kristen.c.accardi@intel.com>
*
*/
#define dev_fmt(fmt) "pciehp: " fmt
#include <linux/kernel.h>
#include <linux/types.h>
PCI: pciehp: Resume to D0 on enable/disable pciehp's IRQ thread ensures accessibility of the port by runtime resuming its parent to D0. However when the slot is enabled/disabled, the port itself needs to be in D0 because its secondary bus is accessed in: pciehp_check_link_status(), pciehp_configure_device() (both called from board_added()) and pciehp_unconfigure_device() (called from remove_board()). Thus, acquire a runtime PM ref on enable/disablement of the slot. Yinghai Lu additionally discovered that some SkyLake servers feature a Power Controller for their PCIe hotplug ports (PCIe r3.1, sec 6.7.1.8) which requires the port to be in D0 when invoking pciehp_power_on_slot() (likewise called from board_added()). If slot power is turned on while in D3hot, link training later fails: https://lkml.kernel.org/r/20170205073454.GA253@wunner.de The spec is silent about such a requirement, but it seems prudent to assume that any hotplug port with a Power Controller may need this. The present commit holds a runtime PM ref whenever slot power is turned on and off, but it doesn't keep the port in D0 as long as slot power is on. If vendors determine that's necessary, they need to amend pciehp to acquire a runtime PM ref in pciehp_power_on_slot() and release one in pciehp_power_off_slot(). Signed-off-by: Lukas Wunner <lukas@wunner.de> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com> Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Cc: Mika Westerberg <mika.westerberg@linux.intel.com> Cc: Ashok Raj <ashok.raj@intel.com> Cc: Keith Busch <keith.busch@intel.com> Cc: Yinghai Lu <yinghai@kernel.org>
2018-07-20 06:27:56 +08:00
#include <linux/pm_runtime.h>
#include <linux/pci.h>
#include "pciehp.h"
/* The following routines constitute the bulk of the
hotplug controller logic
*/
PCI: pciehp: Differentiate between surprise and safe removal When removing PCI devices below a hotplug bridge, pciehp marks them as disconnected if the card is no longer present in the slot or it quiesces them if the card is still present (by disabling INTx interrupts, bus mastering and SERR# reporting). To detect whether the card is still present, pciehp checks the Presence Detect State bit in the Slot Status register. The problem with this approach is that even if the card is present, the link to it may be down, and it that case it would be better to mark the devices as disconnected instead of trying to quiesce them. Moreover, if the card in the slot was quickly replaced by another one, the Presence Detect State bit would be set, yet trying to quiesce the new card's devices would be wrong and the correct thing to do is to mark the previous card's devices as disconnected. Instead of looking at the Presence Detect State bit, it is better to differentiate whether the card was surprise removed versus safely removed (via sysfs or an Attention Button press). On surprise removal, the devices should be marked as disconnected, whereas on safe removal it is correct to quiesce the devices. The knowledge whether a surprise removal or a safe removal is at hand does exist further up in the call stack: A surprise removal is initiated by pciehp_handle_presence_or_link_change(), a safe removal by pciehp_handle_disable_request(). Pass that information down to pciehp_unconfigure_device() and use it in lieu of the Presence Detect State bit. While there, add kernel-doc to pciehp_unconfigure_device() and pciehp_configure_device(). Tested-by: Alexandru Gagniuc <mr.nuke.me@gmail.com> Signed-off-by: Lukas Wunner <lukas@wunner.de> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com> Cc: Keith Busch <keith.busch@intel.com>
2018-07-31 13:50:37 +08:00
#define SAFE_REMOVAL true
#define SURPRISE_REMOVAL false
static void set_slot_off(struct controller *ctrl)
{
/*
* Turn off slot, turn on attention indicator, turn off power
* indicator
*/
if (POWER_CTRL(ctrl)) {
pciehp_power_off_slot(ctrl);
/*
* After turning power off, we must wait for at least 1 second
* before taking any action that relies on power having been
* removed from the slot/adapter.
*/
msleep(1000);
}
pciehp_set_indicators(ctrl, PCI_EXP_SLTCTL_PWR_IND_OFF,
PCI_EXP_SLTCTL_ATTN_IND_ON);
}
/**
* board_added - Called after a board has been added to the system.
* @ctrl: PCIe hotplug controller where board is added
*
* Turns power on for the board.
* Configures board.
*/
static int board_added(struct controller *ctrl)
{
int retval = 0;
struct pci_bus *parent = ctrl->pcie->port->subordinate;
if (POWER_CTRL(ctrl)) {
/* Power on slot */
retval = pciehp_power_on_slot(ctrl);
if (retval)
return retval;
}
pciehp_set_indicators(ctrl, PCI_EXP_SLTCTL_PWR_IND_BLINK,
INDICATOR_NOOP);
/* Check link training status */
retval = pciehp_check_link_status(ctrl);
if (retval) {
ctrl_err(ctrl, "Failed to check link status\n");
goto err_exit;
}
/* Check for a power fault */
if (ctrl->power_fault_detected || pciehp_query_power_fault(ctrl)) {
ctrl_err(ctrl, "Slot(%s): Power fault\n", slot_name(ctrl));
retval = -EIO;
goto err_exit;
}
retval = pciehp_configure_device(ctrl);
if (retval) {
if (retval != -EEXIST) {
ctrl_err(ctrl, "Cannot add device at %04x:%02x:00\n",
pci_domain_nr(parent), parent->number);
goto err_exit;
}
}
pciehp_set_indicators(ctrl, PCI_EXP_SLTCTL_PWR_IND_ON,
PCI_EXP_SLTCTL_ATTN_IND_OFF);
return 0;
err_exit:
set_slot_off(ctrl);
return retval;
}
/**
* remove_board - Turn off slot and Power Indicator
* @ctrl: PCIe hotplug controller where board is being removed
PCI: pciehp: Differentiate between surprise and safe removal When removing PCI devices below a hotplug bridge, pciehp marks them as disconnected if the card is no longer present in the slot or it quiesces them if the card is still present (by disabling INTx interrupts, bus mastering and SERR# reporting). To detect whether the card is still present, pciehp checks the Presence Detect State bit in the Slot Status register. The problem with this approach is that even if the card is present, the link to it may be down, and it that case it would be better to mark the devices as disconnected instead of trying to quiesce them. Moreover, if the card in the slot was quickly replaced by another one, the Presence Detect State bit would be set, yet trying to quiesce the new card's devices would be wrong and the correct thing to do is to mark the previous card's devices as disconnected. Instead of looking at the Presence Detect State bit, it is better to differentiate whether the card was surprise removed versus safely removed (via sysfs or an Attention Button press). On surprise removal, the devices should be marked as disconnected, whereas on safe removal it is correct to quiesce the devices. The knowledge whether a surprise removal or a safe removal is at hand does exist further up in the call stack: A surprise removal is initiated by pciehp_handle_presence_or_link_change(), a safe removal by pciehp_handle_disable_request(). Pass that information down to pciehp_unconfigure_device() and use it in lieu of the Presence Detect State bit. While there, add kernel-doc to pciehp_unconfigure_device() and pciehp_configure_device(). Tested-by: Alexandru Gagniuc <mr.nuke.me@gmail.com> Signed-off-by: Lukas Wunner <lukas@wunner.de> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com> Cc: Keith Busch <keith.busch@intel.com>
2018-07-31 13:50:37 +08:00
* @safe_removal: whether the board is safely removed (versus surprise removed)
*/
static void remove_board(struct controller *ctrl, bool safe_removal)
{
pciehp_unconfigure_device(ctrl, safe_removal);
if (POWER_CTRL(ctrl)) {
pciehp_power_off_slot(ctrl);
/*
* After turning power off, we must wait for at least 1 second
* before taking any action that relies on power having been
* removed from the slot/adapter.
*/
msleep(1000);
/* Ignore link or presence changes caused by power off */
atomic_and(~(PCI_EXP_SLTSTA_DLLSC | PCI_EXP_SLTSTA_PDC),
&ctrl->pending_events);
}
pciehp_set_indicators(ctrl, PCI_EXP_SLTCTL_PWR_IND_OFF,
INDICATOR_NOOP);
}
static int pciehp_enable_slot(struct controller *ctrl);
static int pciehp_disable_slot(struct controller *ctrl, bool safe_removal);
PCI: pciehp: Enable/disable exclusively from IRQ thread Besides the IRQ thread, there are several other places in the driver which enable or disable the slot: - pciehp_probe() enables the slot if it's occupied and the pciehp_force module parameter is used. - pciehp_resume() enables or disables the slot after system sleep. - pciehp_queue_pushbutton_work() enables or disables the slot after the 5 second delay following an Attention Button press. - pciehp_sysfs_enable_slot() and pciehp_sysfs_disable_slot() enable or disable the slot on sysfs write. This requires locking and complicates pciehp's state machine. A simplification can be achieved by enabling and disabling the slot exclusively from the IRQ thread. Amend the functions listed above to request slot enable/disablement from the IRQ thread by either synthesizing a Presence Detect Changed event or, in the case of a disable user request (via sysfs or an Attention Button press), submitting a newly introduced force disable request. The latter is needed because the slot shall be forced off despite being occupied. For this force disable request, avoid colliding with Slot Status register bits by using a bit number greater than 16. For synchronous execution of requests (on sysfs write), wait for the request to finish and retrieve the result. There can only ever be one sysfs write in flight due to the locking in kernfs_fop_write(), hence there is no risk of returning the result of a different sysfs request to user space. The POWERON_STATE and POWEROFF_STATE is now no longer entered by the above-listed functions, but solely by the IRQ thread when it begins a power transition. Afterwards, it moves to STATIC_STATE. The same applies to canceling the Attention Button work, it likewise becomes an IRQ thread only operation. An immediate consequence is that the POWERON_STATE and POWEROFF_STATE is never observed by the IRQ thread itself, only by functions called in a different context, such as pciehp_sysfs_enable_slot(). So remove handling of these states from pciehp_handle_button_press() and pciehp_handle_link_change() which are exclusively called from the IRQ thread. Signed-off-by: Lukas Wunner <lukas@wunner.de> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
2018-07-20 06:27:46 +08:00
void pciehp_request(struct controller *ctrl, int action)
{
atomic_or(action, &ctrl->pending_events);
if (!pciehp_poll_mode)
irq_wake_thread(ctrl->pcie->irq, ctrl);
}
void pciehp_queue_pushbutton_work(struct work_struct *work)
{
struct controller *ctrl = container_of(work, struct controller,
button_work.work);
mutex_lock(&ctrl->state_lock);
switch (ctrl->state) {
case BLINKINGOFF_STATE:
