Merge tag 'trace-v7.3-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-tracegrafted

Pull tracing fixes from Steven Rostedt:

 - Don't destroy user event fields when removal fails

   User event fields are destroyed before the event is removed from
   visibility. But that can fail leaving the still visible event with no
   fields. Move the destroying of the fields to after the event is
   successfully removed from visibility.

 - Initialize function graph state is fork before calling
   copy_exec_state()

   For non-CLONE_VM forks, copy_exec_state() allocates a new
   task_exec_state. If that allocation fails, ftrace_graph_exit_task()
   will free the tasks ret_stack pointer. Since that pointer is still
   using the parent's ret_stack, it mistakenly frees the parent's
   pointer too.

   Call ftrace_graph_init() on the task first which will NULL out the
   new tasks's ret_stack and if the copy fails, it will not free
   anything.

 - Remove FGRAPH_MAX_INDEX

   The macro FGRAPH_MAX_INDEX was added but never used. Remove it.

 - Save ent_size in function graph printing of nested functions

   The function graph tracer needs to look at the next event to see if
   the next event is the return of the current function entry. If it is,
   it prints a single line:

	ktime_get();

   Otherwise it prints it like a nested function:

	tick_nohz_irq_exit() {
	    ktime_get();
	    kcpustat_irq_exit();
	}

   In order to look at the next event, it must save the current event so
   that it has the information to print from it. It saves the event in
   the iterator descriptor called "ent". What it doesn't save is the
   ent_size of the event which is now used to know if the function graph
   arguments are to be printed. The peek doesn't save the size so the
   size used happens to be that of the size of the last event that was
   seen.

   Save the entry event size in the iterator descriptor so that the
   correct size is used.

 - Fix several errors with freeing data in the histogram code

   The histogram code had a lot of leaked or or incorrect accounting
   when failures happen. Correct them.

 - Fix histogram regression of .percent and .graph modifiers

   Up until 6.3 histogram values could have "percent" or "graph"
   modifiers that changed how they were printed. But a change that added
   restricting histograms values from being strings, stack traces and
   other modifiers inadvertently prevented them from using the percent
   and graph modifiers, which were legal use cases for values.

   Put back the percent and graph modifiers.

 - Fix various typos in the comments

 - Set the trace_clock before initializing a histogram with clock
   argument

   The histogram API allows the user to specific which trace clock to
   use via a "clock=" string. The histogram is set up first before the
   clock is checked. If the passed in clock is not valid, it exits
   without fully fixing up the histogram leaving it on the list and a
   use-after-free can trigger.

   Update the clock argument first and if it fails then exit gracefully
   before the histogram trigger is placed on any lists.

 - Restore :mod: trailer after parsing in ftrace_set_clr_event

   The function ftrace_set_clr_event() modifies the parse string and
   needs to put it back to what was passed in. It searches for ":mod:"
   via a strsep() but fails to put back the first ':' in the string.

   Add back the ':' in the passed in string.

 - Take trace_array reference when opening a tracer options file

   The options files are dynamically created and some tracers add their
   own options. When a tracer adds their own list of options, the
   trace_array holding them has an array to hold the list of options for
   each tracer. This array increases in size via a krealloc(), and the
   new entry gets a newly allocated array to hold the options of the new
   tracer being added.

   The element in each entry of the tracer's option array holds a
   pointer back to the trace_array, a pointer to the tracer it is
   associated to, a pointer to the flags of the option.

   The issue is that these arrays are freed when the trace_array is
   freed when its instance it represents is removed from the instances
   directory. There's a race that an open of one of these options files
   can happen when the instance is being removed.

   Add a new helper function to be called by the open function of the
   options file to iterate all existing trace_arrays under a lock and
   find the one that has the given option element in one of it's tracer
   arrays. If found, then update the associated trace_array's reference
   counter to keep it from being freed. If not found, have the open call
   return -ENODEV.

 - Disable interrupts when acquiring the lock in rb_wake_up_waiters()

   The function rb_wake_up_waiters() assumes it will be called in
   interrupt context and does not disable irqs when taking
   cpu_buffer->reader_lock, which can be called in hard interrupt
   context. The issue is in PREEMPT_RT, this function is called in
   thread context leaving this lock open to a deadlock.

   Take the lock with interrupts disabled.

 - Use rcu_assign_pointer() for tmp_ops filter hash

   The tmp_ops used in update_ftrace_direct_mod() assigns its
   filter_hash field directly, but that field is annotated as __rcu and
   sparse complains. Assign it with rcu_assign_pointer()

 - Fix use-after-free in enable_trigger_private_data_free()

   The trace_event_call is accessed through the event_trigger_data's
   trace_event_file pointer to put the trace_event_call on freeing. The
   issue is that the trace_event_file data may have been freed already
   causing a use-after-free. Add a field to the event_trigger_data that
   points directly to the trace_event_call so that it can decrement its
   reference directly without needing to go through the
   trace_event_file.

