linux/drivers/pinctrl/intel/pinctrl-intel.c

1774 lines
44 KiB
C
Raw Normal View History

// SPDX-License-Identifier: GPL-2.0
/*
* Intel pinctrl/GPIO core driver.
*
* Copyright (C) 2015, Intel Corporation
* Authors: Mathias Nyman <mathias.nyman@linux.intel.com>
* Mika Westerberg <mika.westerberg@linux.intel.com>
*/
#include <linux/acpi.h>
#include <linux/gpio/driver.h>
#include <linux/interrupt.h>
#include <linux/log2.h>
#include <linux/module.h>
#include <linux/platform_device.h>
#include <linux/property.h>
#include <linux/time.h>
#include <linux/pinctrl/pinctrl.h>
#include <linux/pinctrl/pinmux.h>
#include <linux/pinctrl/pinconf.h>
#include <linux/pinctrl/pinconf-generic.h>
#include "../core.h"
#include "pinctrl-intel.h"
/* Offset from regs */
#define REVID 0x000
#define REVID_SHIFT 16
#define REVID_MASK GENMASK(31, 16)
#define PADBAR 0x00c
#define PADOWN_BITS 4
#define PADOWN_SHIFT(p) ((p) % 8 * PADOWN_BITS)
#define PADOWN_MASK(p) (GENMASK(3, 0) << PADOWN_SHIFT(p))
#define PADOWN_GPP(p) ((p) / 8)
/* Offset from pad_regs */
#define PADCFG0 0x000
#define PADCFG0_RXEVCFG_SHIFT 25
#define PADCFG0_RXEVCFG_MASK GENMASK(26, 25)
#define PADCFG0_RXEVCFG_LEVEL 0
#define PADCFG0_RXEVCFG_EDGE 1
#define PADCFG0_RXEVCFG_DISABLED 2
#define PADCFG0_RXEVCFG_EDGE_BOTH 3
#define PADCFG0_PREGFRXSEL BIT(24)
#define PADCFG0_RXINV BIT(23)
#define PADCFG0_GPIROUTIOXAPIC BIT(20)
#define PADCFG0_GPIROUTSCI BIT(19)
#define PADCFG0_GPIROUTSMI BIT(18)
#define PADCFG0_GPIROUTNMI BIT(17)
#define PADCFG0_PMODE_SHIFT 10
#define PADCFG0_PMODE_MASK GENMASK(13, 10)
#define PADCFG0_PMODE_GPIO 0
#define PADCFG0_GPIORXDIS BIT(9)
#define PADCFG0_GPIOTXDIS BIT(8)
#define PADCFG0_GPIORXSTATE BIT(1)
#define PADCFG0_GPIOTXSTATE BIT(0)
#define PADCFG1 0x004
#define PADCFG1_TERM_UP BIT(13)
#define PADCFG1_TERM_SHIFT 10
#define PADCFG1_TERM_MASK GENMASK(12, 10)
#define PADCFG1_TERM_20K BIT(2)
#define PADCFG1_TERM_5K BIT(1)
#define PADCFG1_TERM_1K BIT(0)
#define PADCFG1_TERM_833 (BIT(1) | BIT(0))
#define PADCFG2 0x008
#define PADCFG2_DEBEN BIT(0)
#define PADCFG2_DEBOUNCE_SHIFT 1
#define PADCFG2_DEBOUNCE_MASK GENMASK(4, 1)
#define DEBOUNCE_PERIOD_NSEC 31250
struct intel_pad_context {
u32 padcfg0;
u32 padcfg1;
u32 padcfg2;
};
struct intel_community_context {
u32 *intmask;
u32 *hostown;
};
#define pin_to_padno(c, p) ((p) - (c)->pin_base)
#define padgroup_offset(g, p) ((p) - (g)->base)
static struct intel_community *intel_get_community(struct intel_pinctrl *pctrl,
unsigned int pin)
{
struct intel_community *community;
int i;
for (i = 0; i < pctrl->ncommunities; i++) {
community = &pctrl->communities[i];
if (pin >= community->pin_base &&
pin < community->pin_base + community->npins)
return community;
}
dev_warn(pctrl->dev, "failed to find community for pin %u\n", pin);
return NULL;
}
static const struct intel_padgroup *
intel_community_get_padgroup(const struct intel_community *community,
unsigned int pin)
{
int i;
for (i = 0; i < community->ngpps; i++) {
const struct intel_padgroup *padgrp = &community->gpps[i];
if (pin >= padgrp->base && pin < padgrp->base + padgrp->size)
return padgrp;
}
return NULL;
}
static void __iomem *intel_get_padcfg(struct intel_pinctrl *pctrl,
unsigned int pin, unsigned int reg)
{
const struct intel_community *community;
unsigned int padno;
size_t nregs;
community = intel_get_community(pctrl, pin);
if (!community)
return NULL;
padno = pin_to_padno(community, pin);
nregs = (community->features & PINCTRL_FEATURE_DEBOUNCE) ? 4 : 2;
if (reg >= nregs * 4)
return NULL;
return community->pad_regs + reg + padno * nregs * 4;
}
static bool intel_pad_owned_by_host(struct intel_pinctrl *pctrl, unsigned int pin)
{
const struct intel_community *community;
const struct intel_padgroup *padgrp;
unsigned int gpp, offset, gpp_offset;
void __iomem *padown;
community = intel_get_community(pctrl, pin);
if (!community)
return false;
if (!community->padown_offset)
return true;
padgrp = intel_community_get_padgroup(community, pin);
if (!padgrp)
return false;
gpp_offset = padgroup_offset(padgrp, pin);
gpp = PADOWN_GPP(gpp_offset);
offset = community->padown_offset + padgrp->padown_num * 4 + gpp * 4;
padown = community->regs + offset;
return !(readl(padown) & PADOWN_MASK(gpp_offset));
}
static bool intel_pad_acpi_mode(struct intel_pinctrl *pctrl, unsigned int pin)
{
const struct intel_community *community;
const struct intel_padgroup *padgrp;
unsigned int offset, gpp_offset;
void __iomem *hostown;
community = intel_get_community(pctrl, pin);
if (!community)
return true;
if (!community->hostown_offset)
return false;
padgrp = intel_community_get_padgroup(community, pin);
if (!padgrp)
return true;
gpp_offset = padgroup_offset(padgrp, pin);
offset = community->hostown_offset + padgrp->reg_num * 4;
hostown = community->regs + offset;
return !(readl(hostown) & BIT(gpp_offset));
}
/**
* enum - Locking variants of the pad configuration
*
* @PAD_UNLOCKED: pad is fully controlled by the configuration registers
* @PAD_LOCKED: pad configuration registers, except TX state, are locked
* @PAD_LOCKED_TX: pad configuration TX state is locked
* @PAD_LOCKED_FULL: pad configuration registers are locked completely
*
* Locking is considered as read-only mode for corresponding registers and
* their respective fields. That said, TX state bit is locked separately from
* the main locking scheme.
*/
enum {
PAD_UNLOCKED = 0,
PAD_LOCKED = 1,
PAD_LOCKED_TX = 2,
PAD_LOCKED_FULL = PAD_LOCKED | PAD_LOCKED_TX,
};
static int intel_pad_locked(struct intel_pinctrl *pctrl, unsigned int pin)
{
struct intel_community *community;
const struct intel_padgroup *padgrp;
unsigned int offset, gpp_offset;
u32 value;
int ret = PAD_UNLOCKED;
community = intel_get_community(pctrl, pin);
if (!community)
return PAD_LOCKED_FULL;
if (!community->padcfglock_offset)
return PAD_UNLOCKED;
padgrp = intel_community_get_padgroup(community, pin);
if (!padgrp)
return PAD_LOCKED_FULL;
gpp_offset = padgroup_offset(padgrp, pin);
/*
* If PADCFGLOCK and PADCFGLOCKTX bits are both clear for this pad,
* the pad is considered unlocked. Any other case means that it is
* either fully or partially locked.
*/
offset = community->padcfglock_offset + 0 + padgrp->reg_num * 8;
value = readl(community->regs + offset);
if (value & BIT(gpp_offset))
ret |= PAD_LOCKED;
offset = community->padcfglock_offset + 4 + padgrp->reg_num * 8;
value = readl(community->regs + offset);
if (value & BIT(gpp_offset))
ret |= PAD_LOCKED_TX;
return ret;
}
static bool intel_pad_is_unlocked(struct intel_pinctrl *pctrl, unsigned int pin)
{
return (intel_pad_locked(pctrl, pin) & PAD_LOCKED) == PAD_UNLOCKED;
}
static bool intel_pad_usable(struct intel_pinctrl *pctrl, unsigned int pin)
{
return intel_pad_owned_by_host(pctrl, pin) && intel_pad_is_unlocked(pctrl, pin);
}
static int intel_get_groups_count(struct pinctrl_dev *pctldev)
{
struct intel_pinctrl *pctrl = pinctrl_dev_get_drvdata(pctldev);
return pctrl->soc->ngroups;
}
static const char *intel_get_group_name(struct pinctrl_dev *pctldev,
unsigned int group)
{
struct intel_pinctrl *pctrl = pinctrl_dev_get_drvdata(pctldev);
return pctrl->soc->groups[group].name;
}
static int intel_get_group_pins(struct pinctrl_dev *pctldev, unsigned int group,
const unsigned int **pins, unsigned int *npins)
{
struct intel_pinctrl *pctrl = pinctrl_dev_get_drvdata(pctldev);
*pins = pctrl->soc->groups[group].pins;
*npins = pctrl->soc->groups[group].npins;
return 0;
}
static void intel_pin_dbg_show(struct pinctrl_dev *pctldev, struct seq_file *s,
unsigned int pin)
{
struct intel_pinctrl *pctrl = pinctrl_dev_get_drvdata(pctldev);
void __iomem *padcfg;
u32 cfg0, cfg1, mode;
int locked;
bool acpi;
if (!intel_pad_owned_by_host(pctrl, pin)) {
seq_puts(s, "not available");
return;
}
cfg0 = readl(intel_get_padcfg(pctrl, pin, PADCFG0));
cfg1 = readl(intel_get_padcfg(pctrl, pin, PADCFG1));
mode = (cfg0 & PADCFG0_PMODE_MASK) >> PADCFG0_PMODE_SHIFT;
if (mode == PADCFG0_PMODE_GPIO)
seq_puts(s, "GPIO ");
else
seq_printf(s, "mode %d ", mode);
seq_printf(s, "0x%08x 0x%08x", cfg0, cfg1);
/* Dump the additional PADCFG registers if available */
padcfg = intel_get_padcfg(pctrl, pin, PADCFG2);
if (padcfg)
seq_printf(s, " 0x%08x", readl(padcfg));
locked = intel_pad_locked(pctrl, pin);
acpi = intel_pad_acpi_mode(pctrl, pin);
if (locked || acpi) {
seq_puts(s, " [");
if (locked)
seq_puts(s, "LOCKED");
if ((locked & PAD_LOCKED_FULL) == PAD_LOCKED_TX)
seq_puts(s, " tx");
else if ((locked & PAD_LOCKED_FULL) == PAD_LOCKED_FULL)
seq_puts(s, " full");
if (locked && acpi)
seq_puts(s, ", ");
if (acpi)
seq_puts(s, "ACPI");
seq_puts(s, "]");
}
}
static const struct pinctrl_ops intel_pinctrl_ops = {
.get_groups_count = intel_get_groups_count,
.get_group_name = intel_get_group_name,
.get_group_pins = intel_get_group_pins,
.pin_dbg_show = intel_pin_dbg_show,