PCI: pciehp: Enable/disable exclusively from IRQ thread Besides the IRQ thread, there are several other places in the driver which enable or disable the slot: - pciehp_probe() enables the slot if it's occupied and the pciehp_force module parameter is used. - pciehp_resume() enables or disables the slot after system sleep. - pciehp_queue_pushbutton_work() enables or disables the slot after the 5 second delay following an Attention Button press. - pciehp_sysfs_enable_slot() and pciehp_sysfs_disable_slot() enable or disable the slot on sysfs write. This requires locking and complicates pciehp's state machine. A simplification can be achieved by enabling and disabling the slot exclusively from the IRQ thread. Amend the functions listed above to request slot enable/disablement from the IRQ thread by either synthesizing a Presence Detect Changed event or, in the case of a disable user request (via sysfs or an Attention Button press), submitting a newly introduced force disable request. The latter is needed because the slot shall be forced off despite being occupied. For this force disable request, avoid colliding with Slot Status register bits by using a bit number greater than 16. For synchronous execution of requests (on sysfs write), wait for the request to finish and retrieve the result. There can only ever be one sysfs write in flight due to the locking in kernfs_fop_write(), hence there is no risk of returning the result of a different sysfs request to user space. The POWERON_STATE and POWEROFF_STATE is now no longer entered by the above-listed functions, but solely by the IRQ thread when it begins a power transition. Afterwards, it moves to STATIC_STATE. The same applies to canceling the Attention Button work, it likewise becomes an IRQ thread only operation. An immediate consequence is that the POWERON_STATE and POWEROFF_STATE is never observed by the IRQ thread itself, only by functions called in a different context, such as pciehp_sysfs_enable_slot(). So remove handling of these states from pciehp_handle_button_press() and pciehp_handle_link_change() which are exclusively called from the IRQ thread. Signed-off-by: Lukas Wunner <lukas@wunner.de> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
2018-07-20 06:27:46 +08:00
pciehp_request(ctrl, DISABLE_SLOT);
break;
case BLINKINGON_STATE:
PCI: pciehp: Enable/disable exclusively from IRQ thread Besides the IRQ thread, there are several other places in the driver which enable or disable the slot: - pciehp_probe() enables the slot if it's occupied and the pciehp_force module parameter is used. - pciehp_resume() enables or disables the slot after system sleep. - pciehp_queue_pushbutton_work() enables or disables the slot after the 5 second delay following an Attention Button press. - pciehp_sysfs_enable_slot() and pciehp_sysfs_disable_slot() enable or disable the slot on sysfs write. This requires locking and complicates pciehp's state machine. A simplification can be achieved by enabling and disabling the slot exclusively from the IRQ thread. Amend the functions listed above to request slot enable/disablement from the IRQ thread by either synthesizing a Presence Detect Changed event or, in the case of a disable user request (via sysfs or an Attention Button press), submitting a newly introduced force disable request. The latter is needed because the slot shall be forced off despite being occupied. For this force disable request, avoid colliding with Slot Status register bits by using a bit number greater than 16. For synchronous execution of requests (on sysfs write), wait for the request to finish and retrieve the result. There can only ever be one sysfs write in flight due to the locking in kernfs_fop_write(), hence there is no risk of returning the result of a different sysfs request to user space. The POWERON_STATE and POWEROFF_STATE is now no longer entered by the above-listed functions, but solely by the IRQ thread when it begins a power transition. Afterwards, it moves to STATIC_STATE. The same applies to canceling the Attention Button work, it likewise becomes an IRQ thread only operation. An immediate consequence is that the POWERON_STATE and POWEROFF_STATE is never observed by the IRQ thread itself, only by functions called in a different context, such as pciehp_sysfs_enable_slot(). So remove handling of these states from pciehp_handle_button_press() and pciehp_handle_link_change() which are exclusively called from the IRQ thread. Signed-off-by: Lukas Wunner <lukas@wunner.de> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
2018-07-20 06:27:46 +08:00
pciehp_request(ctrl, PCI_EXP_SLTSTA_PDC);
break;
default:
break;
}
mutex_unlock(&ctrl->state_lock);
}
void pciehp_handle_button_press(struct controller *ctrl)
{
mutex_lock(&ctrl->state_lock);
switch (ctrl->state) {
case OFF_STATE:
case ON_STATE:
if (ctrl->state == ON_STATE) {
ctrl->state = BLINKINGOFF_STATE;
ctrl_info(ctrl, "Slot(%s): Powering off due to button press\n",
slot_name(ctrl));
} else {
ctrl->state = BLINKINGON_STATE;
ctrl_info(ctrl, "Slot(%s) Powering on due to button press\n",
slot_name(ctrl));
}
/* blink power indicator and turn off attention */
pciehp_set_indicators(ctrl, PCI_EXP_SLTCTL_PWR_IND_BLINK,
PCI_EXP_SLTCTL_ATTN_IND_OFF);
schedule_delayed_work(&ctrl->button_work, 5 * HZ);
break;
case BLINKINGOFF_STATE:
case BLINKINGON_STATE:
/*
* Cancel if we are still blinking; this means that we
* press the attention again before the 5 sec. limit
* expires to cancel hot-add or hot-remove
*/
ctrl_info(ctrl, "Slot(%s): Button cancel\n", slot_name(ctrl));
cancel_delayed_work(&ctrl->button_work);
if (ctrl->state == BLINKINGOFF_STATE) {
ctrl->state = ON_STATE;
pciehp_set_indicators(ctrl, PCI_EXP_SLTCTL_PWR_IND_ON,
PCI_EXP_SLTCTL_ATTN_IND_OFF);
} else {
ctrl->state = OFF_STATE;
pciehp_set_indicators(ctrl, PCI_EXP_SLTCTL_PWR_IND_OFF,
PCI_EXP_SLTCTL_ATTN_IND_OFF);
}
ctrl_info(ctrl, "Slot(%s): Action canceled due to button press\n",
slot_name(ctrl));
break;
default:
ctrl_err(ctrl, "Slot(%s): Ignoring invalid state %#x\n",
slot_name(ctrl), ctrl->state);
break;
}
mutex_unlock(&ctrl->state_lock);
}
void pciehp_handle_disable_request(struct controller *ctrl)
PCI: pciehp: Enable/disable exclusively from IRQ thread Besides the IRQ thread, there are several other places in the driver which enable or disable the slot: - pciehp_probe() enables the slot if it's occupied and the pciehp_force module parameter is used. - pciehp_resume() enables or disables the slot after system sleep. - pciehp_queue_pushbutton_work() enables or disables the slot after the 5 second delay following an Attention Button press. - pciehp_sysfs_enable_slot() and pciehp_sysfs_disable_slot() enable or disable the slot on sysfs write. This requires locking and complicates pciehp's state machine. A simplification can be achieved by enabling and disabling the slot exclusively from the IRQ thread. Amend the functions listed above to request slot enable/disablement from the IRQ thread by either synthesizing a Presence Detect Changed event or, in the case of a disable user request (via sysfs or an Attention Button press), submitting a newly introduced force disable request. The latter is needed because the slot shall be forced off despite being occupied. For this force disable request, avoid colliding with Slot Status register bits by using a bit number greater than 16. For synchronous execution of requests (on sysfs write), wait for the request to finish and retrieve the result. There can only ever be one sysfs write in flight due to the locking in kernfs_fop_write(), hence there is no risk of returning the result of a different sysfs request to user space. The POWERON_STATE and POWEROFF_STATE is now no longer entered by the above-listed functions, but solely by the IRQ thread when it begins a power transition. Afterwards, it moves to STATIC_STATE. The same applies to canceling the Attention Button work, it likewise becomes an IRQ thread only operation. An immediate consequence is that the POWERON_STATE and POWEROFF_STATE is never observed by the IRQ thread itself, only by functions called in a different context, such as pciehp_sysfs_enable_slot(). So remove handling of these states from pciehp_handle_button_press() and pciehp_handle_link_change() which are exclusively called from the IRQ thread. Signed-off-by: Lukas Wunner <lukas@wunner.de> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
2018-07-20 06:27:46 +08:00
{
mutex_lock(&ctrl->state_lock);
switch (ctrl->state) {
PCI: pciehp: Enable/disable exclusively from IRQ thread Besides the IRQ thread, there are several other places in the driver which enable or disable the slot: - pciehp_probe() enables the slot if it's occupied and the pciehp_force module parameter is used. - pciehp_resume() enables or disables the slot after system sleep. - pciehp_queue_pushbutton_work() enables or disables the slot after the 5 second delay following an Attention Button press. - pciehp_sysfs_enable_slot() and pciehp_sysfs_disable_slot() enable or disable the slot on sysfs write. This requires locking and complicates pciehp's state machine. A simplification can be achieved by enabling and disabling the slot exclusively from the IRQ thread. Amend the functions listed above to request slot enable/disablement from the IRQ thread by either synthesizing a Presence Detect Changed event or, in the case of a disable user request (via sysfs or an Attention Button press), submitting a newly introduced force disable request. The latter is needed because the slot shall be forced off despite being occupied. For this force disable request, avoid colliding with Slot Status register bits by using a bit number greater than 16. For synchronous execution of requests (on sysfs write), wait for the request to finish and retrieve the result. There can only ever be one sysfs write in flight due to the locking in kernfs_fop_write(), hence there is no risk of returning the result of a different sysfs request to user space. The POWERON_STATE and POWEROFF_STATE is now no longer entered by the above-listed functions, but solely by the IRQ thread when it begins a power transition. Afterwards, it moves to STATIC_STATE. The same applies to canceling the Attention Button work, it likewise becomes an IRQ thread only operation. An immediate consequence is that the POWERON_STATE and POWEROFF_STATE is never observed by the IRQ thread itself, only by functions called in a different context, such as pciehp_sysfs_enable_slot(). So remove handling of these states from pciehp_handle_button_press() and pciehp_handle_link_change() which are exclusively called from the IRQ thread. Signed-off-by: Lukas Wunner <lukas@wunner.de> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
2018-07-20 06:27:46 +08:00
case BLINKINGON_STATE:
case BLINKINGOFF_STATE:
cancel_delayed_work(&ctrl->button_work);
break;
PCI: pciehp: Enable/disable exclusively from IRQ thread Besides the IRQ thread, there are several other places in the driver which enable or disable the slot: - pciehp_probe() enables the slot if it's occupied and the pciehp_force module parameter is used. - pciehp_resume() enables or disables the slot after system sleep. - pciehp_queue_pushbutton_work() enables or disables the slot after the 5 second delay following an Attention Button press. - pciehp_sysfs_enable_slot() and pciehp_sysfs_disable_slot() enable or disable the slot on sysfs write. This requires locking and complicates pciehp's state machine. A simplification can be achieved by enabling and disabling the slot exclusively from the IRQ thread. Amend the functions listed above to request slot enable/disablement from the IRQ thread by either synthesizing a Presence Detect Changed event or, in the case of a disable user request (via sysfs or an Attention Button press), submitting a newly introduced force disable request. The latter is needed because the slot shall be forced off despite being occupied. For this force disable request, avoid colliding with Slot Status register bits by using a bit number greater than 16. For synchronous execution of requests (on sysfs write), wait for the request to finish and retrieve the result. There can only ever be one sysfs write in flight due to the locking in kernfs_fop_write(), hence there is no risk of returning the result of a different sysfs request to user space. The POWERON_STATE and POWEROFF_STATE is now no longer entered by the above-listed functions, but solely by the IRQ thread when it begins a power transition. Afterwards, it moves to STATIC_STATE. The same applies to canceling the Attention Button work, it likewise becomes an IRQ thread only operation. An immediate consequence is that the POWERON_STATE and POWEROFF_STATE is never observed by the IRQ thread itself, only by functions called in a different context, such as pciehp_sysfs_enable_slot(). So remove handling of these states from pciehp_handle_button_press() and pciehp_handle_link_change() which are exclusively called from the IRQ thread. Signed-off-by: Lukas Wunner <lukas@wunner.de> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