};
static int intel_get_functions_count(struct pinctrl_dev *pctldev)
{
struct intel_pinctrl *pctrl = pinctrl_dev_get_drvdata(pctldev);
return pctrl->soc->nfunctions;
}
static const char *intel_get_function_name(struct pinctrl_dev *pctldev,
unsigned int function)
{
struct intel_pinctrl *pctrl = pinctrl_dev_get_drvdata(pctldev);
return pctrl->soc->functions[function].name;
}
static int intel_get_function_groups(struct pinctrl_dev *pctldev,
unsigned int function,
const char * const **groups,
unsigned int * const ngroups)
{
struct intel_pinctrl *pctrl = pinctrl_dev_get_drvdata(pctldev);
*groups = pctrl->soc->functions[function].groups;
*ngroups = pctrl->soc->functions[function].ngroups;
return 0;
}
static int intel_pinmux_set_mux(struct pinctrl_dev *pctldev,
unsigned int function, unsigned int group)
{
struct intel_pinctrl *pctrl = pinctrl_dev_get_drvdata(pctldev);
const struct intel_pingroup *grp = &pctrl->soc->groups[group];
unsigned long flags;
int i;
pinctrl: intel: Use raw_spinlock for locking When running -rt kernel and GPIO interrupt happens we get following BUG: sleeping function called from invalid context at kernel/locking/rtmutex.c:931 in_atomic(): 1, irqs_disabled(): 0, pid: 530, name: irq/14-INT3452: Preemption disabled at:[<ffffffff810b4dab>] handle_edge_irq+0x1b/0x190 CPU: 0 PID: 530 Comm: irq/14-INT3452: Not tainted 4.6.2-rt5 #1060 0000000000000000 ffff88007a257d58 ffffffff812d8494 0000000000000000 ffff88017a330000 ffff88007a257d78 ffffffff81083a11 ffff88007a252430 ffff88007a252430 ffff88007a257d90 ffffffff8167ef20 000000000000001a Call Trace: [<ffffffff812d8494>] dump_stack+0x4f/0x6b [<ffffffff81083a11>] ___might_sleep+0xe1/0x160 [<ffffffff8167ef20>] rt_spin_lock+0x20/0x50 [<ffffffff81308c6d>] intel_gpio_irq_ack+0x2d/0x80 [<ffffffff810b4e0b>] handle_edge_irq+0x7b/0x190 [<ffffffff810b0d82>] generic_handle_irq+0x22/0x30 [<ffffffff81307abc>] intel_gpio_irq+0xdc/0x150 [<ffffffff810b2293>] irq_forced_thread_fn+0x23/0x70 [<ffffffff810b250b>] irq_thread+0x13b/0x1d0 [<ffffffff8167b844>] ? __schedule+0x2e4/0x5a0 [<ffffffff810b2270>] ? irq_finalize_oneshot.part.37+0xd0/0xd0 [<ffffffff810b25a0>] ? irq_thread+0x1d0/0x1d0 [<ffffffff810b23d0>] ? wake_threads_waitq+0x30/0x30 [<ffffffff8107e624>] kthread+0xd4/0xf0 [<ffffffff8167ec27>] ? _raw_spin_unlock_irq+0x17/0x40 [<ffffffff8167f592>] ret_from_fork+0x22/0x40 [<ffffffff8107e550>] ? kthread_worker_fn+0x190/0x190 The reason why this happens is because intel_gpio_irq_ack() is called with desc->lock raw_spinlock locked which cannot sleep but our normal spinlock (which is converted to rtmutex in -rt) is allowed to sleep. This causes might_sleep() to trigger. Fix this by converting the normal spinlock to a raw_spinlock. Reported-by: Kim Tatt Chuah <kim.tatt.chuah@intel.com> Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-06-16 08:25:36 +00:00
raw_spin_lock_irqsave(&pctrl->lock, flags);
/*
* All pins in the groups needs to be accessible and writable
* before we can enable the mux for this group.
*/
for (i = 0; i < grp->npins; i++) {
if (!intel_pad_usable(pctrl, grp->pins[i])) {
pinctrl: intel: Use raw_spinlock for locking When running -rt kernel and GPIO interrupt happens we get following BUG: sleeping function called from invalid context at kernel/locking/rtmutex.c:931 in_atomic(): 1, irqs_disabled(): 0, pid: 530, name: irq/14-INT3452: Preemption disabled at:[<ffffffff810b4dab>] handle_edge_irq+0x1b/0x190 CPU: 0 PID: 530 Comm: irq/14-INT3452: Not tainted 4.6.2-rt5 #1060 0000000000000000 ffff88007a257d58 ffffffff812d8494 0000000000000000 ffff88017a330000 ffff88007a257d78 ffffffff81083a11 ffff88007a252430 ffff88007a252430 ffff88007a257d90 ffffffff8167ef20 000000000000001a Call Trace: [<ffffffff812d8494>] dump_stack+0x4f/0x6b [<ffffffff81083a11>] ___might_sleep+0xe1/0x160 [<ffffffff8167ef20>] rt_spin_lock+0x20/0x50 [<ffffffff81308c6d>] intel_gpio_irq_ack+0x2d/0x80 [<ffffffff810b4e0b>] handle_edge_irq+0x7b/0x190 [<ffffffff810b0d82>] generic_handle_irq+0x22/0x30 [<ffffffff81307abc>] intel_gpio_irq+0xdc/0x150 [<ffffffff810b2293>] irq_forced_thread_fn+0x23/0x70 [<ffffffff810b250b>] irq_thread+0x13b/0x1d0 [<ffffffff8167b844>] ? __schedule+0x2e4/0x5a0 [<ffffffff810b2270>] ? irq_finalize_oneshot.part.37+0xd0/0xd0 [<ffffffff810b25a0>] ? irq_thread+0x1d0/0x1d0 [<ffffffff810b23d0>] ? wake_threads_waitq+0x30/0x30 [<ffffffff8107e624>] kthread+0xd4/0xf0 [<ffffffff8167ec27>] ? _raw_spin_unlock_irq+0x17/0x40 [<ffffffff8167f592>] ret_from_fork+0x22/0x40 [<ffffffff8107e550>] ? kthread_worker_fn+0x190/0x190 The reason why this happens is because intel_gpio_irq_ack() is called with desc->lock raw_spinlock locked which cannot sleep but our normal spinlock (which is converted to rtmutex in -rt) is allowed to sleep. This causes might_sleep() to trigger. Fix this by converting the normal spinlock to a raw_spinlock. Reported-by: Kim Tatt Chuah <kim.tatt.chuah@intel.com> Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-06-16 08:25:36 +00:00
raw_spin_unlock_irqrestore(&pctrl->lock, flags);
return -EBUSY;
}
}
/* Now enable the mux setting for each pin in the group */
for (i = 0; i < grp->npins; i++) {
void __iomem *padcfg0;
u32 value;
padcfg0 = intel_get_padcfg(pctrl, grp->pins[i], PADCFG0);
value = readl(padcfg0);
value &= ~PADCFG0_PMODE_MASK;
if (grp->modes)
value |= grp->modes[i] << PADCFG0_PMODE_SHIFT;
else
value |= grp->mode << PADCFG0_PMODE_SHIFT;
writel(value, padcfg0);
}
pinctrl: intel: Use raw_spinlock for locking When running -rt kernel and GPIO interrupt happens we get following BUG: sleeping function called from invalid context at kernel/locking/rtmutex.c:931 in_atomic(): 1, irqs_disabled(): 0, pid: 530, name: irq/14-INT3452: Preemption disabled at:[<ffffffff810b4dab>] handle_edge_irq+0x1b/0x190 CPU: 0 PID: 530 Comm: irq/14-INT3452: Not tainted 4.6.2-rt5 #1060 0000000000000000 ffff88007a257d58 ffffffff812d8494 0000000000000000 ffff88017a330000 ffff88007a257d78 ffffffff81083a11 ffff88007a252430 ffff88007a252430 ffff88007a257d90 ffffffff8167ef20 000000000000001a Call Trace: [<ffffffff812d8494>] dump_stack+0x4f/0x6b [<ffffffff81083a11>] ___might_sleep+0xe1/0x160 [<ffffffff8167ef20>] rt_spin_lock+0x20/0x50 [<ffffffff81308c6d>] intel_gpio_irq_ack+0x2d/0x80 [<ffffffff810b4e0b>] handle_edge_irq+0x7b/0x190 [<ffffffff810b0d82>] generic_handle_irq+0x22/0x30 [<ffffffff81307abc>] intel_gpio_irq+0xdc/0x150 [<ffffffff810b2293>] irq_forced_thread_fn+0x23/0x70 [<ffffffff810b250b>] irq_thread+0x13b/0x1d0 [<ffffffff8167b844>] ? __schedule+0x2e4/0x5a0 [<ffffffff810b2270>] ? irq_finalize_oneshot.part.37+0xd0/0xd0 [<ffffffff810b25a0>] ? irq_thread+0x1d0/0x1d0 [<ffffffff810b23d0>] ? wake_threads_waitq+0x30/0x30 [<ffffffff8107e624>] kthread+0xd4/0xf0 [<ffffffff8167ec27>] ? _raw_spin_unlock_irq+0x17/0x40 [<ffffffff8167f592>] ret_from_fork+0x22/0x40 [<ffffffff8107e550>] ? kthread_worker_fn+0x190/0x190 The reason why this happens is because intel_gpio_irq_ack() is called with desc->lock raw_spinlock locked which cannot sleep but our normal spinlock (which is converted to rtmutex in -rt) is allowed to sleep. This causes might_sleep() to trigger. Fix this by converting the normal spinlock to a raw_spinlock. Reported-by: Kim Tatt Chuah <kim.tatt.chuah@intel.com> Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-06-16 08:25:36 +00:00
raw_spin_unlock_irqrestore(&pctrl->lock, flags);
return 0;
}
static void __intel_gpio_set_direction(void __iomem *padcfg0, bool input)
{
u32 value;
value = readl(padcfg0);
if (input) {
value &= ~PADCFG0_GPIORXDIS;
value |= PADCFG0_GPIOTXDIS;
} else {
value &= ~PADCFG0_GPIOTXDIS;
value |= PADCFG0_GPIORXDIS;
}
writel(value, padcfg0);
}
static int intel_gpio_get_gpio_mode(void __iomem *padcfg0)
{
return (readl(padcfg0) & PADCFG0_PMODE_MASK) >> PADCFG0_PMODE_SHIFT;
}
static void intel_gpio_set_gpio_mode(void __iomem *padcfg0)
{
u32 value;
value = readl(padcfg0);
/* Put the pad into GPIO mode */
value &= ~PADCFG0_PMODE_MASK;
value |= PADCFG0_PMODE_GPIO;
/* Disable input and output buffers */
value &= ~PADCFG0_GPIORXDIS;
value &= ~PADCFG0_GPIOTXDIS;
/* Disable SCI/SMI/NMI generation */
value &= ~(PADCFG0_GPIROUTIOXAPIC | PADCFG0_GPIROUTSCI);
value &= ~(PADCFG0_GPIROUTSMI | PADCFG0_GPIROUTNMI);
writel(value, padcfg0);
}
static int intel_gpio_request_enable(struct pinctrl_dev *pctldev,
struct pinctrl_gpio_range *range,
unsigned int pin)
{
struct intel_pinctrl *pctrl = pinctrl_dev_get_drvdata(pctldev);
void __iomem *padcfg0;
unsigned long flags;
padcfg0 = intel_get_padcfg(pctrl, pin, PADCFG0);
pinctrl: intel: Use raw_spinlock for locking When running -rt kernel and GPIO interrupt happens we get following BUG: sleeping function called from invalid context at kernel/locking/rtmutex.c:931 in_atomic(): 1, irqs_disabled(): 0, pid: 530, name: irq/14-INT3452: Preemption disabled at:[<ffffffff810b4dab>] handle_edge_irq+0x1b/0x190 CPU: 0 PID: 530 Comm: irq/14-INT3452: Not tainted 4.6.2-rt5 #1060 0000000000000000 ffff88007a257d58 ffffffff812d8494 0000000000000000 ffff88017a330000 ffff88007a257d78 ffffffff81083a11 ffff88007a252430 ffff88007a252430 ffff88007a257d90 ffffffff8167ef20 000000000000001a Call Trace: [<ffffffff812d8494>] dump_stack+0x4f/0x6b [<ffffffff81083a11>] ___might_sleep+0xe1/0x160 [<ffffffff8167ef20>] rt_spin_lock+0x20/0x50 [<ffffffff81308c6d>] intel_gpio_irq_ack+0x2d/0x80 [<ffffffff810b4e0b>] handle_edge_irq+0x7b/0x190 [<ffffffff810b0d82>] generic_handle_irq+0x22/0x30 [<ffffffff81307abc>] intel_gpio_irq+0xdc/0x150 [<ffffffff810b2293>] irq_forced_thread_fn+0x23/0x70 [<ffffffff810b250b>] irq_thread+0x13b/0x1d0 [<ffffffff8167b844>] ? __schedule+0x2e4/0x5a0 [<ffffffff810b2270>] ? irq_finalize_oneshot.part.37+0xd0/0xd0 [<ffffffff810b25a0>] ? irq_thread+0x1d0/0x1d0 [<ffffffff810b23d0>] ? wake_threads_waitq+0x30/0x30 [<ffffffff8107e624>] kthread+0xd4/0xf0 [<ffffffff8167ec27>] ? _raw_spin_unlock_irq+0x17/0x40 [<ffffffff8167f592>] ret_from_fork+0x22/0x40 [<ffffffff8107e550>] ? kthread_worker_fn+0x190/0x190 The reason why this happens is because intel_gpio_irq_ack() is called with desc->lock raw_spinlock locked which cannot sleep but our normal spinlock (which is converted to rtmutex in -rt) is allowed to sleep. This causes might_sleep() to trigger. Fix this by converting the normal spinlock to a raw_spinlock. Reported-by: Kim Tatt Chuah <kim.tatt.chuah@intel.com> Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-06-16 08:25:36 +00:00
raw_spin_lock_irqsave(&pctrl->lock, flags);
if (!intel_pad_owned_by_host(pctrl, pin)) {
pinctrl: intel: Use raw_spinlock for locking When running -rt kernel and GPIO interrupt happens we get following BUG: sleeping function called from invalid context at kernel/locking/rtmutex.c:931 in_atomic(): 1, irqs_disabled(): 0, pid: 530, name: irq/14-INT3452: Preemption disabled at:[<ffffffff810b4dab>] handle_edge_irq+0x1b/0x190 CPU: 0 PID: 530 Comm: irq/14-INT3452: Not tainted 4.6.2-rt5 #1060 0000000000000000 ffff88007a257d58 ffffffff812d8494 0000000000000000 ffff88017a330000 ffff88007a257d78 ffffffff81083a11 ffff88007a252430 ffff88007a252430 ffff88007a257d90 ffffffff8167ef20 000000000000001a Call Trace: [<ffffffff812d8494>] dump_stack+0x4f/0x6b [<ffffffff81083a11>] ___might_sleep+0xe1/0x160 [<ffffffff8167ef20>] rt_spin_lock+0x20/0x50 [<ffffffff81308c6d>] intel_gpio_irq_ack+0x2d/0x80 [<ffffffff810b4e0b>] handle_edge_irq+0x7b/0x190 [<ffffffff810b0d82>] generic_handle_irq+0x22/0x30 [<ffffffff81307abc>] intel_gpio_irq+0xdc/0x150 [<ffffffff810b2293>] irq_forced_thread_fn+0x23/0x70 [<ffffffff810b250b>] irq_thread+0x13b/0x1d0 [<ffffffff8167b844>] ? __schedule+0x2e4/0x5a0 [<ffffffff810b2270>] ? irq_finalize_oneshot.part.37+0xd0/0xd0 [<ffffffff810b25a0>] ? irq_thread+0x1d0/0x1d0 [<ffffffff810b23d0>] ? wake_threads_waitq+0x30/0x30 [<ffffffff8107e624>] kthread+0xd4/0xf0 [<ffffffff8167ec27>] ? _raw_spin_unlock_irq+0x17/0x40 [<ffffffff8167f592>] ret_from_fork+0x22/0x40 [<ffffffff8107e550>] ? kthread_worker_fn+0x190/0x190 The reason why this happens is because intel_gpio_irq_ack() is called with desc->lock raw_spinlock locked which cannot sleep but our normal spinlock (which is converted to rtmutex in -rt) is allowed to sleep. This causes might_sleep() to trigger. Fix this by converting the normal spinlock to a raw_spinlock. Reported-by: Kim Tatt Chuah <kim.tatt.chuah@intel.com> Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-06-16 08:25:36 +00:00
raw_spin_unlock_irqrestore(&pctrl->lock, flags);
return -EBUSY;
}
if (!intel_pad_is_unlocked(pctrl, pin)) {
raw_spin_unlock_irqrestore(&pctrl->lock, flags);
return 0;
}
/*
* If pin is already configured in GPIO mode, we assume that
* firmware provides correct settings. In such case we avoid
* potential glitches on the pin. Otherwise, for the pin in
* alternative mode, consumer has to supply respective flags.
*/
if (intel_gpio_get_gpio_mode(padcfg0) == PADCFG0_PMODE_GPIO) {
raw_spin_unlock_irqrestore(&pctrl->lock, flags);
return 0;
}
intel_gpio_set_gpio_mode(padcfg0);
/* Disable TX buffer and enable RX (this will be input) */
__intel_gpio_set_direction(padcfg0, true);
pinctrl: intel: Use raw_spinlock for locking When running -rt kernel and GPIO interrupt happens we get following BUG: sleeping function called from invalid context at kernel/locking/rtmutex.c:931 in_atomic(): 1, irqs_disabled(): 0, pid: 530, name: irq/14-INT3452: Preemption disabled at:[<ffffffff810b4dab>] handle_edge_irq+0x1b/0x190 CPU: 0 PID: 530 Comm: irq/14-INT3452: Not tainted 4.6.2-rt5 #1060 0000000000000000 ffff88007a257d58 ffffffff812d8494 0000000000000000 ffff88017a330000 ffff88007a257d78 ffffffff81083a11 ffff88007a252430 ffff88007a252430 ffff88007a257d90 ffffffff8167ef20 000000000000001a Call Trace: [<ffffffff812d8494>] dump_stack+0x4f/0x6b [<ffffffff81083a11>] ___might_sleep+0xe1/0x160 [<ffffffff8167ef20>] rt_spin_lock+0x20/0x50 [<ffffffff81308c6d>] intel_gpio_irq_ack+0x2d/0x80 [<ffffffff810b4e0b>] handle_edge_irq+0x7b/0x190 [<ffffffff810b0d82>] generic_handle_irq+0x22/0x30 [<ffffffff81307abc>] intel_gpio_irq+0xdc/0x150 [<ffffffff810b2293>] irq_forced_thread_fn+0x23/0x70 [<ffffffff810b250b>] irq_thread+0x13b/0x1d0 [<ffffffff8167b844>] ? __schedule+0x2e4/0x5a0 [<ffffffff810b2270>] ? irq_finalize_oneshot.part.37+0xd0/0xd0 [<ffffffff810b25a0>] ? irq_thread+0x1d0/0x1d0 [<ffffffff810b23d0>] ? wake_threads_waitq+0x30/0x30 [<ffffffff8107e624>] kthread+0xd4/0xf0 [<ffffffff8167ec27>] ? _raw_spin_unlock_irq+0x17/0x40 [<ffffffff8167f592>] ret_from_fork+0x22/0x40 [<ffffffff8107e550>] ? kthread_worker_fn+0x190/0x190 The reason why this happens is because intel_gpio_irq_ack() is called with desc->lock raw_spinlock locked which cannot sleep but our normal spinlock (which is converted to rtmutex in -rt) is allowed to sleep. This causes might_sleep() to trigger. Fix this by converting the normal spinlock to a raw_spinlock. Reported-by: Kim Tatt Chuah <kim.tatt.chuah@intel.com> Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-06-16 08:25:36 +00:00
raw_spin_unlock_irqrestore(&pctrl->lock, flags);
return 0;
}
static int intel_gpio_set_direction(struct pinctrl_dev *pctldev,
struct pinctrl_gpio_range *range,
unsigned int pin, bool input)
{
struct intel_pinctrl *pctrl = pinctrl_dev_get_drvdata(pctldev);
void __iomem *padcfg0;
unsigned long flags;
padcfg0 = intel_get_padcfg(pctrl, pin, PADCFG0);
raw_spin_lock_irqsave(&pctrl->lock, flags);
__intel_gpio_set_direction(padcfg0, input);
pinctrl: intel: Use raw_spinlock for locking When running -rt kernel and GPIO interrupt happens we get following BUG: sleeping function called from invalid context at kernel/locking/rtmutex.c:931 in_atomic(): 1, irqs_disabled(): 0, pid: 530, name: irq/14-INT3452: Preemption disabled at:[<ffffffff810b4dab>] handle_edge_irq+0x1b/0x190 CPU: 0 PID: 530 Comm: irq/14-INT3452: Not tainted 4.6.2-rt5 #1060 0000000000000000 ffff88007a257d58 ffffffff812d8494 0000000000000000 ffff88017a330000 ffff88007a257d78 ffffffff81083a11 ffff88007a252430 ffff88007a252430 ffff88007a257d90 ffffffff8167ef20 000000000000001a Call Trace: [<ffffffff812d8494>] dump_stack+0x4f/0x6b [<ffffffff81083a11>] ___might_sleep+0xe1/0x160 [<ffffffff8167ef20>] rt_spin_lock+0x20/0x50 [<ffffffff81308c6d>] intel_gpio_irq_ack+0x2d/0x80 [<ffffffff810b4e0b>] handle_edge_irq+0x7b/0x190 [<ffffffff810b0d82>] generic_handle_irq+0x22/0x30 [<ffffffff81307abc>] intel_gpio_irq+0xdc/0x150 [<ffffffff810b2293>] irq_forced_thread_fn+0x23/0x70 [<ffffffff810b250b>] irq_thread+0x13b/0x1d0 [<ffffffff8167b844>] ? __schedule+0x2e4/0x5a0 [<ffffffff810b2270>] ? irq_finalize_oneshot.part.37+0xd0/0xd0 [<ffffffff810b25a0>] ? irq_thread+0x1d0/0x1d0 [<ffffffff810b23d0>] ? wake_threads_waitq+0x30/0x30 [<ffffffff8107e624>] kthread+0xd4/0xf0 [<ffffffff8167ec27>] ? _raw_spin_unlock_irq+0x17/0x40 [<ffffffff8167f592>] ret_from_fork+0x22/0x40 [<ffffffff8107e550>] ? kthread_worker_fn+0x190/0x190 The reason why this happens is because intel_gpio_irq_ack() is called with desc->lock raw_spinlock locked which cannot sleep but our normal spinlock (which is converted to rtmutex in -rt) is allowed to sleep. This causes might_sleep() to trigger. Fix this by converting the normal spinlock to a raw_spinlock. Reported-by: Kim Tatt Chuah <kim.tatt.chuah@intel.com> Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-06-16 08:25:36 +00:00
raw_spin_unlock_irqrestore(&pctrl->lock, flags);
return 0;
}
static const struct pinmux_ops intel_pinmux_ops = {
.get_functions_count = intel_get_functions_count,
.get_function_name = intel_get_function_name,
.get_function_groups = intel_get_function_groups,
.set_mux = intel_pinmux_set_mux,