2018-07-20 06:27:46 +08:00
}
ctrl->state = POWEROFF_STATE;
mutex_unlock(&ctrl->state_lock);
PCI: pciehp: Enable/disable exclusively from IRQ thread Besides the IRQ thread, there are several other places in the driver which enable or disable the slot: - pciehp_probe() enables the slot if it's occupied and the pciehp_force module parameter is used. - pciehp_resume() enables or disables the slot after system sleep. - pciehp_queue_pushbutton_work() enables or disables the slot after the 5 second delay following an Attention Button press. - pciehp_sysfs_enable_slot() and pciehp_sysfs_disable_slot() enable or disable the slot on sysfs write. This requires locking and complicates pciehp's state machine. A simplification can be achieved by enabling and disabling the slot exclusively from the IRQ thread. Amend the functions listed above to request slot enable/disablement from the IRQ thread by either synthesizing a Presence Detect Changed event or, in the case of a disable user request (via sysfs or an Attention Button press), submitting a newly introduced force disable request. The latter is needed because the slot shall be forced off despite being occupied. For this force disable request, avoid colliding with Slot Status register bits by using a bit number greater than 16. For synchronous execution of requests (on sysfs write), wait for the request to finish and retrieve the result. There can only ever be one sysfs write in flight due to the locking in kernfs_fop_write(), hence there is no risk of returning the result of a different sysfs request to user space. The POWERON_STATE and POWEROFF_STATE is now no longer entered by the above-listed functions, but solely by the IRQ thread when it begins a power transition. Afterwards, it moves to STATIC_STATE. The same applies to canceling the Attention Button work, it likewise becomes an IRQ thread only operation. An immediate consequence is that the POWERON_STATE and POWEROFF_STATE is never observed by the IRQ thread itself, only by functions called in a different context, such as pciehp_sysfs_enable_slot(). So remove handling of these states from pciehp_handle_button_press() and pciehp_handle_link_change() which are exclusively called from the IRQ thread. Signed-off-by: Lukas Wunner <lukas@wunner.de> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
2018-07-20 06:27:46 +08:00
ctrl->request_result = pciehp_disable_slot(ctrl, SAFE_REMOVAL);
PCI: pciehp: Enable/disable exclusively from IRQ thread Besides the IRQ thread, there are several other places in the driver which enable or disable the slot: - pciehp_probe() enables the slot if it's occupied and the pciehp_force module parameter is used. - pciehp_resume() enables or disables the slot after system sleep. - pciehp_queue_pushbutton_work() enables or disables the slot after the 5 second delay following an Attention Button press. - pciehp_sysfs_enable_slot() and pciehp_sysfs_disable_slot() enable or disable the slot on sysfs write. This requires locking and complicates pciehp's state machine. A simplification can be achieved by enabling and disabling the slot exclusively from the IRQ thread. Amend the functions listed above to request slot enable/disablement from the IRQ thread by either synthesizing a Presence Detect Changed event or, in the case of a disable user request (via sysfs or an Attention Button press), submitting a newly introduced force disable request. The latter is needed because the slot shall be forced off despite being occupied. For this force disable request, avoid colliding with Slot Status register bits by using a bit number greater than 16. For synchronous execution of requests (on sysfs write), wait for the request to finish and retrieve the result. There can only ever be one sysfs write in flight due to the locking in kernfs_fop_write(), hence there is no risk of returning the result of a different sysfs request to user space. The POWERON_STATE and POWEROFF_STATE is now no longer entered by the above-listed functions, but solely by the IRQ thread when it begins a power transition. Afterwards, it moves to STATIC_STATE. The same applies to canceling the Attention Button work, it likewise becomes an IRQ thread only operation. An immediate consequence is that the POWERON_STATE and POWEROFF_STATE is never observed by the IRQ thread itself, only by functions called in a different context, such as pciehp_sysfs_enable_slot(). So remove handling of these states from pciehp_handle_button_press() and pciehp_handle_link_change() which are exclusively called from the IRQ thread. Signed-off-by: Lukas Wunner <lukas@wunner.de> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
2018-07-20 06:27:46 +08:00
}
void pciehp_handle_presence_or_link_change(struct controller *ctrl, u32 events)
{
PCI: pciehp: Prevent deadlock on disconnect This addresses deadlocks in these common cases in hierarchies containing two switches: - All involved ports are runtime suspended and they are unplugged. This can happen easily if the drivers involved automatically enable runtime PM (xHCI for example does that). - System is suspended (e.g., closing the lid on a laptop) with a dock + something else connected, and the dock is unplugged while suspended. These cases lead to the following deadlock: INFO: task irq/126-pciehp:198 blocked for more than 120 seconds. irq/126-pciehp D 0 198 2 0x80000000 Call Trace: schedule+0x2c/0x80 schedule_timeout+0x246/0x350 wait_for_completion+0xb7/0x140 kthread_stop+0x49/0x110 free_irq+0x32/0x70 pcie_shutdown_notification+0x2f/0x50 pciehp_remove+0x27/0x50 pcie_port_remove_service+0x36/0x50 device_release_driver+0x12/0x20 bus_remove_device+0xec/0x160 device_del+0x13b/0x350 device_unregister+0x1a/0x60 remove_iter+0x1e/0x30 device_for_each_child+0x56/0x90 pcie_port_device_remove+0x22/0x40 pcie_portdrv_remove+0x20/0x60 pci_device_remove+0x3e/0xc0 device_release_driver_internal+0x18c/0x250 device_release_driver+0x12/0x20 pci_stop_bus_device+0x6f/0x90 pci_stop_bus_device+0x31/0x90 pci_stop_and_remove_bus_device+0x12/0x20 pciehp_unconfigure_device+0x88/0x140 pciehp_disable_slot+0x6a/0x110 pciehp_handle_presence_or_link_change+0x263/0x400 pciehp_ist+0x1c9/0x1d0 irq_thread_fn+0x24/0x60 irq_thread+0xeb/0x190 kthread+0x120/0x140 INFO: task irq/190-pciehp:2288 blocked for more than 120 seconds. irq/190-pciehp D 0 2288 2 0x80000000 Call Trace: __schedule+0x2a2/0x880 schedule+0x2c/0x80 schedule_preempt_disabled+0xe/0x10 mutex_lock+0x2c/0x30 pci_lock_rescan_remove+0x15/0x20 pciehp_unconfigure_device+0x4d/0x140 pciehp_disable_slot+0x6a/0x110 pciehp_handle_presence_or_link_change+0x263/0x400 pciehp_ist+0x1c9/0x1d0 irq_thread_fn+0x24/0x60 irq_thread+0xeb/0x190 kthread+0x120/0x140 What happens here is that the whole hierarchy is runtime resumed and the parent PCIe downstream port, which got the hot-remove event, starts removing devices below it, taking pci_lock_rescan_remove() lock. When the child PCIe port is runtime resumed it calls pciehp_check_presence() which ends up calling pciehp_card_present() and pciehp_check_link_active(). Both of these use pcie_capability_read_word(), which notices that the underlying device is already gone and returns PCIBIOS_DEVICE_NOT_FOUND with the capability value set to 0. When pciehp gets this value it thinks that its child device is also hot-removed and schedules its IRQ thread to handle the event. The deadlock happens when the child's IRQ thread runs and tries to acquire pci_lock_rescan_remove() which is already taken by the parent and the parent waits for the child's IRQ thread to finish. Prevent this from happening by checking the return value of pcie_capability_read_word() and if it is PCIBIOS_DEVICE_NOT_FOUND stop performing any hot-removal activities. [bhelgaas: add common scenarios to commit log] Link: https://lore.kernel.org/r/20191029170022.57528-2-mika.westerberg@linux.intel.com Tested-by: Kai-Heng Feng <kai.heng.feng@canonical.com> Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
2019-10-30 01:00:22 +08:00
int present, link_active;
PCI: pciehp: Become resilient to missed events A hotplug port's Slot Status register does not count how often each type of event occurred, it only records the fact *that* an event has occurred. Previously pciehp queued a work item for each event. But if it missed an event, e.g. removal of a card in-between two back-to-back insertions, it queued up the wrong work item or no work item at all. Commit fad214b0aa72 ("PCI: pciehp: Process all hotplug events before looking for new ones") sought to improve the situation by shrinking the window during which events may be missed. But Stefan Roese reports unbalanced Card present and Link Up events, suggesting that we're still missing events if they occur very rapidly. Bjorn Helgaas responds that he considers pciehp's event handling "baroque" and calls for its simplification and rationalization: https://lkml.kernel.org/r/20180202192045.GA53759@bhelgaas-glaptop.roam.corp.google.com It gets worse once a hotplug port is runtime suspended: The port can signal an interrupt while it and its parents are in D3hot, i.e. while it is inaccessible. By the time we've runtime resumed all parents to D0 and read the port's Slot Status register, we may have missed an arbitrary number of events. Event handling therefore needs to be reworked to become resilient to missed events. Assume that a Presence Detect Changed event has occurred. Consider the following truth table: - Slot is in OFF_STATE and is currently empty. => Do nothing. (The event is trailing a Link Down or we've missed an insertion and subsequent removal.) - Slot is in OFF_STATE and is currently occupied. => Turn the slot on. - Slot is in ON_STATE and is currently empty. => Turn the slot off. - Slot is in ON_STATE and is currently occupied. => Turn the slot off, (Be cautious and assume the card in then back on. the slot isn't the same as before.) This leads to the following simple algorithm: 1 If the slot is in ON_STATE, turn it off unconditionally. 2 If the slot is currently occupied, turn it on. Because those actions are now carried out synchronously, rather than by scheduled work items, pciehp reacts to the *current* situation and missed events no longer matter. Data Link Layer State Changed events can be handled identically to Presence Detect Changed events. Note that in the above truth table, a Link Up trailing a Card present event didn't have to be accounted for: It is filtered out by pciehp_check_link_status(). As for Attention Button Pressed events, PCIe r4.0, sec 6.7.1.5 says: "Once the Power Indicator begins blinking, a 5-second abort interval exists during which a second depression of the Attention Button cancels the operation." In other words, the user can only expect the system to react to a button press after it starts blinking. Missed button presses that occur in-between are irrelevant. Signed-off-by: Lukas Wunner <lukas@wunner.de> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com> Cc: Stefan Roese <sr@denx.de> Cc: Mayurkumar Patel <mayurkumar.patel@intel.com> Cc: Mika Westerberg <mika.westerberg@linux.intel.com> Cc: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
2018-07-20 06:27:49 +08:00
/*
* If the slot is on and presence or link has changed, turn it off.
* Even if it's occupied again, we cannot assume the card is the same.
*/
mutex_lock(&ctrl->state_lock);
switch (ctrl->state) {
case BLINKINGOFF_STATE:
cancel_delayed_work(&ctrl->button_work);
/* fall through */
case ON_STATE:
ctrl->state = POWEROFF_STATE;
mutex_unlock(&ctrl->state_lock);
PCI: pciehp: Become resilient to missed events A hotplug port's Slot Status register does not count how often each type of event occurred, it only records the fact *that* an event has occurred. Previously pciehp queued a work item for each event. But if it missed an event, e.g. removal of a card in-between two back-to-back insertions, it queued up the wrong work item or no work item at all. Commit fad214b0aa72 ("PCI: pciehp: Process all hotplug events before looking for new ones") sought to improve the situation by shrinking the window during which events may be missed. But Stefan Roese reports unbalanced Card present and Link Up events, suggesting that we're still missing events if they occur very rapidly. Bjorn Helgaas responds that he considers pciehp's event handling "baroque" and calls for its simplification and rationalization: https://lkml.kernel.org/r/20180202192045.GA53759@bhelgaas-glaptop.roam.corp.google.com It gets worse once a hotplug port is runtime suspended: The port can signal an interrupt while it and its parents are in D3hot, i.e. while it is inaccessible. By the time we've runtime resumed all parents to D0 and read the port's Slot Status register, we may have missed an arbitrary number of events. Event handling therefore needs to be reworked to become resilient to missed events. Assume that a Presence Detect Changed event has occurred. Consider the following truth table: - Slot is in OFF_STATE and is currently empty. => Do nothing. (The event is trailing a Link Down or we've missed an insertion and subsequent removal.) - Slot is in OFF_STATE and is currently occupied. => Turn the slot on. - Slot is in ON_STATE and is currently empty. => Turn the slot off. - Slot is in ON_STATE and is currently occupied. => Turn the slot off, (Be cautious and assume the card in then back on. the slot isn't the same as before.) This leads to the following simple algorithm: 1 If the slot is in ON_STATE, turn it off unconditionally. 2 If the slot is currently occupied, turn it on. Because those actions are now carried out synchronously, rather than by scheduled work items, pciehp reacts to the *current* situation and missed events no longer matter. Data Link Layer State Changed events can be handled identically to Presence Detect Changed events. Note that in the above truth table, a Link Up trailing a Card present event didn't have to be accounted for: It is filtered out by pciehp_check_link_status(). As for Attention Button Pressed events, PCIe r4.0, sec 6.7.1.5 says: "Once the Power Indicator begins blinking, a 5-second abort interval exists during which a second depression of the Attention Button cancels the operation." In other words, the user can only expect the system to react to a button press after it starts blinking. Missed button presses that occur in-between are irrelevant. Signed-off-by: Lukas Wunner <lukas@wunner.de> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com> Cc: Stefan Roese <sr@denx.de> Cc: Mayurkumar Patel <mayurkumar.patel@intel.com> Cc: Mika Westerberg <mika.westerberg@linux.intel.com> Cc: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
2018-07-20 06:27:49 +08:00
if (events & PCI_EXP_SLTSTA_DLLSC)
ctrl_info(ctrl, "Slot(%s): Link Down\n",
slot_name(ctrl));
PCI: pciehp: Become resilient to missed events A hotplug port's Slot Status register does not count how often each type of event occurred, it only records the fact *that* an event has occurred. Previously pciehp queued a work item for each event. But if it missed an event, e.g. removal of a card in-between two back-to-back insertions, it queued up the wrong work item or no work item at all. Commit fad214b0aa72 ("PCI: pciehp: Process all hotplug events before looking for new ones") sought to improve the situation by shrinking the window during which events may be missed. But Stefan Roese reports unbalanced Card present and Link Up events, suggesting that we're still missing events if they occur very rapidly. Bjorn Helgaas responds that he considers pciehp's event handling "baroque" and calls for its simplification and rationalization: https://lkml.kernel.org/r/20180202192045.GA53759@bhelgaas-glaptop.roam.corp.google.com It gets worse once a hotplug port is runtime suspended: The port can signal an interrupt while it and its parents are in D3hot, i.e. while it is inaccessible. By the time we've runtime resumed all parents to D0 and read the port's Slot Status register, we may have missed an arbitrary number of events. Event handling therefore needs to be reworked to become resilient to missed events. Assume that a Presence Detect Changed event has occurred. Consider the following truth table: - Slot is in OFF_STATE and is currently empty. => Do nothing. (The event is trailing a Link Down or we've missed an insertion and subsequent removal.) - Slot is in OFF_STATE and is currently occupied. => Turn the slot on. - Slot is in ON_STATE and is currently empty. => Turn the slot off. - Slot is in ON_STATE and is currently occupied. => Turn the slot off, (Be cautious and assume the card in then back on. the slot isn't the same as before.) This leads to the following simple algorithm: 1 If the slot is in ON_STATE, turn it off unconditionally. 2 If the slot is currently occupied, turn it on. Because those actions are now carried out synchronously, rather than by scheduled work items, pciehp reacts to the *current* situation and missed events no longer matter. Data Link Layer State Changed events can be handled identically to Presence Detect Changed events. Note that in the above truth table, a Link Up trailing a Card present event didn't have to be accounted for: It is filtered out by pciehp_check_link_status(). As for Attention Button Pressed events, PCIe r4.0, sec 6.7.1.5 says: "Once the Power Indicator begins blinking, a 5-second abort interval exists during which a second depression of the Attention Button cancels the operation." In other words, the user can only expect the system to react to a button press after it starts blinking. Missed button presses that occur in-between are irrelevant. Signed-off-by: Lukas Wunner <lukas@wunner.de> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com> Cc: Stefan Roese <sr@denx.de> Cc: Mayurkumar Patel <mayurkumar.patel@intel.com> Cc: Mika Westerberg <mika.westerberg@linux.intel.com> Cc: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
2018-07-20 06:27:49 +08:00
if (events & PCI_EXP_SLTSTA_PDC)
ctrl_info(ctrl, "Slot(%s): Card not present\n",
slot_name(ctrl));
pciehp_disable_slot(ctrl, SURPRISE_REMOVAL);
break;
default:
mutex_unlock(&ctrl->state_lock);
break;
}
PCI: pciehp: Become resilient to missed events A hotplug port's Slot Status register does not count how often each type of event occurred, it only records the fact *that* an event has occurred. Previously pciehp queued a work item for each event. But if it missed an event, e.g. removal of a card in-between two back-to-back insertions, it queued up the wrong work item or no work item at all. Commit fad214b0aa72 ("PCI: pciehp: Process all hotplug events before looking for new ones") sought to improve the situation by shrinking the window during which events may be missed. But Stefan Roese reports unbalanced Card present and Link Up events, suggesting that we're still missing events if they occur very rapidly. Bjorn Helgaas responds that he considers pciehp's event handling "baroque" and calls for its simplification and rationalization: https://lkml.kernel.org/r/20180202192045.GA53759@bhelgaas-glaptop.roam.corp.google.com It gets worse once a hotplug port is runtime suspended: The port can signal an interrupt while it and its parents are in D3hot, i.e. while it is inaccessible. By the time we've runtime resumed all parents to D0 and read the port's Slot Status register, we may have missed an arbitrary number of events. Event handling therefore needs to be reworked to become resilient to missed events. Assume that a Presence Detect Changed event has occurred. Consider the following truth table: - Slot is in OFF_STATE and is currently empty. => Do nothing. (The event is trailing a Link Down or we've missed an insertion and subsequent removal.) - Slot is in OFF_STATE and is currently occupied. => Turn the slot on. - Slot is in ON_STATE and is currently empty. => Turn the slot off. - Slot is in ON_STATE and is currently occupied. => Turn the slot off, (Be cautious and assume the card in then back on. the slot isn't the same as before.) This leads to the following simple algorithm: 1 If the slot is in ON_STATE, turn it off unconditionally. 2 If the slot is currently occupied, turn it on. Because those actions are now carried out synchronously, rather than by scheduled work items, pciehp reacts to the *current* situation and missed events no longer matter. Data Link Layer State Changed events can be handled identically to Presence Detect Changed events. Note that in the above truth table, a Link Up trailing a Card present event didn't have to be accounted for: It is filtered out by pciehp_check_link_status(). As for Attention Button Pressed events, PCIe r4.0, sec 6.7.1.5 says: "Once the Power Indicator begins blinking, a 5-second abort interval exists during which a second depression of the Attention Button cancels the operation." In other words, the user can only expect the system to react to a button press after it starts blinking. Missed button presses that occur in-between are irrelevant. Signed-off-by: Lukas Wunner <lukas@wunner.de> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com> Cc: Stefan Roese <sr@denx.de> Cc: Mayurkumar Patel <mayurkumar.patel@intel.com> Cc: Mika Westerberg <mika.westerberg@linux.intel.com> Cc: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
2018-07-20 06:27:49 +08:00
/* Turn the slot on if it's occupied or link is up */
mutex_lock(&ctrl->state_lock);
PCI: pciehp: Tolerate Presence Detect hardwired to zero The WiGig Bus Extension (WBE) specification allows tunneling PCIe over IEEE 802.11. A product implementing this spec is the wil6210 from Wilocity (now part of Qualcomm Atheros). It integrates a PCIe switch with a wireless network adapter: 00.0-+ [1ae9:0101] Upstream Port +-00.0-+ [1ae9:0200] Downstream Port | +-00.0 [168c:0034] Atheros AR9462 Wireless Network Adapter +-02.0 [1ae9:0201] Downstream Port +-03.0 [1ae9:0201] Downstream Port Wirelessly attached devices presumably appear below the hotplug ports with device ID [1ae9:0201]. Oddly, the Downstream Port [1ae9:0200] leading to the wireless network adapter is likewise Hotplug Capable, but has its Presence Detect State bit hardwired to zero. Even if the Link Active bit is set, Presence Detect is zero, so this cannot be caused by in-band presence detection but only by broken hardware. pciehp assumes an empty slot if Presence Detect State is zero, regardless of Link Active being one. Consequently, up until v4.18 it removes the wireless network adapter in pciehp_resume(). From v4.19 it already does so in pciehp_probe(). Be lenient towards broken hardware and assume the slot is occupied if Link Active is set: Introduce pciehp_card_present_or_link_active() and use it in lieu of pciehp_get_adapter_status() everywhere, except in pciehp_handle_presence_or_link_change() whose log messages depend on which of Presence Detect State or Link Active is set. Remove the Presence Detect State check from __pciehp_enable_slot() because it is only called if either of Presence Detect State or Link Active is set. Caution: There is a possibility that broken hardware exists which has working Presence Detect but hardwires Link Active to one. On such hardware the slot will now incorrectly be considered always occupied. If such hardware is discovered, this commit can be rolled back and a quirk can be added which sets is_hotplug_bridge = 0 for [1ae9:0200]. Link: https://bugzilla.kernel.org/show_bug.cgi?id=200839 Reported-and-tested-by: David Yang <mmyangfl@gmail.com> Signed-off-by: Lukas Wunner <lukas@wunner.de> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com> Cc: Rajat Jain <rajatja@google.com> Cc: Ashok Raj <ashok.raj@intel.com>
2018-09-08 15:59:01 +08:00
present = pciehp_card_present(ctrl);