.gpio_request_enable = intel_gpio_request_enable,
.gpio_set_direction = intel_gpio_set_direction,
};
static int intel_config_get_pull(struct intel_pinctrl *pctrl, unsigned int pin,
enum pin_config_param param, u32 *arg)
{
const struct intel_community *community;
void __iomem *padcfg1;
unsigned long flags;
u32 value, term;
community = intel_get_community(pctrl, pin);
padcfg1 = intel_get_padcfg(pctrl, pin, PADCFG1);
raw_spin_lock_irqsave(&pctrl->lock, flags);
value = readl(padcfg1);
raw_spin_unlock_irqrestore(&pctrl->lock, flags);
term = (value & PADCFG1_TERM_MASK) >> PADCFG1_TERM_SHIFT;
switch (param) {
case PIN_CONFIG_BIAS_DISABLE:
if (term)
return -EINVAL;
break;
case PIN_CONFIG_BIAS_PULL_UP:
if (!term || !(value & PADCFG1_TERM_UP))
return -EINVAL;
switch (term) {
case PADCFG1_TERM_833:
*arg = 833;
break;
case PADCFG1_TERM_1K:
*arg = 1000;
break;
case PADCFG1_TERM_5K:
*arg = 5000;
break;
case PADCFG1_TERM_20K:
*arg = 20000;
break;
}
break;
case PIN_CONFIG_BIAS_PULL_DOWN:
if (!term || value & PADCFG1_TERM_UP)
return -EINVAL;
switch (term) {
case PADCFG1_TERM_833:
if (!(community->features & PINCTRL_FEATURE_1K_PD))
return -EINVAL;
*arg = 833;
break;
case PADCFG1_TERM_1K:
if (!(community->features & PINCTRL_FEATURE_1K_PD))
return -EINVAL;
*arg = 1000;
break;
case PADCFG1_TERM_5K:
*arg = 5000;
break;
case PADCFG1_TERM_20K:
*arg = 20000;
break;
}
break;
default:
return -EINVAL;
}
return 0;
}
static int intel_config_get_debounce(struct intel_pinctrl *pctrl, unsigned int pin,
enum pin_config_param param, u32 *arg)
{
void __iomem *padcfg2;
unsigned long flags;
unsigned long v;
u32 value2;
padcfg2 = intel_get_padcfg(pctrl, pin, PADCFG2);
if (!padcfg2)
return -ENOTSUPP;
raw_spin_lock_irqsave(&pctrl->lock, flags);
value2 = readl(padcfg2);
raw_spin_unlock_irqrestore(&pctrl->lock, flags);
if (!(value2 & PADCFG2_DEBEN))
return -EINVAL;
v = (value2 & PADCFG2_DEBOUNCE_MASK) >> PADCFG2_DEBOUNCE_SHIFT;
*arg = BIT(v) * DEBOUNCE_PERIOD_NSEC / NSEC_PER_USEC;
return 0;
}
static int intel_config_get(struct pinctrl_dev *pctldev, unsigned int pin,
unsigned long *config)
{
struct intel_pinctrl *pctrl = pinctrl_dev_get_drvdata(pctldev);
enum pin_config_param param = pinconf_to_config_param(*config);
u32 arg = 0;
int ret;
if (!intel_pad_owned_by_host(pctrl, pin))
return -ENOTSUPP;
switch (param) {
case PIN_CONFIG_BIAS_DISABLE:
case PIN_CONFIG_BIAS_PULL_UP:
case PIN_CONFIG_BIAS_PULL_DOWN:
ret = intel_config_get_pull(pctrl, pin, param, &arg);
if (ret)
return ret;
break;
case PIN_CONFIG_INPUT_DEBOUNCE:
ret = intel_config_get_debounce(pctrl, pin, param, &arg);
if (ret)
return ret;
break;
default:
return -ENOTSUPP;
}
*config = pinconf_to_config_packed(param, arg);
return 0;
}
static int intel_config_set_pull(struct intel_pinctrl *pctrl, unsigned int pin,
unsigned long config)
{
unsigned int param = pinconf_to_config_param(config);
unsigned int arg = pinconf_to_config_argument(config);
const struct intel_community *community;
void __iomem *padcfg1;
unsigned long flags;
int ret = 0;
u32 value;
community = intel_get_community(pctrl, pin);
padcfg1 = intel_get_padcfg(pctrl, pin, PADCFG1);
raw_spin_lock_irqsave(&pctrl->lock, flags);
value = readl(padcfg1);
switch (param) {
case PIN_CONFIG_BIAS_DISABLE:
value &= ~(PADCFG1_TERM_MASK | PADCFG1_TERM_UP);
break;
case PIN_CONFIG_BIAS_PULL_UP:
value &= ~PADCFG1_TERM_MASK;
value |= PADCFG1_TERM_UP;
/* Set default strength value in case none is given */
if (arg == 1)
arg = 5000;
switch (arg) {
case 20000:
value |= PADCFG1_TERM_20K << PADCFG1_TERM_SHIFT;
break;
case 5000:
value |= PADCFG1_TERM_5K << PADCFG1_TERM_SHIFT;
break;
case 1000:
value |= PADCFG1_TERM_1K << PADCFG1_TERM_SHIFT;
break;
case 833:
value |= PADCFG1_TERM_833 << PADCFG1_TERM_SHIFT;
break;
default:
ret = -EINVAL;
}
break;
case PIN_CONFIG_BIAS_PULL_DOWN:
value &= ~(PADCFG1_TERM_UP | PADCFG1_TERM_MASK);
/* Set default strength value in case none is given */
if (arg == 1)
arg = 5000;
switch (arg) {
case 20000:
value |= PADCFG1_TERM_20K << PADCFG1_TERM_SHIFT;
break;
case 5000:
value |= PADCFG1_TERM_5K << PADCFG1_TERM_SHIFT;
break;
case 1000:
if (!(community->features & PINCTRL_FEATURE_1K_PD)) {
ret = -EINVAL;
break;
}
value |= PADCFG1_TERM_1K << PADCFG1_TERM_SHIFT;
break;
case 833:
if (!(community->features & PINCTRL_FEATURE_1K_PD)) {
ret = -EINVAL;
break;
}
value |= PADCFG1_TERM_833 << PADCFG1_TERM_SHIFT;
break;
default:
ret = -EINVAL;
}
break;
}
if (!ret)
writel(value, padcfg1);
pinctrl: intel: Use raw_spinlock for locking When running -rt kernel and GPIO interrupt happens we get following BUG: sleeping function called from invalid context at kernel/locking/rtmutex.c:931 in_atomic(): 1, irqs_disabled(): 0, pid: 530, name: irq/14-INT3452: Preemption disabled at:[<ffffffff810b4dab>] handle_edge_irq+0x1b/0x190 CPU: 0 PID: 530 Comm: irq/14-INT3452: Not tainted 4.6.2-rt5 #1060 0000000000000000 ffff88007a257d58 ffffffff812d8494 0000000000000000 ffff88017a330000 ffff88007a257d78 ffffffff81083a11 ffff88007a252430 ffff88007a252430 ffff88007a257d90 ffffffff8167ef20 000000000000001a Call Trace: [<ffffffff812d8494>] dump_stack+0x4f/0x6b [<ffffffff81083a11>] ___might_sleep+0xe1/0x160 [<ffffffff8167ef20>] rt_spin_lock+0x20/0x50 [<ffffffff81308c6d>] intel_gpio_irq_ack+0x2d/0x80 [<ffffffff810b4e0b>] handle_edge_irq+0x7b/0x190 [<ffffffff810b0d82>] generic_handle_irq+0x22/0x30 [<ffffffff81307abc>] intel_gpio_irq+0xdc/0x150 [<ffffffff810b2293>] irq_forced_thread_fn+0x23/0x70 [<ffffffff810b250b>] irq_thread+0x13b/0x1d0 [<ffffffff8167b844>] ? __schedule+0x2e4/0x5a0 [<ffffffff810b2270>] ? irq_finalize_oneshot.part.37+0xd0/0xd0 [<ffffffff810b25a0>] ? irq_thread+0x1d0/0x1d0 [<ffffffff810b23d0>] ? wake_threads_waitq+0x30/0x30 [<ffffffff8107e624>] kthread+0xd4/0xf0 [<ffffffff8167ec27>] ? _raw_spin_unlock_irq+0x17/0x40 [<ffffffff8167f592>] ret_from_fork+0x22/0x40 [<ffffffff8107e550>] ? kthread_worker_fn+0x190/0x190 The reason why this happens is because intel_gpio_irq_ack() is called with desc->lock raw_spinlock locked which cannot sleep but our normal spinlock (which is converted to rtmutex in -rt) is allowed to sleep. This causes might_sleep() to trigger. Fix this by converting the normal spinlock to a raw_spinlock. Reported-by: Kim Tatt Chuah <kim.tatt.chuah@intel.com> Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-06-16 08:25:36 +00:00
raw_spin_unlock_irqrestore(&pctrl->lock, flags);
return ret;
}
static int intel_config_set_debounce(struct intel_pinctrl *pctrl,
unsigned int pin, unsigned int debounce)
{
void __iomem *padcfg0, *padcfg2;
unsigned long flags;
u32 value0, value2;
padcfg2 = intel_get_padcfg(pctrl, pin, PADCFG2);
if (!padcfg2)
return -ENOTSUPP;
padcfg0 = intel_get_padcfg(pctrl, pin, PADCFG0);
raw_spin_lock_irqsave(&pctrl->lock, flags);
value0 = readl(padcfg0);
value2 = readl(padcfg2);
/* Disable glitch filter and debouncer */
value0 &= ~PADCFG0_PREGFRXSEL;
value2 &= ~(PADCFG2_DEBEN | PADCFG2_DEBOUNCE_MASK);
if (debounce) {
unsigned long v;
v = order_base_2(debounce * NSEC_PER_USEC / DEBOUNCE_PERIOD_NSEC);
if (v < 3 || v > 15) {
raw_spin_unlock_irqrestore(&pctrl->lock, flags);
return -EINVAL;
}
/* Enable glitch filter and debouncer */
value0 |= PADCFG0_PREGFRXSEL;
value2 |= v << PADCFG2_DEBOUNCE_SHIFT;
value2 |= PADCFG2_DEBEN;
}
writel(value0, padcfg0);
writel(value2, padcfg2);
raw_spin_unlock_irqrestore(&pctrl->lock, flags);
return 0;
}
static int intel_config_set(struct pinctrl_dev *pctldev, unsigned int pin,
unsigned long *configs, unsigned int nconfigs)
{
struct intel_pinctrl *pctrl = pinctrl_dev_get_drvdata(pctldev);
int i, ret;
if (!intel_pad_usable(pctrl, pin))
return -ENOTSUPP;
for (i = 0; i < nconfigs; i++) {
switch (pinconf_to_config_param(configs[i])) {
case PIN_CONFIG_BIAS_DISABLE:
case PIN_CONFIG_BIAS_PULL_UP:
case PIN_CONFIG_BIAS_PULL_DOWN:
ret = intel_config_set_pull(pctrl, pin, configs[i]);
if (ret)
return ret;
break;
case PIN_CONFIG_INPUT_DEBOUNCE:
ret = intel_config_set_debounce(pctrl, pin,
pinconf_to_config_argument(configs[i]));
if (ret)
return ret;
break;
default:
return -ENOTSUPP;
}
}
return 0;
}
static const struct pinconf_ops intel_pinconf_ops = {
.is_generic = true,
.pin_config_get = intel_config_get,
.pin_config_set = intel_config_set,
};
static const struct pinctrl_desc intel_pinctrl_desc = {
.pctlops = &intel_pinctrl_ops,
.pmxops = &intel_pinmux_ops,
.confops = &intel_pinconf_ops,
.owner = THIS_MODULE,
};
/**
* intel_gpio_to_pin() - Translate from GPIO offset to pin number
* @pctrl: Pinctrl structure
* @offset: GPIO offset from gpiolib
* @community: Community is filled here if not %NULL
* @padgrp: Pad group is filled here if not %NULL
*
* When coming through gpiolib irqchip, the GPIO offset is not
* automatically translated to pinctrl pin number. This function can be
* used to find out the corresponding pinctrl pin.