PCI: pciehp: Become resilient to missed events A hotplug port's Slot Status register does not count how often each type of event occurred, it only records the fact *that* an event has occurred. Previously pciehp queued a work item for each event. But if it missed an event, e.g. removal of a card in-between two back-to-back insertions, it queued up the wrong work item or no work item at all. Commit fad214b0aa72 ("PCI: pciehp: Process all hotplug events before looking for new ones") sought to improve the situation by shrinking the window during which events may be missed. But Stefan Roese reports unbalanced Card present and Link Up events, suggesting that we're still missing events if they occur very rapidly. Bjorn Helgaas responds that he considers pciehp's event handling "baroque" and calls for its simplification and rationalization: https://lkml.kernel.org/r/20180202192045.GA53759@bhelgaas-glaptop.roam.corp.google.com It gets worse once a hotplug port is runtime suspended: The port can signal an interrupt while it and its parents are in D3hot, i.e. while it is inaccessible. By the time we've runtime resumed all parents to D0 and read the port's Slot Status register, we may have missed an arbitrary number of events. Event handling therefore needs to be reworked to become resilient to missed events. Assume that a Presence Detect Changed event has occurred. Consider the following truth table: - Slot is in OFF_STATE and is currently empty. => Do nothing. (The event is trailing a Link Down or we've missed an insertion and subsequent removal.) - Slot is in OFF_STATE and is currently occupied. => Turn the slot on. - Slot is in ON_STATE and is currently empty. => Turn the slot off. - Slot is in ON_STATE and is currently occupied. => Turn the slot off, (Be cautious and assume the card in then back on. the slot isn't the same as before.) This leads to the following simple algorithm: 1 If the slot is in ON_STATE, turn it off unconditionally. 2 If the slot is currently occupied, turn it on. Because those actions are now carried out synchronously, rather than by scheduled work items, pciehp reacts to the *current* situation and missed events no longer matter. Data Link Layer State Changed events can be handled identically to Presence Detect Changed events. Note that in the above truth table, a Link Up trailing a Card present event didn't have to be accounted for: It is filtered out by pciehp_check_link_status(). As for Attention Button Pressed events, PCIe r4.0, sec 6.7.1.5 says: "Once the Power Indicator begins blinking, a 5-second abort interval exists during which a second depression of the Attention Button cancels the operation." In other words, the user can only expect the system to react to a button press after it starts blinking. Missed button presses that occur in-between are irrelevant. Signed-off-by: Lukas Wunner <lukas@wunner.de> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com> Cc: Stefan Roese <sr@denx.de> Cc: Mayurkumar Patel <mayurkumar.patel@intel.com> Cc: Mika Westerberg <mika.westerberg@linux.intel.com> Cc: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
2018-07-20 06:27:49 +08:00
link_active = pciehp_check_link_active(ctrl);
PCI: pciehp: Prevent deadlock on disconnect This addresses deadlocks in these common cases in hierarchies containing two switches: - All involved ports are runtime suspended and they are unplugged. This can happen easily if the drivers involved automatically enable runtime PM (xHCI for example does that). - System is suspended (e.g., closing the lid on a laptop) with a dock + something else connected, and the dock is unplugged while suspended. These cases lead to the following deadlock: INFO: task irq/126-pciehp:198 blocked for more than 120 seconds. irq/126-pciehp D 0 198 2 0x80000000 Call Trace: schedule+0x2c/0x80 schedule_timeout+0x246/0x350 wait_for_completion+0xb7/0x140 kthread_stop+0x49/0x110 free_irq+0x32/0x70 pcie_shutdown_notification+0x2f/0x50 pciehp_remove+0x27/0x50 pcie_port_remove_service+0x36/0x50 device_release_driver+0x12/0x20 bus_remove_device+0xec/0x160 device_del+0x13b/0x350 device_unregister+0x1a/0x60 remove_iter+0x1e/0x30 device_for_each_child+0x56/0x90 pcie_port_device_remove+0x22/0x40 pcie_portdrv_remove+0x20/0x60 pci_device_remove+0x3e/0xc0 device_release_driver_internal+0x18c/0x250 device_release_driver+0x12/0x20 pci_stop_bus_device+0x6f/0x90 pci_stop_bus_device+0x31/0x90 pci_stop_and_remove_bus_device+0x12/0x20 pciehp_unconfigure_device+0x88/0x140 pciehp_disable_slot+0x6a/0x110 pciehp_handle_presence_or_link_change+0x263/0x400 pciehp_ist+0x1c9/0x1d0 irq_thread_fn+0x24/0x60 irq_thread+0xeb/0x190 kthread+0x120/0x140 INFO: task irq/190-pciehp:2288 blocked for more than 120 seconds. irq/190-pciehp D 0 2288 2 0x80000000 Call Trace: __schedule+0x2a2/0x880 schedule+0x2c/0x80 schedule_preempt_disabled+0xe/0x10 mutex_lock+0x2c/0x30 pci_lock_rescan_remove+0x15/0x20 pciehp_unconfigure_device+0x4d/0x140 pciehp_disable_slot+0x6a/0x110 pciehp_handle_presence_or_link_change+0x263/0x400 pciehp_ist+0x1c9/0x1d0 irq_thread_fn+0x24/0x60 irq_thread+0xeb/0x190 kthread+0x120/0x140 What happens here is that the whole hierarchy is runtime resumed and the parent PCIe downstream port, which got the hot-remove event, starts removing devices below it, taking pci_lock_rescan_remove() lock. When the child PCIe port is runtime resumed it calls pciehp_check_presence() which ends up calling pciehp_card_present() and pciehp_check_link_active(). Both of these use pcie_capability_read_word(), which notices that the underlying device is already gone and returns PCIBIOS_DEVICE_NOT_FOUND with the capability value set to 0. When pciehp gets this value it thinks that its child device is also hot-removed and schedules its IRQ thread to handle the event. The deadlock happens when the child's IRQ thread runs and tries to acquire pci_lock_rescan_remove() which is already taken by the parent and the parent waits for the child's IRQ thread to finish. Prevent this from happening by checking the return value of pcie_capability_read_word() and if it is PCIBIOS_DEVICE_NOT_FOUND stop performing any hot-removal activities. [bhelgaas: add common scenarios to commit log] Link: https://lore.kernel.org/r/20191029170022.57528-2-mika.westerberg@linux.intel.com Tested-by: Kai-Heng Feng <kai.heng.feng@canonical.com> Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
2019-10-30 01:00:22 +08:00
if (present <= 0 && link_active <= 0) {
mutex_unlock(&ctrl->state_lock);
PCI: pciehp: Become resilient to missed events A hotplug port's Slot Status register does not count how often each type of event occurred, it only records the fact *that* an event has occurred. Previously pciehp queued a work item for each event. But if it missed an event, e.g. removal of a card in-between two back-to-back insertions, it queued up the wrong work item or no work item at all. Commit fad214b0aa72 ("PCI: pciehp: Process all hotplug events before looking for new ones") sought to improve the situation by shrinking the window during which events may be missed. But Stefan Roese reports unbalanced Card present and Link Up events, suggesting that we're still missing events if they occur very rapidly. Bjorn Helgaas responds that he considers pciehp's event handling "baroque" and calls for its simplification and rationalization: https://lkml.kernel.org/r/20180202192045.GA53759@bhelgaas-glaptop.roam.corp.google.com It gets worse once a hotplug port is runtime suspended: The port can signal an interrupt while it and its parents are in D3hot, i.e. while it is inaccessible. By the time we've runtime resumed all parents to D0 and read the port's Slot Status register, we may have missed an arbitrary number of events. Event handling therefore needs to be reworked to become resilient to missed events. Assume that a Presence Detect Changed event has occurred. Consider the following truth table: - Slot is in OFF_STATE and is currently empty. => Do nothing. (The event is trailing a Link Down or we've missed an insertion and subsequent removal.) - Slot is in OFF_STATE and is currently occupied. => Turn the slot on. - Slot is in ON_STATE and is currently empty. => Turn the slot off. - Slot is in ON_STATE and is currently occupied. => Turn the slot off, (Be cautious and assume the card in then back on. the slot isn't the same as before.) This leads to the following simple algorithm: 1 If the slot is in ON_STATE, turn it off unconditionally. 2 If the slot is currently occupied, turn it on. Because those actions are now carried out synchronously, rather than by scheduled work items, pciehp reacts to the *current* situation and missed events no longer matter. Data Link Layer State Changed events can be handled identically to Presence Detect Changed events. Note that in the above truth table, a Link Up trailing a Card present event didn't have to be accounted for: It is filtered out by pciehp_check_link_status(). As for Attention Button Pressed events, PCIe r4.0, sec 6.7.1.5 says: "Once the Power Indicator begins blinking, a 5-second abort interval exists during which a second depression of the Attention Button cancels the operation." In other words, the user can only expect the system to react to a button press after it starts blinking. Missed button presses that occur in-between are irrelevant. Signed-off-by: Lukas Wunner <lukas@wunner.de> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com> Cc: Stefan Roese <sr@denx.de> Cc: Mayurkumar Patel <mayurkumar.patel@intel.com> Cc: Mika Westerberg <mika.westerberg@linux.intel.com> Cc: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
2018-07-20 06:27:49 +08:00
return;
}
switch (ctrl->state) {
PCI: pciehp: Enable/disable exclusively from IRQ thread Besides the IRQ thread, there are several other places in the driver which enable or disable the slot: - pciehp_probe() enables the slot if it's occupied and the pciehp_force module parameter is used. - pciehp_resume() enables or disables the slot after system sleep. - pciehp_queue_pushbutton_work() enables or disables the slot after the 5 second delay following an Attention Button press. - pciehp_sysfs_enable_slot() and pciehp_sysfs_disable_slot() enable or disable the slot on sysfs write. This requires locking and complicates pciehp's state machine. A simplification can be achieved by enabling and disabling the slot exclusively from the IRQ thread. Amend the functions listed above to request slot enable/disablement from the IRQ thread by either synthesizing a Presence Detect Changed event or, in the case of a disable user request (via sysfs or an Attention Button press), submitting a newly introduced force disable request. The latter is needed because the slot shall be forced off despite being occupied. For this force disable request, avoid colliding with Slot Status register bits by using a bit number greater than 16. For synchronous execution of requests (on sysfs write), wait for the request to finish and retrieve the result. There can only ever be one sysfs write in flight due to the locking in kernfs_fop_write(), hence there is no risk of returning the result of a different sysfs request to user space. The POWERON_STATE and POWEROFF_STATE is now no longer entered by the above-listed functions, but solely by the IRQ thread when it begins a power transition. Afterwards, it moves to STATIC_STATE. The same applies to canceling the Attention Button work, it likewise becomes an IRQ thread only operation. An immediate consequence is that the POWERON_STATE and POWEROFF_STATE is never observed by the IRQ thread itself, only by functions called in a different context, such as pciehp_sysfs_enable_slot(). So remove handling of these states from pciehp_handle_button_press() and pciehp_handle_link_change() which are exclusively called from the IRQ thread. Signed-off-by: Lukas Wunner <lukas@wunner.de> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
2018-07-20 06:27:46 +08:00
case BLINKINGON_STATE:
cancel_delayed_work(&ctrl->button_work);
/* fall through */