*/
static int intel_gpio_to_pin(struct intel_pinctrl *pctrl, unsigned int offset,
const struct intel_community **community,
const struct intel_padgroup **padgrp)
{
int i;
for (i = 0; i < pctrl->ncommunities; i++) {
const struct intel_community *comm = &pctrl->communities[i];
int j;
for (j = 0; j < comm->ngpps; j++) {
const struct intel_padgroup *pgrp = &comm->gpps[j];
if (pgrp->gpio_base == INTEL_GPIO_BASE_NOMAP)
continue;
if (offset >= pgrp->gpio_base &&
offset < pgrp->gpio_base + pgrp->size) {
int pin;
pin = pgrp->base + offset - pgrp->gpio_base;
if (community)
*community = comm;
if (padgrp)
*padgrp = pgrp;
return pin;
}
}
}
return -EINVAL;
}
pinctrl: intel: remap the pin number to gpio offset for irq enabled pin On Asus X571GT, GPIO 297 is configured as an interrupt and serves for the touchpad. The touchpad will report input events much less than expected after S3 suspend/resume, which results in extremely slow cursor movement. However, the number of interrupts observed from /proc/interrupts increases much more than expected even no touching touchpad. This is due to the value of PADCFG0 of PIN 225 for the interrupt has been changed from 0x80800102 to 0x80100102. The GPIROUTIOXAPIC is toggled on which results in the spurious interrupts. The PADCFG0 of PIN 225 is expected to be saved during suspend, but the 297 is saved instead because the gpiochip_line_is_irq() expect the GPIO offset but what's really passed to it is PIN number. In this case, the /sys/kernel/debug/pinctrl/INT3450:00/gpio-ranges shows 288: INT3450:00 GPIOS [436 - 459] PINS [216 - 239] So gpiochip_line_is_irq() returns true for GPIO offset 297, the suspend routine spuriously saves the content for PIN 297 which we expect to save for PIN 225. This commit maps the PIN number to GPIO offset first in the intel_pinctrl_should_save() to make sure the values for the specific PINs can be correctly saved and then restored. Fixes: c538b9436751 ("pinctrl: intel: Only restore pins that are used by the driver") Signed-off-by: Chris Chiu <chiu@endlessm.com> Acked-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
2019-08-16 09:38:38 +00:00
/**
* intel_pin_to_gpio() - Translate from pin number to GPIO offset
* @pctrl: Pinctrl structure
* @pin: pin number
*
* Translate the pin number of pinctrl to GPIO offset
*/
static __maybe_unused int intel_pin_to_gpio(struct intel_pinctrl *pctrl, int pin)
pinctrl: intel: remap the pin number to gpio offset for irq enabled pin On Asus X571GT, GPIO 297 is configured as an interrupt and serves for the touchpad. The touchpad will report input events much less than expected after S3 suspend/resume, which results in extremely slow cursor movement. However, the number of interrupts observed from /proc/interrupts increases much more than expected even no touching touchpad. This is due to the value of PADCFG0 of PIN 225 for the interrupt has been changed from 0x80800102 to 0x80100102. The GPIROUTIOXAPIC is toggled on which results in the spurious interrupts. The PADCFG0 of PIN 225 is expected to be saved during suspend, but the 297 is saved instead because the gpiochip_line_is_irq() expect the GPIO offset but what's really passed to it is PIN number. In this case, the /sys/kernel/debug/pinctrl/INT3450:00/gpio-ranges shows 288: INT3450:00 GPIOS [436 - 459] PINS [216 - 239] So gpiochip_line_is_irq() returns true for GPIO offset 297, the suspend routine spuriously saves the content for PIN 297 which we expect to save for PIN 225. This commit maps the PIN number to GPIO offset first in the intel_pinctrl_should_save() to make sure the values for the specific PINs can be correctly saved and then restored. Fixes: c538b9436751 ("pinctrl: intel: Only restore pins that are used by the driver") Signed-off-by: Chris Chiu <chiu@endlessm.com> Acked-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
2019-08-16 09:38:38 +00:00
{
const struct intel_community *community;
const struct intel_padgroup *padgrp;
community = intel_get_community(pctrl, pin);
if (!community)
return -EINVAL;
padgrp = intel_community_get_padgroup(community, pin);
if (!padgrp)
return -EINVAL;
return pin - padgrp->base + padgrp->gpio_base;
}
static int intel_gpio_get(struct gpio_chip *chip, unsigned int offset)
{
struct intel_pinctrl *pctrl = gpiochip_get_data(chip);
void __iomem *reg;
u32 padcfg0;
int pin;
pin = intel_gpio_to_pin(pctrl, offset, NULL, NULL);
if (pin < 0)
return -EINVAL;
reg = intel_get_padcfg(pctrl, pin, PADCFG0);
if (!reg)
return -EINVAL;
padcfg0 = readl(reg);
if (!(padcfg0 & PADCFG0_GPIOTXDIS))
return !!(padcfg0 & PADCFG0_GPIOTXSTATE);
return !!(padcfg0 & PADCFG0_GPIORXSTATE);
}
static void intel_gpio_set(struct gpio_chip *chip, unsigned int offset,
int value)
{
struct intel_pinctrl *pctrl = gpiochip_get_data(chip);
unsigned long flags;
void __iomem *reg;
u32 padcfg0;
int pin;
pin = intel_gpio_to_pin(pctrl, offset, NULL, NULL);
if (pin < 0)
return;
reg = intel_get_padcfg(pctrl, pin, PADCFG0);
if (!reg)
return;
raw_spin_lock_irqsave(&pctrl->lock, flags);
padcfg0 = readl(reg);
if (value)
padcfg0 |= PADCFG0_GPIOTXSTATE;
else
padcfg0 &= ~PADCFG0_GPIOTXSTATE;
writel(padcfg0, reg);
raw_spin_unlock_irqrestore(&pctrl->lock, flags);
}
static int intel_gpio_get_direction(struct gpio_chip *chip, unsigned int offset)
{
struct intel_pinctrl *pctrl = gpiochip_get_data(chip);
unsigned long flags;
void __iomem *reg;
u32 padcfg0;
int pin;
pin = intel_gpio_to_pin(pctrl, offset, NULL, NULL);
if (pin < 0)
return -EINVAL;
reg = intel_get_padcfg(pctrl, pin, PADCFG0);
if (!reg)
return -EINVAL;
raw_spin_lock_irqsave(&pctrl->lock, flags);
padcfg0 = readl(reg);
raw_spin_unlock_irqrestore(&pctrl->lock, flags);
if (padcfg0 & PADCFG0_PMODE_MASK)
return -EINVAL;
if (padcfg0 & PADCFG0_GPIOTXDIS)
return GPIO_LINE_DIRECTION_IN;
return GPIO_LINE_DIRECTION_OUT;
}
static int intel_gpio_direction_input(struct gpio_chip *chip, unsigned int offset)
{
return pinctrl_gpio_direction_input(chip->base + offset);
}
static int intel_gpio_direction_output(struct gpio_chip *chip, unsigned int offset,
int value)
{
intel_gpio_set(chip, offset, value);
return pinctrl_gpio_direction_output(chip->base + offset);
}
static const struct gpio_chip intel_gpio_chip = {
.owner = THIS_MODULE,
.request = gpiochip_generic_request,
.free = gpiochip_generic_free,
.get_direction = intel_gpio_get_direction,
.direction_input = intel_gpio_direction_input,
.direction_output = intel_gpio_direction_output,
.get = intel_gpio_get,
.set = intel_gpio_set,
.set_config = gpiochip_generic_config,
};
static void intel_gpio_irq_ack(struct irq_data *d)
{
struct gpio_chip *gc = irq_data_get_irq_chip_data(d);
struct intel_pinctrl *pctrl = gpiochip_get_data(gc);
const struct intel_community *community;
const struct intel_padgroup *padgrp;
int pin;
pin = intel_gpio_to_pin(pctrl, irqd_to_hwirq(d), &community, &padgrp);
if (pin >= 0) {
unsigned int gpp, gpp_offset, is_offset;
gpp = padgrp->reg_num;
gpp_offset = padgroup_offset(padgrp, pin);
is_offset = community->is_offset + gpp * 4;
raw_spin_lock(&pctrl->lock);
writel(BIT(gpp_offset), community->regs + is_offset);
raw_spin_unlock(&pctrl->lock);
}
}
static void intel_gpio_irq_mask_unmask(struct irq_data *d, bool mask)
{
struct gpio_chip *gc = irq_data_get_irq_chip_data(d);
struct intel_pinctrl *pctrl = gpiochip_get_data(gc);
const struct intel_community *community;
const struct intel_padgroup *padgrp;
int pin;
pin = intel_gpio_to_pin(pctrl, irqd_to_hwirq(d), &community, &padgrp);
if (pin >= 0) {
unsigned int gpp, gpp_offset;
unsigned long flags;
void __iomem *reg, *is;
u32 value;
gpp = padgrp->reg_num;
gpp_offset = padgroup_offset(padgrp, pin);
reg = community->regs + community->ie_offset + gpp * 4;
is = community->regs + community->is_offset + gpp * 4;
raw_spin_lock_irqsave(&pctrl->lock, flags);
/* Clear interrupt status first to avoid unexpected interrupt */
writel(BIT(gpp_offset), is);
value = readl(reg);
if (mask)
value &= ~BIT(gpp_offset);
else
value |= BIT(gpp_offset);
writel(value, reg);
raw_spin_unlock_irqrestore(&pctrl->lock, flags);
}
}
static void intel_gpio_irq_mask(struct irq_data *d)
{
intel_gpio_irq_mask_unmask(d, true);
}
static void intel_gpio_irq_unmask(struct irq_data *d)
{
intel_gpio_irq_mask_unmask(d, false);
}
static int intel_gpio_irq_type(struct irq_data *d, unsigned int type)
{
struct gpio_chip *gc = irq_data_get_irq_chip_data(d);
struct intel_pinctrl *pctrl = gpiochip_get_data(gc);
unsigned int pin = intel_gpio_to_pin(pctrl, irqd_to_hwirq(d), NULL, NULL);
unsigned long flags;
void __iomem *reg;
u32 value;
reg = intel_get_padcfg(pctrl, pin, PADCFG0);
if (!reg)
return -EINVAL;
/*
* If the pin is in ACPI mode it is still usable as a GPIO but it
* cannot be used as IRQ because GPI_IS status bit will not be
* updated by the host controller hardware.
*/
if (intel_pad_acpi_mode(pctrl, pin)) {
dev_warn(pctrl->dev, "pin %u cannot be used as IRQ\n", pin);
return -EPERM;
}
pinctrl: intel: Use raw_spinlock for locking When running -rt kernel and GPIO interrupt happens we get following BUG: sleeping function called from invalid context at kernel/locking/rtmutex.c:931 in_atomic(): 1, irqs_disabled(): 0, pid: 530, name: irq/14-INT3452: Preemption disabled at:[<ffffffff810b4dab>] handle_edge_irq+0x1b/0x190 CPU: 0 PID: 530 Comm: irq/14-INT3452: Not tainted 4.6.2-rt5 #1060 0000000000000000 ffff88007a257d58 ffffffff812d8494 0000000000000000 ffff88017a330000 ffff88007a257d78 ffffffff81083a11 ffff88007a252430 ffff88007a252430 ffff88007a257d90 ffffffff8167ef20 000000000000001a Call Trace: [<ffffffff812d8494>] dump_stack+0x4f/0x6b [<ffffffff81083a11>] ___might_sleep+0xe1/0x160 [<ffffffff8167ef20>] rt_spin_lock+0x20/0x50 [<ffffffff81308c6d>] intel_gpio_irq_ack+0x2d/0x80 [<ffffffff810b4e0b>] handle_edge_irq+0x7b/0x190 [<ffffffff810b0d82>] generic_handle_irq+0x22/0x30 [<ffffffff81307abc>] intel_gpio_irq+0xdc/0x150 [<ffffffff810b2293>] irq_forced_thread_fn+0x23/0x70 [<ffffffff810b250b>] irq_thread+0x13b/0x1d0 [<ffffffff8167b844>] ? __schedule+0x2e4/0x5a0 [<ffffffff810b2270>] ? irq_finalize_oneshot.part.37+0xd0/0xd0 [<ffffffff810b25a0>] ? irq_thread+0x1d0/0x1d0 [<ffffffff810b23d0>] ? wake_threads_waitq+0x30/0x30 [<ffffffff8107e624>] kthread+0xd4/0xf0 [<ffffffff8167ec27>] ? _raw_spin_unlock_irq+0x17/0x40 [<ffffffff8167f592>] ret_from_fork+0x22/0x40 [<ffffffff8107e550>] ? kthread_worker_fn+0x190/0x190 The reason why this happens is because intel_gpio_irq_ack() is called with desc->lock raw_spinlock locked which cannot sleep but our normal spinlock (which is converted to rtmutex in -rt) is allowed to sleep. This causes might_sleep() to trigger. Fix this by converting the normal spinlock to a raw_spinlock. Reported-by: Kim Tatt Chuah <kim.tatt.chuah@intel.com> Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-06-16 08:25:36 +00:00
raw_spin_lock_irqsave(&pctrl->lock, flags);
intel_gpio_set_gpio_mode(reg);
/* Disable TX buffer and enable RX (this will be input) */
__intel_gpio_set_direction(reg, true);
value = readl(reg);
value &= ~(PADCFG0_RXEVCFG_MASK | PADCFG0_RXINV);
if ((type & IRQ_TYPE_EDGE_BOTH) == IRQ_TYPE_EDGE_BOTH) {
value |= PADCFG0_RXEVCFG_EDGE_BOTH << PADCFG0_RXEVCFG_SHIFT;
} else if (type & IRQ_TYPE_EDGE_FALLING) {
value |= PADCFG0_RXEVCFG_EDGE << PADCFG0_RXEVCFG_SHIFT;
value |= PADCFG0_RXINV;
} else if (type & IRQ_TYPE_EDGE_RISING) {
value |= PADCFG0_RXEVCFG_EDGE << PADCFG0_RXEVCFG_SHIFT;
} else if (type & IRQ_TYPE_LEVEL_MASK) {
if (type & IRQ_TYPE_LEVEL_LOW)
value |= PADCFG0_RXINV;
} else {
value |= PADCFG0_RXEVCFG_DISABLED << PADCFG0_RXEVCFG_SHIFT;
}
writel(value, reg);
if (type & IRQ_TYPE_EDGE_BOTH)
irq_set_handler_locked(d, handle_edge_irq);
else if (type & IRQ_TYPE_LEVEL_MASK)
irq_set_handler_locked(d, handle_level_irq);
pinctrl: intel: Use raw_spinlock for locking When running -rt kernel and GPIO interrupt happens we get following BUG: sleeping function called from invalid context at kernel/locking/rtmutex.c:931 in_atomic(): 1, irqs_disabled(): 0, pid: 530, name: irq/14-INT3452: Preemption disabled at:[<ffffffff810b4dab>] handle_edge_irq+0x1b/0x190 CPU: 0 PID: 530 Comm: irq/14-INT3452: Not tainted 4.6.2-rt5 #1060 0000000000000000 ffff88007a257d58 ffffffff812d8494 0000000000000000 ffff88017a330000 ffff88007a257d78 ffffffff81083a11 ffff88007a252430 ffff88007a252430 ffff88007a257d90 ffffffff8167ef20 000000000000001a Call Trace: [<ffffffff812d8494>] dump_stack+0x4f/0x6b [<ffffffff81083a11>] ___might_sleep+0xe1/0x160 [<ffffffff8167ef20>] rt_spin_lock+0x20/0x50 [<ffffffff81308c6d>] intel_gpio_irq_ack+0x2d/0x80 [<ffffffff810b4e0b>] handle_edge_irq+0x7b/0x190 [<ffffffff810b0d82>] generic_handle_irq+0x22/0x30 [<ffffffff81307abc>] intel_gpio_irq+0xdc/0x150 [<ffffffff810b2293>] irq_forced_thread_fn+0x23/0x70 [<ffffffff810b250b>] irq_thread+0x13b/0x1d0 [<ffffffff8167b844>] ? __schedule+0x2e4/0x5a0 [<ffffffff810b2270>] ? irq_finalize_oneshot.part.37+0xd0/0xd0 [<ffffffff810b25a0>] ? irq_thread+0x1d0/0x1d0 [<ffffffff810b23d0>] ? wake_threads_waitq+0x30/0x30 [<ffffffff8107e624>] kthread+0xd4/0xf0 [<ffffffff8167ec27>] ? _raw_spin_unlock_irq+0x17/0x40 [<ffffffff8167f592>] ret_from_fork+0x22/0x40 [<ffffffff8107e550>] ? kthread_worker_fn+0x190/0x190 The reason why this happens is because intel_gpio_irq_ack() is called with desc->lock raw_spinlock locked which cannot sleep but our normal spinlock (which is converted to rtmutex in -rt) is allowed to sleep. This causes might_sleep() to trigger. Fix this by converting the normal spinlock to a raw_spinlock. Reported-by: Kim Tatt Chuah <kim.tatt.chuah@intel.com> Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-06-16 08:25:36 +00:00
raw_spin_unlock_irqrestore(&pctrl->lock, flags);
return 0;
}
static int intel_gpio_irq_wake(struct irq_data *d, unsigned int on)
{
struct gpio_chip *gc = irq_data_get_irq_chip_data(d);
struct intel_pinctrl *pctrl = gpiochip_get_data(gc);
unsigned int pin = intel_gpio_to_pin(pctrl, irqd_to_hwirq(d), NULL, NULL);
if (on)
enable_irq_wake(pctrl->irq);
else
disable_irq_wake(pctrl->irq);
dev_dbg(pctrl->dev, "%sable wake for pin %u\n", on ? "en" : "dis", pin);
return 0;
}
static int intel_gpio_community_irq_handler(struct intel_pinctrl *pctrl,
const struct intel_community *community)
{
struct gpio_chip *gc = &pctrl->chip;
unsigned int gpp;
int ret = 0;
for (gpp = 0; gpp < community->ngpps; gpp++) {
const struct intel_padgroup *padgrp = &community->gpps[gpp];
unsigned long pending, enabled, gpp_offset;
unsigned long flags;
raw_spin_lock_irqsave(&pctrl->lock, flags);
pending = readl(community->regs + community->is_offset +
padgrp->reg_num * 4);
enabled = readl(community->regs + community->ie_offset +
padgrp->reg_num * 4);
raw_spin_unlock_irqrestore(&pctrl->lock, flags);
/* Only interrupts that are enabled */
pending &= enabled;
for_each_set_bit(gpp_offset, &pending, padgrp->size) {
unsigned int irq;
irq = irq_find_mapping(gc->irq.domain,
padgrp->gpio_base + gpp_offset);
generic_handle_irq(irq);
}
ret += pending ? 1 : 0;
}
return ret;
}
static irqreturn_t intel_gpio_irq(int irq, void *data)
{
const struct intel_community *community;
struct intel_pinctrl *pctrl = data;
unsigned int i;
int ret = 0;
/* Need to check all communities for pending interrupts */
for (i = 0; i < pctrl->ncommunities; i++) {
community = &pctrl->communities[i];
ret += intel_gpio_community_irq_handler(pctrl, community);
}
return IRQ_RETVAL(ret);
}
static int intel_gpio_add_community_ranges(struct intel_pinctrl *pctrl,
const struct intel_community *community)
{
int ret = 0, i;
for (i = 0; i < community->ngpps; i++) {
const struct intel_padgroup *gpp = &community->gpps[i];
if (gpp->gpio_base == INTEL_GPIO_BASE_NOMAP)
continue;
ret = gpiochip_add_pin_range(&pctrl->chip, dev_name(pctrl->dev),
gpp->gpio_base, gpp->base,
gpp->size);
if (ret)
return ret;
}
return ret;
}
static int intel_gpio_add_pin_ranges(struct gpio_chip *gc)
{
struct intel_pinctrl *pctrl = gpiochip_get_data(gc);
int ret, i;
for (i = 0; i < pctrl->ncommunities; i++) {
struct intel_community *community = &pctrl->communities[i];
ret = intel_gpio_add_community_ranges(pctrl, community);
if (ret) {
dev_err(pctrl->dev, "failed to add GPIO pin range\n");
return ret;
}
}
return 0;
}
static unsigned int intel_gpio_ngpio(const struct intel_pinctrl *pctrl)
{
const struct intel_community *community;
unsigned int ngpio = 0;
int i, j;
for (i = 0; i < pctrl->ncommunities; i++) {
community = &pctrl->communities[i];
for (j = 0; j < community->ngpps; j++) {
const struct intel_padgroup *gpp = &community->gpps[j];
if (gpp->gpio_base == INTEL_GPIO_BASE_NOMAP)
continue;
if (gpp->gpio_base + gpp->size > ngpio)
ngpio = gpp->gpio_base + gpp->size;
}
}
return ngpio;
}
static int intel_gpio_probe(struct intel_pinctrl *pctrl, int irq)
{
int ret;
struct gpio_irq_chip *girq;
pctrl->chip = intel_gpio_chip;
/* Setup GPIO chip */
pctrl->chip.ngpio = intel_gpio_ngpio(pctrl);
pctrl->chip.label = dev_name(pctrl->dev);
gpio: change member .dev to .parent The name .dev in a struct is normally reserved for a struct device that is let us say a superclass to the thing described by the struct. struct gpio_chip stands out by confusingly using a struct device *dev to point to the parent device (such as a platform_device) that represents the hardware. As we want to give gpio_chip:s real devices, this is not working. We need to rename this member to parent. This was done by two coccinelle scripts, I guess it is possible to combine them into one, but I don't know such stuff. They look like this: @@ struct gpio_chip *var; @@ -var->dev +var->parent and: @@ struct gpio_chip var; @@ -var.dev +var.parent and: @@ struct bgpio_chip *var; @@ -var->gc.dev +var->gc.parent Plus a few instances of bgpio that I couldn't figure out how to teach Coccinelle to rewrite. This patch hits all over the place, but I *strongly* prefer this solution to any piecemal approaches that just exercise patch mechanics all over the place. It mainly hits drivers/gpio and drivers/pinctrl which is my own backyard anyway. Cc: Haavard Skinnemoen <hskinnemoen@gmail.com> Cc: Rafał Miłecki <zajec5@gmail.com> Cc: Richard Purdie <rpurdie@rpsys.net> Cc: Mauro Carvalho Chehab <mchehab@osg.samsung.com> Cc: Alek Du <alek.du@intel.com> Cc: Jaroslav Kysela <perex@perex.cz> Cc: Takashi Iwai <tiwai@suse.com> Acked-by: Dmitry Torokhov <dmitry.torokhov@gmail.com> Acked-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Acked-by: Lee Jones <lee.jones@linaro.org> Acked-by: Jiri Kosina <jkosina@suse.cz> Acked-by: Hans-Christian Egtvedt <egtvedt@samfundet.no> Acked-by: Jacek Anaszewski <j.anaszewski@samsung.com> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2015-11-04 08:56:26 +00:00
pctrl->chip.parent = pctrl->dev;
pctrl->chip.base = -1;
pctrl->chip.add_pin_ranges = intel_gpio_add_pin_ranges;
pctrl->irq = irq;
/* Setup IRQ chip */
pctrl->irqchip.name = dev_name(pctrl->dev);
pctrl->irqchip.irq_ack = intel_gpio_irq_ack;
pctrl->irqchip.irq_mask = intel_gpio_irq_mask;
pctrl->irqchip.irq_unmask = intel_gpio_irq_unmask;
pctrl->irqchip.irq_set_type = intel_gpio_irq_type;
pctrl->irqchip.irq_set_wake = intel_gpio_irq_wake;
pctrl->irqchip.flags = IRQCHIP_MASK_ON_SUSPEND;
/*
* On some platforms several GPIO controllers share the same interrupt
* line.