PCI: pciehp: Become resilient to missed events A hotplug port's Slot Status register does not count how often each type of event occurred, it only records the fact *that* an event has occurred. Previously pciehp queued a work item for each event. But if it missed an event, e.g. removal of a card in-between two back-to-back insertions, it queued up the wrong work item or no work item at all. Commit fad214b0aa72 ("PCI: pciehp: Process all hotplug events before looking for new ones") sought to improve the situation by shrinking the window during which events may be missed. But Stefan Roese reports unbalanced Card present and Link Up events, suggesting that we're still missing events if they occur very rapidly. Bjorn Helgaas responds that he considers pciehp's event handling "baroque" and calls for its simplification and rationalization: https://lkml.kernel.org/r/20180202192045.GA53759@bhelgaas-glaptop.roam.corp.google.com It gets worse once a hotplug port is runtime suspended: The port can signal an interrupt while it and its parents are in D3hot, i.e. while it is inaccessible. By the time we've runtime resumed all parents to D0 and read the port's Slot Status register, we may have missed an arbitrary number of events. Event handling therefore needs to be reworked to become resilient to missed events. Assume that a Presence Detect Changed event has occurred. Consider the following truth table: - Slot is in OFF_STATE and is currently empty. => Do nothing. (The event is trailing a Link Down or we've missed an insertion and subsequent removal.) - Slot is in OFF_STATE and is currently occupied. => Turn the slot on. - Slot is in ON_STATE and is currently empty. => Turn the slot off. - Slot is in ON_STATE and is currently occupied. => Turn the slot off, (Be cautious and assume the card in then back on. the slot isn't the same as before.) This leads to the following simple algorithm: 1 If the slot is in ON_STATE, turn it off unconditionally. 2 If the slot is currently occupied, turn it on. Because those actions are now carried out synchronously, rather than by scheduled work items, pciehp reacts to the *current* situation and missed events no longer matter. Data Link Layer State Changed events can be handled identically to Presence Detect Changed events. Note that in the above truth table, a Link Up trailing a Card present event didn't have to be accounted for: It is filtered out by pciehp_check_link_status(). As for Attention Button Pressed events, PCIe r4.0, sec 6.7.1.5 says: "Once the Power Indicator begins blinking, a 5-second abort interval exists during which a second depression of the Attention Button cancels the operation." In other words, the user can only expect the system to react to a button press after it starts blinking. Missed button presses that occur in-between are irrelevant. Signed-off-by: Lukas Wunner <lukas@wunner.de> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com> Cc: Stefan Roese <sr@denx.de> Cc: Mayurkumar Patel <mayurkumar.patel@intel.com> Cc: Mika Westerberg <mika.westerberg@linux.intel.com> Cc: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
2018-07-20 06:27:49 +08:00
case OFF_STATE:
ctrl->state = POWERON_STATE;
mutex_unlock(&ctrl->state_lock);
PCI: pciehp: Become resilient to missed events A hotplug port's Slot Status register does not count how often each type of event occurred, it only records the fact *that* an event has occurred. Previously pciehp queued a work item for each event. But if it missed an event, e.g. removal of a card in-between two back-to-back insertions, it queued up the wrong work item or no work item at all. Commit fad214b0aa72 ("PCI: pciehp: Process all hotplug events before looking for new ones") sought to improve the situation by shrinking the window during which events may be missed. But Stefan Roese reports unbalanced Card present and Link Up events, suggesting that we're still missing events if they occur very rapidly. Bjorn Helgaas responds that he considers pciehp's event handling "baroque" and calls for its simplification and rationalization: https://lkml.kernel.org/r/20180202192045.GA53759@bhelgaas-glaptop.roam.corp.google.com It gets worse once a hotplug port is runtime suspended: The port can signal an interrupt while it and its parents are in D3hot, i.e. while it is inaccessible. By the time we've runtime resumed all parents to D0 and read the port's Slot Status register, we may have missed an arbitrary number of events. Event handling therefore needs to be reworked to become resilient to missed events. Assume that a Presence Detect Changed event has occurred. Consider the following truth table: - Slot is in OFF_STATE and is currently empty. => Do nothing. (The event is trailing a Link Down or we've missed an insertion and subsequent removal.) - Slot is in OFF_STATE and is currently occupied. => Turn the slot on. - Slot is in ON_STATE and is currently empty. => Turn the slot off. - Slot is in ON_STATE and is currently occupied. => Turn the slot off, (Be cautious and assume the card in then back on. the slot isn't the same as before.) This leads to the following simple algorithm: 1 If the slot is in ON_STATE, turn it off unconditionally. 2 If the slot is currently occupied, turn it on. Because those actions are now carried out synchronously, rather than by scheduled work items, pciehp reacts to the *current* situation and missed events no longer matter. Data Link Layer State Changed events can be handled identically to Presence Detect Changed events. Note that in the above truth table, a Link Up trailing a Card present event didn't have to be accounted for: It is filtered out by pciehp_check_link_status(). As for Attention Button Pressed events, PCIe r4.0, sec 6.7.1.5 says: "Once the Power Indicator begins blinking, a 5-second abort interval exists during which a second depression of the Attention Button cancels the operation." In other words, the user can only expect the system to react to a button press after it starts blinking. Missed button presses that occur in-between are irrelevant. Signed-off-by: Lukas Wunner <lukas@wunner.de> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com> Cc: Stefan Roese <sr@denx.de> Cc: Mayurkumar Patel <mayurkumar.patel@intel.com> Cc: Mika Westerberg <mika.westerberg@linux.intel.com> Cc: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
2018-07-20 06:27:49 +08:00
if (present)
ctrl_info(ctrl, "Slot(%s): Card present\n",
slot_name(ctrl));
PCI: pciehp: Become resilient to missed events A hotplug port's Slot Status register does not count how often each type of event occurred, it only records the fact *that* an event has occurred. Previously pciehp queued a work item for each event. But if it missed an event, e.g. removal of a card in-between two back-to-back insertions, it queued up the wrong work item or no work item at all. Commit fad214b0aa72 ("PCI: pciehp: Process all hotplug events before looking for new ones") sought to improve the situation by shrinking the window during which events may be missed. But Stefan Roese reports unbalanced Card present and Link Up events, suggesting that we're still missing events if they occur very rapidly. Bjorn Helgaas responds that he considers pciehp's event handling "baroque" and calls for its simplification and rationalization: https://lkml.kernel.org/r/20180202192045.GA53759@bhelgaas-glaptop.roam.corp.google.com It gets worse once a hotplug port is runtime suspended: The port can signal an interrupt while it and its parents are in D3hot, i.e. while it is inaccessible. By the time we've runtime resumed all parents to D0 and read the port's Slot Status register, we may have missed an arbitrary number of events. Event handling therefore needs to be reworked to become resilient to missed events. Assume that a Presence Detect Changed event has occurred. Consider the following truth table: - Slot is in OFF_STATE and is currently empty. => Do nothing. (The event is trailing a Link Down or we've missed an insertion and subsequent removal.) - Slot is in OFF_STATE and is currently occupied. => Turn the slot on. - Slot is in ON_STATE and is currently empty. => Turn the slot off. - Slot is in ON_STATE and is currently occupied. => Turn the slot off, (Be cautious and assume the card in then back on. the slot isn't the same as before.) This leads to the following simple algorithm: 1 If the slot is in ON_STATE, turn it off unconditionally. 2 If the slot is currently occupied, turn it on. Because those actions are now carried out synchronously, rather than by scheduled work items, pciehp reacts to the *current* situation and missed events no longer matter. Data Link Layer State Changed events can be handled identically to Presence Detect Changed events. Note that in the above truth table, a Link Up trailing a Card present event didn't have to be accounted for: It is filtered out by pciehp_check_link_status(). As for Attention Button Pressed events, PCIe r4.0, sec 6.7.1.5 says: "Once the Power Indicator begins blinking, a 5-second abort interval exists during which a second depression of the Attention Button cancels the operation." In other words, the user can only expect the system to react to a button press after it starts blinking. Missed button presses that occur in-between are irrelevant. Signed-off-by: Lukas Wunner <lukas@wunner.de> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com> Cc: Stefan Roese <sr@denx.de> Cc: Mayurkumar Patel <mayurkumar.patel@intel.com> Cc: Mika Westerberg <mika.westerberg@linux.intel.com> Cc: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
2018-07-20 06:27:49 +08:00
if (link_active)
ctrl_info(ctrl, "Slot(%s): Link Up\n",
slot_name(ctrl));
ctrl->request_result = pciehp_enable_slot(ctrl);
PCI: pciehp: Become resilient to missed events A hotplug port's Slot Status register does not count how often each type of event occurred, it only records the fact *that* an event has occurred. Previously pciehp queued a work item for each event. But if it missed an event, e.g. removal of a card in-between two back-to-back insertions, it queued up the wrong work item or no work item at all. Commit fad214b0aa72 ("PCI: pciehp: Process all hotplug events before looking for new ones") sought to improve the situation by shrinking the window during which events may be missed. But Stefan Roese reports unbalanced Card present and Link Up events, suggesting that we're still missing events if they occur very rapidly. Bjorn Helgaas responds that he considers pciehp's event handling "baroque" and calls for its simplification and rationalization: https://lkml.kernel.org/r/20180202192045.GA53759@bhelgaas-glaptop.roam.corp.google.com It gets worse once a hotplug port is runtime suspended: The port can signal an interrupt while it and its parents are in D3hot, i.e. while it is inaccessible. By the time we've runtime resumed all parents to D0 and read the port's Slot Status register, we may have missed an arbitrary number of events. Event handling therefore needs to be reworked to become resilient to missed events. Assume that a Presence Detect Changed event has occurred. Consider the following truth table: - Slot is in OFF_STATE and is currently empty. => Do nothing. (The event is trailing a Link Down or we've missed an insertion and subsequent removal.) - Slot is in OFF_STATE and is currently occupied. => Turn the slot on. - Slot is in ON_STATE and is currently empty. => Turn the slot off. - Slot is in ON_STATE and is currently occupied. => Turn the slot off, (Be cautious and assume the card in then back on. the slot isn't the same as before.) This leads to the following simple algorithm: 1 If the slot is in ON_STATE, turn it off unconditionally. 2 If the slot is currently occupied, turn it on. Because those actions are now carried out synchronously, rather than by scheduled work items, pciehp reacts to the *current* situation and missed events no longer matter. Data Link Layer State Changed events can be handled identically to Presence Detect Changed events. Note that in the above truth table, a Link Up trailing a Card present event didn't have to be accounted for: It is filtered out by pciehp_check_link_status(). As for Attention Button Pressed events, PCIe r4.0, sec 6.7.1.5 says: "Once the Power Indicator begins blinking, a 5-second abort interval exists during which a second depression of the Attention Button cancels the operation." In other words, the user can only expect the system to react to a button press after it starts blinking. Missed button presses that occur in-between are irrelevant. Signed-off-by: Lukas Wunner <lukas@wunner.de> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com> Cc: Stefan Roese <sr@denx.de> Cc: Mayurkumar Patel <mayurkumar.patel@intel.com> Cc: Mika Westerberg <mika.westerberg@linux.intel.com> Cc: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
2018-07-20 06:27:49 +08:00
break;
default:
mutex_unlock(&ctrl->state_lock);
break;
}
}
static int __pciehp_enable_slot(struct controller *ctrl)
{
u8 getstatus = 0;
if (MRL_SENS(ctrl)) {
pciehp_get_latch_status(ctrl, &getstatus);
if (getstatus) {
ctrl_info(ctrl, "Slot(%s): Latch open\n",
slot_name(ctrl));
return -ENODEV;
}
}
if (POWER_CTRL(ctrl)) {
pciehp_get_power_status(ctrl, &getstatus);
if (getstatus) {