*/
pinctrl: intel: Prevent force threading of the interrupt handler The pinctrl-intel needs to use request_irq() instead of chained interrupt handling because it shares the interrupt with multiple GPIO host controllers found on Intel CPUs. In -rt all such interrupts are forced to run in thread context which triggers following warning: WARNING: CPU: 0 PID: 530 at kernel/irq/handle.c:151 handle_irq_event_percpu+0x23d/0x240 irq 348 handler irq_default_primary_handler+0x0/0x10 enabled interrupts Modules linked in: CPU: 0 PID: 530 Comm: irq/14-INT3452: Not tainted 4.6.2-rt5 #1060 0000000000000000 ffff88007a257c98 ffffffff812d8494 ffff88007a257ce8 0000000000000000 ffff88007a257cd8 ffffffff8105e554 000000977a257d90 ffff88007a37a380 000000000000015c 0000000000000002 0000000000000000 Call Trace: [<ffffffff812d8494>] dump_stack+0x4f/0x6b [<ffffffff8105e554>] __warn+0xe4/0x100 [<ffffffff8105e5bf>] warn_slowpath_fmt+0x4f/0x60 [<ffffffff810b18f0>] ? __synchronize_hardirq+0x60/0x60 [<ffffffff810b17fd>] handle_irq_event_percpu+0x23d/0x240 [<ffffffff810b1862>] handle_irq_event+0x62/0x90 [<ffffffff810b4e1f>] handle_edge_irq+0x8f/0x190 [<ffffffff810b0d82>] generic_handle_irq+0x22/0x30 [<ffffffff81307abc>] intel_gpio_irq+0xdc/0x150 [<ffffffff810b2293>] irq_forced_thread_fn+0x23/0x70 [<ffffffff810b250b>] irq_thread+0x13b/0x1d0 [<ffffffff8167b844>] ? __schedule+0x2e4/0x5a0 [<ffffffff810b2270>] ? irq_finalize_oneshot.part.37+0xd0/0xd0 [<ffffffff810b25a0>] ? irq_thread+0x1d0/0x1d0 [<ffffffff810b23d0>] ? wake_threads_waitq+0x30/0x30 [<ffffffff8107e624>] kthread+0xd4/0xf0 [<ffffffff8167ec27>] ? _raw_spin_unlock_irq+0x17/0x40 [<ffffffff8167f592>] ret_from_fork+0x22/0x40 [<ffffffff8107e550>] ? kthread_worker_fn+0x190/0x190 The handle_irq_event_* functions (and I suppose generic_handle_irq()) is expected to be called with interrupts disabled and they rightfully complain here because we run in thread context with interrupts enabled. Fix this by adding IRQF_NO_THREAD flag when the master interrupt is requested. This prevents forced threading of the interrupt used by the GPIO host controllers. Reported-by: Kim Tatt Chuah <kim.tatt.chuah@intel.com> Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-06-16 08:25:37 +00:00
ret = devm_request_irq(pctrl->dev, irq, intel_gpio_irq,
IRQF_SHARED | IRQF_NO_THREAD,
dev_name(pctrl->dev), pctrl);
if (ret) {
dev_err(pctrl->dev, "failed to request interrupt\n");
return ret;
}
girq = &pctrl->chip.irq;
girq->chip = &pctrl->irqchip;
/* This will let us handle the IRQ in the driver */
girq->parent_handler = NULL;
girq->num_parents = 0;
girq->default_type = IRQ_TYPE_NONE;
girq->handler = handle_bad_irq;
ret = devm_gpiochip_add_data(pctrl->dev, &pctrl->chip, pctrl);
if (ret) {
dev_err(pctrl->dev, "failed to register gpiochip\n");
return ret;
}
return 0;
}
static int intel_pinctrl_add_padgroups(struct intel_pinctrl *pctrl,
struct intel_community *community)
{
struct intel_padgroup *gpps;
unsigned int npins = community->npins;
unsigned int padown_num = 0;
size_t ngpps, i;
if (community->gpps)
ngpps = community->ngpps;
else
ngpps = DIV_ROUND_UP(community->npins, community->gpp_size);
gpps = devm_kcalloc(pctrl->dev, ngpps, sizeof(*gpps), GFP_KERNEL);
if (!gpps)
return -ENOMEM;
for (i = 0; i < ngpps; i++) {
if (community->gpps) {
gpps[i] = community->gpps[i];
} else {
unsigned int gpp_size = community->gpp_size;
gpps[i].reg_num = i;
gpps[i].base = community->pin_base + i * gpp_size;
gpps[i].size = min(gpp_size, npins);
npins -= gpps[i].size;
}
if (gpps[i].size > 32)
return -EINVAL;
/* Special treatment for GPIO base */
switch (gpps[i].gpio_base) {
case INTEL_GPIO_BASE_MATCH:
gpps[i].gpio_base = gpps[i].base;
break;
case INTEL_GPIO_BASE_ZERO:
gpps[i].gpio_base = 0;
break;
case INTEL_GPIO_BASE_NOMAP:
default:
break;
}
gpps[i].padown_num = padown_num;
/*
* In older hardware the number of padown registers per
* group is fixed regardless of the group size.
*/
if (community->gpp_num_padown_regs)
padown_num += community->gpp_num_padown_regs;
else
padown_num += DIV_ROUND_UP(gpps[i].size * 4, 32);
}
community->ngpps = ngpps;
community->gpps = gpps;
return 0;
}
static int intel_pinctrl_pm_init(struct intel_pinctrl *pctrl)
{
#ifdef CONFIG_PM_SLEEP
const struct intel_pinctrl_soc_data *soc = pctrl->soc;
struct intel_community_context *communities;
struct intel_pad_context *pads;
int i;
pads = devm_kcalloc(pctrl->dev, soc->npins, sizeof(*pads), GFP_KERNEL);
if (!pads)
return -ENOMEM;
communities = devm_kcalloc(pctrl->dev, pctrl->ncommunities,
sizeof(*communities), GFP_KERNEL);
if (!communities)
return -ENOMEM;
for (i = 0; i < pctrl->ncommunities; i++) {
struct intel_community *community = &pctrl->communities[i];
u32 *intmask, *hostown;
intmask = devm_kcalloc(pctrl->dev, community->ngpps,
sizeof(*intmask), GFP_KERNEL);
if (!intmask)
return -ENOMEM;
communities[i].intmask = intmask;
hostown = devm_kcalloc(pctrl->dev, community->ngpps,
sizeof(*hostown), GFP_KERNEL);
if (!hostown)
return -ENOMEM;
communities[i].hostown = hostown;
}
pctrl->context.pads = pads;
pctrl->context.communities = communities;
#endif
return 0;
}
static int intel_pinctrl_probe(struct platform_device *pdev,
const struct intel_pinctrl_soc_data *soc_data)
{
struct intel_pinctrl *pctrl;
int i, ret, irq;
pctrl = devm_kzalloc(&pdev->dev, sizeof(*pctrl), GFP_KERNEL);
if (!pctrl)
return -ENOMEM;
pctrl->dev = &pdev->dev;
pctrl->soc = soc_data;
pinctrl: intel: Use raw_spinlock for locking When running -rt kernel and GPIO interrupt happens we get following BUG: sleeping function called from invalid context at kernel/locking/rtmutex.c:931 in_atomic(): 1, irqs_disabled(): 0, pid: 530, name: irq/14-INT3452: Preemption disabled at:[<ffffffff810b4dab>] handle_edge_irq+0x1b/0x190 CPU: 0 PID: 530 Comm: irq/14-INT3452: Not tainted 4.6.2-rt5 #1060 0000000000000000 ffff88007a257d58 ffffffff812d8494 0000000000000000 ffff88017a330000 ffff88007a257d78 ffffffff81083a11 ffff88007a252430 ffff88007a252430 ffff88007a257d90 ffffffff8167ef20 000000000000001a Call Trace: [<ffffffff812d8494>] dump_stack+0x4f/0x6b [<ffffffff81083a11>] ___might_sleep+0xe1/0x160 [<ffffffff8167ef20>] rt_spin_lock+0x20/0x50 [<ffffffff81308c6d>] intel_gpio_irq_ack+0x2d/0x80 [<ffffffff810b4e0b>] handle_edge_irq+0x7b/0x190 [<ffffffff810b0d82>] generic_handle_irq+0x22/0x30 [<ffffffff81307abc>] intel_gpio_irq+0xdc/0x150 [<ffffffff810b2293>] irq_forced_thread_fn+0x23/0x70 [<ffffffff810b250b>] irq_thread+0x13b/0x1d0 [<ffffffff8167b844>] ? __schedule+0x2e4/0x5a0 [<ffffffff810b2270>] ? irq_finalize_oneshot.part.37+0xd0/0xd0 [<ffffffff810b25a0>] ? irq_thread+0x1d0/0x1d0 [<ffffffff810b23d0>] ? wake_threads_waitq+0x30/0x30 [<ffffffff8107e624>] kthread+0xd4/0xf0 [<ffffffff8167ec27>] ? _raw_spin_unlock_irq+0x17/0x40 [<ffffffff8167f592>] ret_from_fork+0x22/0x40 [<ffffffff8107e550>] ? kthread_worker_fn+0x190/0x190 The reason why this happens is because intel_gpio_irq_ack() is called with desc->lock raw_spinlock locked which cannot sleep but our normal spinlock (which is converted to rtmutex in -rt) is allowed to sleep. This causes might_sleep() to trigger. Fix this by converting the normal spinlock to a raw_spinlock. Reported-by: Kim Tatt Chuah <kim.tatt.chuah@intel.com> Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-06-16 08:25:36 +00:00
raw_spin_lock_init(&pctrl->lock);
/*
* Make a copy of the communities which we can use to hold pointers
* to the registers.
*/
pctrl->ncommunities = pctrl->soc->ncommunities;
pctrl->communities = devm_kcalloc(&pdev->dev, pctrl->ncommunities,
sizeof(*pctrl->communities), GFP_KERNEL);
if (!pctrl->communities)
return -ENOMEM;
for (i = 0; i < pctrl->ncommunities; i++) {
struct intel_community *community = &pctrl->communities[i];
void __iomem *regs;
u32 padbar;
*community = pctrl->soc->communities[i];
regs = devm_platform_ioremap_resource(pdev, community->barno);
if (IS_ERR(regs))
return PTR_ERR(regs);
/*
* Determine community features based on the revision if
* not specified already.