ctrl_info(ctrl, "Slot(%s): Already enabled\n",
slot_name(ctrl));
return 0;
}
}
return board_added(ctrl);
}
static int pciehp_enable_slot(struct controller *ctrl)
{
int ret;
PCI: pciehp: Resume to D0 on enable/disable pciehp's IRQ thread ensures accessibility of the port by runtime resuming its parent to D0. However when the slot is enabled/disabled, the port itself needs to be in D0 because its secondary bus is accessed in: pciehp_check_link_status(), pciehp_configure_device() (both called from board_added()) and pciehp_unconfigure_device() (called from remove_board()). Thus, acquire a runtime PM ref on enable/disablement of the slot. Yinghai Lu additionally discovered that some SkyLake servers feature a Power Controller for their PCIe hotplug ports (PCIe r3.1, sec 6.7.1.8) which requires the port to be in D0 when invoking pciehp_power_on_slot() (likewise called from board_added()). If slot power is turned on while in D3hot, link training later fails: https://lkml.kernel.org/r/20170205073454.GA253@wunner.de The spec is silent about such a requirement, but it seems prudent to assume that any hotplug port with a Power Controller may need this. The present commit holds a runtime PM ref whenever slot power is turned on and off, but it doesn't keep the port in D0 as long as slot power is on. If vendors determine that's necessary, they need to amend pciehp to acquire a runtime PM ref in pciehp_power_on_slot() and release one in pciehp_power_off_slot(). Signed-off-by: Lukas Wunner <lukas@wunner.de> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com> Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Cc: Mika Westerberg <mika.westerberg@linux.intel.com> Cc: Ashok Raj <ashok.raj@intel.com> Cc: Keith Busch <keith.busch@intel.com> Cc: Yinghai Lu <yinghai@kernel.org>
2018-07-20 06:27:56 +08:00
pm_runtime_get_sync(&ctrl->pcie->port->dev);
ret = __pciehp_enable_slot(ctrl);
if (ret && ATTN_BUTTN(ctrl))
/* may be blinking */
pciehp_set_indicators(ctrl, PCI_EXP_SLTCTL_PWR_IND_OFF,
INDICATOR_NOOP);
PCI: pciehp: Resume to D0 on enable/disable pciehp's IRQ thread ensures accessibility of the port by runtime resuming its parent to D0. However when the slot is enabled/disabled, the port itself needs to be in D0 because its secondary bus is accessed in: pciehp_check_link_status(), pciehp_configure_device() (both called from board_added()) and pciehp_unconfigure_device() (called from remove_board()). Thus, acquire a runtime PM ref on enable/disablement of the slot. Yinghai Lu additionally discovered that some SkyLake servers feature a Power Controller for their PCIe hotplug ports (PCIe r3.1, sec 6.7.1.8) which requires the port to be in D0 when invoking pciehp_power_on_slot() (likewise called from board_added()). If slot power is turned on while in D3hot, link training later fails: https://lkml.kernel.org/r/20170205073454.GA253@wunner.de The spec is silent about such a requirement, but it seems prudent to assume that any hotplug port with a Power Controller may need this. The present commit holds a runtime PM ref whenever slot power is turned on and off, but it doesn't keep the port in D0 as long as slot power is on. If vendors determine that's necessary, they need to amend pciehp to acquire a runtime PM ref in pciehp_power_on_slot() and release one in pciehp_power_off_slot(). Signed-off-by: Lukas Wunner <lukas@wunner.de> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com> Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Cc: Mika Westerberg <mika.westerberg@linux.intel.com> Cc: Ashok Raj <ashok.raj@intel.com> Cc: Keith Busch <keith.busch@intel.com> Cc: Yinghai Lu <yinghai@kernel.org>
2018-07-20 06:27:56 +08:00
pm_runtime_put(&ctrl->pcie->port->dev);
mutex_lock(&ctrl->state_lock);
ctrl->state = ret ? OFF_STATE : ON_STATE;
mutex_unlock(&ctrl->state_lock);
return ret;
}
static int __pciehp_disable_slot(struct controller *ctrl, bool safe_removal)
{
u8 getstatus = 0;
if (POWER_CTRL(ctrl)) {
pciehp_get_power_status(ctrl, &getstatus);
if (!getstatus) {
ctrl_info(ctrl, "Slot(%s): Already disabled\n",
slot_name(ctrl));
return -EINVAL;
}
}
remove_board(ctrl, safe_removal);
return 0;
}
static int pciehp_disable_slot(struct controller *ctrl, bool safe_removal)
{
int ret;
PCI: pciehp: Resume to D0 on enable/disable pciehp's IRQ thread ensures accessibility of the port by runtime resuming its parent to D0. However when the slot is enabled/disabled, the port itself needs to be in D0 because its secondary bus is accessed in: pciehp_check_link_status(), pciehp_configure_device() (both called from board_added()) and pciehp_unconfigure_device() (called from remove_board()). Thus, acquire a runtime PM ref on enable/disablement of the slot. Yinghai Lu additionally discovered that some SkyLake servers feature a Power Controller for their PCIe hotplug ports (PCIe r3.1, sec 6.7.1.8) which requires the port to be in D0 when invoking pciehp_power_on_slot() (likewise called from board_added()). If slot power is turned on while in D3hot, link training later fails: https://lkml.kernel.org/r/20170205073454.GA253@wunner.de The spec is silent about such a requirement, but it seems prudent to assume that any hotplug port with a Power Controller may need this. The present commit holds a runtime PM ref whenever slot power is turned on and off, but it doesn't keep the port in D0 as long as slot power is on. If vendors determine that's necessary, they need to amend pciehp to acquire a runtime PM ref in pciehp_power_on_slot() and release one in pciehp_power_off_slot(). Signed-off-by: Lukas Wunner <lukas@wunner.de> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com> Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Cc: Mika Westerberg <mika.westerberg@linux.intel.com> Cc: Ashok Raj <ashok.raj@intel.com> Cc: Keith Busch <keith.busch@intel.com> Cc: Yinghai Lu <yinghai@kernel.org>
2018-07-20 06:27:56 +08:00
pm_runtime_get_sync(&ctrl->pcie->port->dev);
ret = __pciehp_disable_slot(ctrl, safe_removal);
PCI: pciehp: Resume to D0 on enable/disable pciehp's IRQ thread ensures accessibility of the port by runtime resuming its parent to D0. However when the slot is enabled/disabled, the port itself needs to be in D0 because its secondary bus is accessed in: pciehp_check_link_status(), pciehp_configure_device() (both called from board_added()) and pciehp_unconfigure_device() (called from remove_board()). Thus, acquire a runtime PM ref on enable/disablement of the slot. Yinghai Lu additionally discovered that some SkyLake servers feature a Power Controller for their PCIe hotplug ports (PCIe r3.1, sec 6.7.1.8) which requires the port to be in D0 when invoking pciehp_power_on_slot() (likewise called from board_added()). If slot power is turned on while in D3hot, link training later fails: https://lkml.kernel.org/r/20170205073454.GA253@wunner.de The spec is silent about such a requirement, but it seems prudent to assume that any hotplug port with a Power Controller may need this. The present commit holds a runtime PM ref whenever slot power is turned on and off, but it doesn't keep the port in D0 as long as slot power is on. If vendors determine that's necessary, they need to amend pciehp to acquire a runtime PM ref in pciehp_power_on_slot() and release one in pciehp_power_off_slot(). Signed-off-by: Lukas Wunner <lukas@wunner.de> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com> Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Cc: Mika Westerberg <mika.westerberg@linux.intel.com> Cc: Ashok Raj <ashok.raj@intel.com> Cc: Keith Busch <keith.busch@intel.com> Cc: Yinghai Lu <yinghai@kernel.org>
2018-07-20 06:27:56 +08:00
pm_runtime_put(&ctrl->pcie->port->dev);
mutex_lock(&ctrl->state_lock);
ctrl->state = OFF_STATE;
mutex_unlock(&ctrl->state_lock);
return ret;
}
int pciehp_sysfs_enable_slot(struct hotplug_slot *hotplug_slot)
{
PCI: hotplug: Embed hotplug_slot When the PCI hotplug core and its first user, cpqphp, were introduced in February 2002 with historic commit a8a2069f432c, cpqphp allocated a slot struct for its internal use plus a hotplug_slot struct to be registered with the hotplug core and linked the two with pointers: https://git.kernel.org/tglx/history/c/a8a2069f432c Nowadays, the predominant pattern in the tree is to embed ("subclass") such structures in one another and cast to the containing struct with container_of(). But it wasn't until July 2002 that container_of() was introduced with historic commit ec4f214232cf: https://git.kernel.org/tglx/history/c/ec4f214232cf pnv_php, introduced in 2016, did the right thing and embedded struct hotplug_slot in its internal struct pnv_php_slot, but all other drivers cargo-culted cpqphp's design and linked separate structs with pointers. Embedding structs is preferrable to linking them with pointers because it requires fewer allocations, thereby reducing overhead and simplifying error paths. Casting an embedded struct to the containing struct becomes a cheap subtraction rather than a dereference. And having fewer pointers reduces the risk of them pointing nowhere either accidentally or due to an attack. Convert all drivers to embed struct hotplug_slot in their internal slot struct. The "private" pointer in struct hotplug_slot thereby becomes unused, so drop it. Signed-off-by: Lukas Wunner <lukas@wunner.de> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com> Reviewed-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Acked-by: Tyrel Datwyler <tyreld@linux.vnet.ibm.com> # drivers/pci/hotplug/rpa* Acked-by: Sebastian Ott <sebott@linux.ibm.com> # drivers/pci/hotplug/s390* Acked-by: Andy Shevchenko <andy.shevchenko@gmail.com> # drivers/platform/x86 Cc: Len Brown <lenb@kernel.org> Cc: Scott Murray <scott@spiteful.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Cc: Oliver OHalloran <oliveroh@au1.ibm.com> Cc: Gavin Shan <gwshan@linux.vnet.ibm.com> Cc: Gerald Schaefer <gerald.schaefer@de.ibm.com> Cc: Corentin Chary <corentin.chary@gmail.com> Cc: Darren Hart <dvhart@infradead.org>
2018-09-08 15:59:01 +08:00
struct controller *ctrl = to_ctrl(hotplug_slot);
mutex_lock(&ctrl->state_lock);
switch (ctrl->state) {
case BLINKINGON_STATE:
case OFF_STATE:
mutex_unlock(&ctrl->state_lock);
PCI: pciehp: Enable/disable exclusively from IRQ thread Besides the IRQ thread, there are several other places in the driver which enable or disable the slot: - pciehp_probe() enables the slot if it's occupied and the pciehp_force module parameter is used. - pciehp_resume() enables or disables the slot after system sleep. - pciehp_queue_pushbutton_work() enables or disables the slot after the 5 second delay following an Attention Button press. - pciehp_sysfs_enable_slot() and pciehp_sysfs_disable_slot() enable or disable the slot on sysfs write. This requires locking and complicates pciehp's state machine. A simplification can be achieved by enabling and disabling the slot exclusively from the IRQ thread. Amend the functions listed above to request slot enable/disablement from the IRQ thread by either synthesizing a Presence Detect Changed event or, in the case of a disable user request (via sysfs or an Attention Button press), submitting a newly introduced force disable request. The latter is needed because the slot shall be forced off despite being occupied. For this force disable request, avoid colliding with Slot Status register bits by using a bit number greater than 16. For synchronous execution of requests (on sysfs write), wait for the request to finish and retrieve the result. There can only ever be one sysfs write in flight due to the locking in kernfs_fop_write(), hence there is no risk of returning the result of a different sysfs request to user space. The POWERON_STATE and POWEROFF_STATE is now no longer entered by the above-listed functions, but solely by the IRQ thread when it begins a power transition. Afterwards, it moves to STATIC_STATE. The same applies to canceling the Attention Button work, it likewise becomes an IRQ thread only operation. An immediate consequence is that the POWERON_STATE and POWEROFF_STATE is never observed by the IRQ thread itself, only by functions called in a different context, such as pciehp_sysfs_enable_slot(). So remove handling of these states from pciehp_handle_button_press() and pciehp_handle_link_change() which are exclusively called from the IRQ thread. Signed-off-by: Lukas Wunner <lukas@wunner.de> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
2018-07-20 06:27:46 +08:00
/*
* The IRQ thread becomes a no-op if the user pulls out the
* card before the thread wakes up, so initialize to -ENODEV.