*/
if (!community->features) {
u32 rev;
rev = (readl(regs + REVID) & REVID_MASK) >> REVID_SHIFT;
if (rev >= 0x94) {
community->features |= PINCTRL_FEATURE_DEBOUNCE;
community->features |= PINCTRL_FEATURE_1K_PD;
}
}
/* Read offset of the pad configuration registers */
padbar = readl(regs + PADBAR);
community->regs = regs;
community->pad_regs = regs + padbar;
ret = intel_pinctrl_add_padgroups(pctrl, community);
if (ret)
return ret;
}
irq = platform_get_irq(pdev, 0);
if (irq < 0)
return irq;
ret = intel_pinctrl_pm_init(pctrl);
if (ret)
return ret;
pctrl->pctldesc = intel_pinctrl_desc;
pctrl->pctldesc.name = dev_name(&pdev->dev);
pctrl->pctldesc.pins = pctrl->soc->pins;
pctrl->pctldesc.npins = pctrl->soc->npins;
pctrl->pctldev = devm_pinctrl_register(&pdev->dev, &pctrl->pctldesc,
pctrl);
if (IS_ERR(pctrl->pctldev)) {
dev_err(&pdev->dev, "failed to register pinctrl driver\n");
return PTR_ERR(pctrl->pctldev);
}
ret = intel_gpio_probe(pctrl, irq);
if (ret)
return ret;
platform_set_drvdata(pdev, pctrl);
return 0;
}
int intel_pinctrl_probe_by_hid(struct platform_device *pdev)
{
const struct intel_pinctrl_soc_data *data;
data = device_get_match_data(&pdev->dev);
if (!data)
return -ENODATA;
return intel_pinctrl_probe(pdev, data);
}
EXPORT_SYMBOL_GPL(intel_pinctrl_probe_by_hid);
int intel_pinctrl_probe_by_uid(struct platform_device *pdev)
{
const struct intel_pinctrl_soc_data *data;
data = intel_pinctrl_get_soc_data(pdev);
if (IS_ERR(data))
return PTR_ERR(data);
return intel_pinctrl_probe(pdev, data);
}
EXPORT_SYMBOL_GPL(intel_pinctrl_probe_by_uid);
const struct intel_pinctrl_soc_data *intel_pinctrl_get_soc_data(struct platform_device *pdev)
{
const struct intel_pinctrl_soc_data *data = NULL;
const struct intel_pinctrl_soc_data **table;
struct acpi_device *adev;
unsigned int i;
adev = ACPI_COMPANION(&pdev->dev);
if (adev) {
const void *match = device_get_match_data(&pdev->dev);
table = (const struct intel_pinctrl_soc_data **)match;
for (i = 0; table[i]; i++) {
if (!strcmp(adev->pnp.unique_id, table[i]->uid)) {
data = table[i];
break;
}
}
} else {
const struct platform_device_id *id;
id = platform_get_device_id(pdev);
if (!id)
return ERR_PTR(-ENODEV);
table = (const struct intel_pinctrl_soc_data **)id->driver_data;
data = table[pdev->id];
}
return data ?: ERR_PTR(-ENODATA);
}
EXPORT_SYMBOL_GPL(intel_pinctrl_get_soc_data);
#ifdef CONFIG_PM_SLEEP
static bool intel_pinctrl_should_save(struct intel_pinctrl *pctrl, unsigned int pin)
{
const struct pin_desc *pd = pin_desc_get(pctrl->pctldev, pin);
if (!pd || !intel_pad_usable(pctrl, pin))
return false;
/*
* Only restore the pin if it is actually in use by the kernel (or
* by userspace). It is possible that some pins are used by the
* BIOS during resume and those are not always locked down so leave
* them alone.
*/
if (pd->mux_owner || pd->gpio_owner ||
pinctrl: intel: remap the pin number to gpio offset for irq enabled pin On Asus X571GT, GPIO 297 is configured as an interrupt and serves for the touchpad. The touchpad will report input events much less than expected after S3 suspend/resume, which results in extremely slow cursor movement. However, the number of interrupts observed from /proc/interrupts increases much more than expected even no touching touchpad. This is due to the value of PADCFG0 of PIN 225 for the interrupt has been changed from 0x80800102 to 0x80100102. The GPIROUTIOXAPIC is toggled on which results in the spurious interrupts. The PADCFG0 of PIN 225 is expected to be saved during suspend, but the 297 is saved instead because the gpiochip_line_is_irq() expect the GPIO offset but what's really passed to it is PIN number. In this case, the /sys/kernel/debug/pinctrl/INT3450:00/gpio-ranges shows 288: INT3450:00 GPIOS [436 - 459] PINS [216 - 239] So gpiochip_line_is_irq() returns true for GPIO offset 297, the suspend routine spuriously saves the content for PIN 297 which we expect to save for PIN 225. This commit maps the PIN number to GPIO offset first in the intel_pinctrl_should_save() to make sure the values for the specific PINs can be correctly saved and then restored. Fixes: c538b9436751 ("pinctrl: intel: Only restore pins that are used by the driver") Signed-off-by: Chris Chiu <chiu@endlessm.com> Acked-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
2019-08-16 09:38:38 +00:00
gpiochip_line_is_irq(&pctrl->chip, intel_pin_to_gpio(pctrl, pin)))
return true;
return false;
}
int intel_pinctrl_suspend_noirq(struct device *dev)
{
struct intel_pinctrl *pctrl = dev_get_drvdata(dev);
struct intel_community_context *communities;
struct intel_pad_context *pads;
int i;
pads = pctrl->context.pads;
for (i = 0; i < pctrl->soc->npins; i++) {
const struct pinctrl_pin_desc *desc = &pctrl->soc->pins[i];
void __iomem *padcfg;
u32 val;
if (!intel_pinctrl_should_save(pctrl, desc->number))
continue;
val = readl(intel_get_padcfg(pctrl, desc->number, PADCFG0));
pads[i].padcfg0 = val & ~PADCFG0_GPIORXSTATE;
val = readl(intel_get_padcfg(pctrl, desc->number, PADCFG1));
pads[i].padcfg1 = val;
padcfg = intel_get_padcfg(pctrl, desc->number, PADCFG2);
if (padcfg)
pads[i].padcfg2 = readl(padcfg);
}
communities = pctrl->context.communities;
for (i = 0; i < pctrl->ncommunities; i++) {
struct intel_community *community = &pctrl->communities[i];
void __iomem *base;
unsigned int gpp;
base = community->regs + community->ie_offset;
for (gpp = 0; gpp < community->ngpps; gpp++)
communities[i].intmask[gpp] = readl(base + gpp * 4);
base = community->regs + community->hostown_offset;
for (gpp = 0; gpp < community->ngpps; gpp++)
communities[i].hostown[gpp] = readl(base + gpp * 4);
}
return 0;
}
EXPORT_SYMBOL_GPL(intel_pinctrl_suspend_noirq);
static void intel_gpio_irq_init(struct intel_pinctrl *pctrl)
{
size_t i;
for (i = 0; i < pctrl->ncommunities; i++) {
const struct intel_community *community;
void __iomem *base;
unsigned int gpp;
community = &pctrl->communities[i];
base = community->regs;
for (gpp = 0; gpp < community->ngpps; gpp++) {
/* Mask and clear all interrupts */
writel(0, base + community->ie_offset + gpp * 4);
writel(0xffff, base + community->is_offset + gpp * 4);
}
}
}
static bool intel_gpio_update_reg(void __iomem *reg, u32 mask, u32 value)
{
u32 curr, updated;
curr = readl(reg);
updated = (curr & ~mask) | (value & mask);
if (curr == updated)
return false;
writel(updated, reg);
return true;
}
static void intel_restore_hostown(struct intel_pinctrl *pctrl, unsigned int c,
void __iomem *base, unsigned int gpp, u32 saved)
{
const struct intel_community *community = &pctrl->communities[c];
const struct intel_padgroup *padgrp = &community->gpps[gpp];
struct device *dev = pctrl->dev;
const char *dummy;
u32 requested = 0;
unsigned int i;
if (padgrp->gpio_base == INTEL_GPIO_BASE_NOMAP)
return;
for_each_requested_gpio_in_range(&pctrl->chip, i, padgrp->gpio_base, padgrp->size, dummy)
requested |= BIT(i);
if (!intel_gpio_update_reg(base + gpp * 4, requested, saved))
return;
dev_dbg(dev, "restored hostown %u/%u %#08x\n", c, gpp, readl(base + gpp * 4));
}
static void intel_restore_intmask(struct intel_pinctrl *pctrl, unsigned int c,
void __iomem *base, unsigned int gpp, u32 saved)
{
struct device *dev = pctrl->dev;
if (!intel_gpio_update_reg(base + gpp * 4, ~0U, saved))
return;
dev_dbg(dev, "restored mask %u/%u %#08x\n", c, gpp, readl(base + gpp * 4));
}
static void intel_restore_padcfg(struct intel_pinctrl *pctrl, unsigned int pin,
unsigned int reg, u32 saved)
{
u32 mask = (reg == PADCFG0) ? PADCFG0_GPIORXSTATE : 0;
unsigned int n = reg / sizeof(u32);
struct device *dev = pctrl->dev;
void __iomem *padcfg;
padcfg = intel_get_padcfg(pctrl, pin, reg);
if (!padcfg)
return;
if (!intel_gpio_update_reg(padcfg, ~mask, saved))
return;
dev_dbg(dev, "restored pin %u padcfg%u %#08x\n", pin, n, readl(padcfg));
}
int intel_pinctrl_resume_noirq(struct device *dev)
{
struct intel_pinctrl *pctrl = dev_get_drvdata(dev);
const struct intel_community_context *communities;
const struct intel_pad_context *pads;
int i;
/* Mask all interrupts */
intel_gpio_irq_init(pctrl);
pads = pctrl->context.pads;
for (i = 0; i < pctrl->soc->npins; i++) {
const struct pinctrl_pin_desc *desc = &pctrl->soc->pins[i];
if (!intel_pinctrl_should_save(pctrl, desc->number))
continue;
intel_restore_padcfg(pctrl, desc->number, PADCFG0, pads[i].padcfg0);
intel_restore_padcfg(pctrl, desc->number, PADCFG1, pads[i].padcfg1);
intel_restore_padcfg(pctrl, desc->number, PADCFG2, pads[i].padcfg2);
}
communities = pctrl->context.communities;
for (i = 0; i < pctrl->ncommunities; i++) {
struct intel_community *community = &pctrl->communities[i];
void __iomem *base;
unsigned int gpp;
base = community->regs + community->ie_offset;
for (gpp = 0; gpp < community->ngpps; gpp++)
intel_restore_intmask(pctrl, i, base, gpp, communities[i].intmask[gpp]);
base = community->regs + community->hostown_offset;
for (gpp = 0; gpp < community->ngpps; gpp++)
intel_restore_hostown(pctrl, i, base, gpp, communities[i].hostown[gpp]);
}
return 0;
}
EXPORT_SYMBOL_GPL(intel_pinctrl_resume_noirq);
#endif
MODULE_AUTHOR("Mathias Nyman <mathias.nyman@linux.intel.com>");
MODULE_AUTHOR("Mika Westerberg <mika.westerberg@linux.intel.com>");
MODULE_DESCRIPTION("Intel pinctrl/GPIO core driver");
MODULE_LICENSE("GPL v2");