*/
ctrl->request_result = -ENODEV;
pciehp_request(ctrl, PCI_EXP_SLTSTA_PDC);
wait_event(ctrl->requester,
!atomic_read(&ctrl->pending_events) &&
!ctrl->ist_running);
PCI: pciehp: Enable/disable exclusively from IRQ thread Besides the IRQ thread, there are several other places in the driver which enable or disable the slot: - pciehp_probe() enables the slot if it's occupied and the pciehp_force module parameter is used. - pciehp_resume() enables or disables the slot after system sleep. - pciehp_queue_pushbutton_work() enables or disables the slot after the 5 second delay following an Attention Button press. - pciehp_sysfs_enable_slot() and pciehp_sysfs_disable_slot() enable or disable the slot on sysfs write. This requires locking and complicates pciehp's state machine. A simplification can be achieved by enabling and disabling the slot exclusively from the IRQ thread. Amend the functions listed above to request slot enable/disablement from the IRQ thread by either synthesizing a Presence Detect Changed event or, in the case of a disable user request (via sysfs or an Attention Button press), submitting a newly introduced force disable request. The latter is needed because the slot shall be forced off despite being occupied. For this force disable request, avoid colliding with Slot Status register bits by using a bit number greater than 16. For synchronous execution of requests (on sysfs write), wait for the request to finish and retrieve the result. There can only ever be one sysfs write in flight due to the locking in kernfs_fop_write(), hence there is no risk of returning the result of a different sysfs request to user space. The POWERON_STATE and POWEROFF_STATE is now no longer entered by the above-listed functions, but solely by the IRQ thread when it begins a power transition. Afterwards, it moves to STATIC_STATE. The same applies to canceling the Attention Button work, it likewise becomes an IRQ thread only operation. An immediate consequence is that the POWERON_STATE and POWEROFF_STATE is never observed by the IRQ thread itself, only by functions called in a different context, such as pciehp_sysfs_enable_slot(). So remove handling of these states from pciehp_handle_button_press() and pciehp_handle_link_change() which are exclusively called from the IRQ thread. Signed-off-by: Lukas Wunner <lukas@wunner.de> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
2018-07-20 06:27:46 +08:00
return ctrl->request_result;
case POWERON_STATE:
ctrl_info(ctrl, "Slot(%s): Already in powering on state\n",
slot_name(ctrl));
break;
case BLINKINGOFF_STATE:
case ON_STATE:
case POWEROFF_STATE:
ctrl_info(ctrl, "Slot(%s): Already enabled\n",
slot_name(ctrl));
break;
default:
ctrl_err(ctrl, "Slot(%s): Invalid state %#x\n",
slot_name(ctrl), ctrl->state);
break;
}
mutex_unlock(&ctrl->state_lock);
return -ENODEV;
}
int pciehp_sysfs_disable_slot(struct hotplug_slot *hotplug_slot)
{
PCI: hotplug: Embed hotplug_slot When the PCI hotplug core and its first user, cpqphp, were introduced in February 2002 with historic commit a8a2069f432c, cpqphp allocated a slot struct for its internal use plus a hotplug_slot struct to be registered with the hotplug core and linked the two with pointers: https://git.kernel.org/tglx/history/c/a8a2069f432c Nowadays, the predominant pattern in the tree is to embed ("subclass") such structures in one another and cast to the containing struct with container_of(). But it wasn't until July 2002 that container_of() was introduced with historic commit ec4f214232cf: https://git.kernel.org/tglx/history/c/ec4f214232cf pnv_php, introduced in 2016, did the right thing and embedded struct hotplug_slot in its internal struct pnv_php_slot, but all other drivers cargo-culted cpqphp's design and linked separate structs with pointers. Embedding structs is preferrable to linking them with pointers because it requires fewer allocations, thereby reducing overhead and simplifying error paths. Casting an embedded struct to the containing struct becomes a cheap subtraction rather than a dereference. And having fewer pointers reduces the risk of them pointing nowhere either accidentally or due to an attack. Convert all drivers to embed struct hotplug_slot in their internal slot struct. The "private" pointer in struct hotplug_slot thereby becomes unused, so drop it. Signed-off-by: Lukas Wunner <lukas@wunner.de> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com> Reviewed-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Acked-by: Tyrel Datwyler <tyreld@linux.vnet.ibm.com> # drivers/pci/hotplug/rpa* Acked-by: Sebastian Ott <sebott@linux.ibm.com> # drivers/pci/hotplug/s390* Acked-by: Andy Shevchenko <andy.shevchenko@gmail.com> # drivers/platform/x86 Cc: Len Brown <lenb@kernel.org> Cc: Scott Murray <scott@spiteful.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Cc: Oliver OHalloran <oliveroh@au1.ibm.com> Cc: Gavin Shan <gwshan@linux.vnet.ibm.com> Cc: Gerald Schaefer <gerald.schaefer@de.ibm.com> Cc: Corentin Chary <corentin.chary@gmail.com> Cc: Darren Hart <dvhart@infradead.org>
2018-09-08 15:59:01 +08:00
struct controller *ctrl = to_ctrl(hotplug_slot);
mutex_lock(&ctrl->state_lock);
switch (ctrl->state) {
case BLINKINGOFF_STATE:
case ON_STATE:
mutex_unlock(&ctrl->state_lock);
PCI: pciehp: Enable/disable exclusively from IRQ thread Besides the IRQ thread, there are several other places in the driver which enable or disable the slot: - pciehp_probe() enables the slot if it's occupied and the pciehp_force module parameter is used. - pciehp_resume() enables or disables the slot after system sleep. - pciehp_queue_pushbutton_work() enables or disables the slot after the 5 second delay following an Attention Button press. - pciehp_sysfs_enable_slot() and pciehp_sysfs_disable_slot() enable or disable the slot on sysfs write. This requires locking and complicates pciehp's state machine. A simplification can be achieved by enabling and disabling the slot exclusively from the IRQ thread. Amend the functions listed above to request slot enable/disablement from the IRQ thread by either synthesizing a Presence Detect Changed event or, in the case of a disable user request (via sysfs or an Attention Button press), submitting a newly introduced force disable request. The latter is needed because the slot shall be forced off despite being occupied. For this force disable request, avoid colliding with Slot Status register bits by using a bit number greater than 16. For synchronous execution of requests (on sysfs write), wait for the request to finish and retrieve the result. There can only ever be one sysfs write in flight due to the locking in kernfs_fop_write(), hence there is no risk of returning the result of a different sysfs request to user space. The POWERON_STATE and POWEROFF_STATE is now no longer entered by the above-listed functions, but solely by the IRQ thread when it begins a power transition. Afterwards, it moves to STATIC_STATE. The same applies to canceling the Attention Button work, it likewise becomes an IRQ thread only operation. An immediate consequence is that the POWERON_STATE and POWEROFF_STATE is never observed by the IRQ thread itself, only by functions called in a different context, such as pciehp_sysfs_enable_slot(). So remove handling of these states from pciehp_handle_button_press() and pciehp_handle_link_change() which are exclusively called from the IRQ thread. Signed-off-by: Lukas Wunner <lukas@wunner.de> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
2018-07-20 06:27:46 +08:00
pciehp_request(ctrl, DISABLE_SLOT);
wait_event(ctrl->requester,
!atomic_read(&ctrl->pending_events) &&
!ctrl->ist_running);
PCI: pciehp: Enable/disable exclusively from IRQ thread Besides the IRQ thread, there are several other places in the driver which enable or disable the slot: - pciehp_probe() enables the slot if it's occupied and the pciehp_force module parameter is used. - pciehp_resume() enables or disables the slot after system sleep. - pciehp_queue_pushbutton_work() enables or disables the slot after the 5 second delay following an Attention Button press. - pciehp_sysfs_enable_slot() and pciehp_sysfs_disable_slot() enable or disable the slot on sysfs write. This requires locking and complicates pciehp's state machine. A simplification can be achieved by enabling and disabling the slot exclusively from the IRQ thread. Amend the functions listed above to request slot enable/disablement from the IRQ thread by either synthesizing a Presence Detect Changed event or, in the case of a disable user request (via sysfs or an Attention Button press), submitting a newly introduced force disable request. The latter is needed because the slot shall be forced off despite being occupied. For this force disable request, avoid colliding with Slot Status register bits by using a bit number greater than 16. For synchronous execution of requests (on sysfs write), wait for the request to finish and retrieve the result. There can only ever be one sysfs write in flight due to the locking in kernfs_fop_write(), hence there is no risk of returning the result of a different sysfs request to user space. The POWERON_STATE and POWEROFF_STATE is now no longer entered by the above-listed functions, but solely by the IRQ thread when it begins a power transition. Afterwards, it moves to STATIC_STATE. The same applies to canceling the Attention Button work, it likewise becomes an IRQ thread only operation. An immediate consequence is that the POWERON_STATE and POWEROFF_STATE is never observed by the IRQ thread itself, only by functions called in a different context, such as pciehp_sysfs_enable_slot(). So remove handling of these states from pciehp_handle_button_press() and pciehp_handle_link_change() which are exclusively called from the IRQ thread. Signed-off-by: Lukas Wunner <lukas@wunner.de> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
2018-07-20 06:27:46 +08:00
return ctrl->request_result;
case POWEROFF_STATE:
ctrl_info(ctrl, "Slot(%s): Already in powering off state\n",
slot_name(ctrl));
break;
case BLINKINGON_STATE:
case OFF_STATE:
case POWERON_STATE:
ctrl_info(ctrl, "Slot(%s): Already disabled\n",
slot_name(ctrl));
break;
default:
ctrl_err(ctrl, "Slot(%s): Invalid state %#x\n",
slot_name(ctrl), ctrl->state);
break;
}
mutex_unlock(&ctrl->state_lock);
return -ENODEV;
}