 - Fix accounting of buffer data remote headers

   trace_buffer_desc_size() and trace_remote_alloc_buffer() undercount
   the number of pages is needed for the asked for size as it doesn't
   take into account the meta data on each page. Add a helper function
   to do the calculation properly and use that in these functions.

 - Catch nr_page_va overflow in ring_buffer_desc sizing

   The number of pages per remote ring buffer is capped by
   ring_buffer_desc::nr_page_va (32 bits). A buffer_size large enough to
   overflow that field would silently allocate a descriptor smaller than
   what was asked for.

 - Do not resize the subbuf order if any per_cpu buffer is disabled

   The mmapping of ring buffers disables resizing the subbuffers, but it
   is done per-cpu whereas the subbuf size change is done for all the
   per_cpu buffers under the buffer->mutex. It could change the size of
   some while the mapping is happening on others. Have the resize of the
   subbuf order check all the per_cpu buffers under the lock to see if
   any of them is disabled before starting and causing an inconsistency
   between buffers that are being mapped.

* tag 'trace-v7.3-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace: (25 commits)
  ring-buffer: Check resize_disabled before publishing the new subbuf order
  tracing/remotes: Catch nr_page_va overflow in ring_buffer_desc sizing
  tracing/remotes: Account for ring buffer page header in size calculation
  tracing: Don't dereference trace_event_file in deferred trigger free
  ftrace: Use rcu_assign_pointer() for tmp_ops filter hash
  ring-buffer: Acquire the lock with irqsave in rb_wake_up_waiters()
  tracing: Take trace_array reference when opening a tracer options file
  tracing: Fix ring_buffer_read_page_size() kernel-doc
  tracing: Restore :mod: trailer after parsing in ftrace_set_clr_event()
  tracing: Fix memory corruption from a "STACKTRACE" histogram key
  tracing: Fix memory corruption from the histogram stacktrace modifier
  tracing: Undo the registration when enabling the histogram trigger fails
  tracing: Take the reference before publishing the named histogram trigger
  tracing: Set the trace clock before registering the histogram trigger
  tracing: Fix typo "preceeded" in comment
  tracing: Fix typo "availabe" in comment
  tracing: Let histogram values keep the percent and graph modifiers
  tracing: Keep the entry count when the histogram stats allocation fails
  tracing: Free histogram the field rejected for a bad modifier
  tracing: Free histogram the var ref when its initialization fails
  ...

5 files changed, 1206 insertions, 0 deletions

diff --git a/drivers/siox/Kconfig b/drivers/siox/Kconfig
new file mode 100644
index 000000000..4326db32c
--- /dev/null
+++ b/drivers/siox/Kconfig
@@ -0,0 +1,19 @@
+# SPDX-License-Identifier: GPL-2.0-only
+menuconfig SIOX
+	tristate "Eckelmann SIOX Support"
+	help
+	  SIOX stands for Serial Input Output eXtension and is a synchronous
+	  bus system invented by Eckelmann AG. It is used in their control and
+	  remote monitoring systems for commercial and industrial refrigeration
+	  to drive additional I/O units.
+
+	  Unless you know better, it is probably safe to say "no" here.
+
+if SIOX
+
+config SIOX_BUS_GPIO
+	tristate "SIOX GPIO bus driver"
+	help
+	  SIOX bus driver that controls the four bus lines using GPIOs.
+
+endif
diff --git a/drivers/siox/Makefile b/drivers/siox/Makefile
new file mode 100644
index 000000000..8b5ebb1a8
--- /dev/null
+++ b/drivers/siox/Makefile
@@ -0,0 +1,3 @@
+# SPDX-License-Identifier: GPL-2.0-only
+obj-$(CONFIG_SIOX) += siox-core.o
+obj-$(CONFIG_SIOX_BUS_GPIO) += siox-bus-gpio.o
diff --git a/drivers/siox/siox-bus-gpio.c b/drivers/siox/siox-bus-gpio.c
new file mode 100644
index 000000000..a0b62397e
--- /dev/null
+++ b/drivers/siox/siox-bus-gpio.c
@@ -0,0 +1,151 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2015-2017 Pengutronix, Uwe Kleine-König <kernel@pengutronix.de>
+ */
+
+#include <linux/gpio/consumer.h>
+#include <linux/module.h>
+#include <linux/platform_device.h>
+
+#include <linux/delay.h>
+
+#include "siox.h"
+
+#define DRIVER_NAME "siox-gpio"
+
+struct siox_gpio_ddata {
+	struct gpio_desc *din;
+	struct gpio_desc *dout;
+	struct gpio_desc *dclk;
+	struct gpio_desc *dld;
+};
+
+static unsigned int siox_clkhigh_ns = 1000;
+static unsigned int siox_loadhigh_ns;
+static unsigned int siox_bytegap_ns;
+
+static int siox_gpio_pushpull(struct siox_master *smaster,
+			      size_t setbuf_len, const u8 setbuf[],
+			      size_t getbuf_len, u8 getbuf[])
+{
+	struct siox_gpio_ddata *ddata = siox_master_get_devdata(smaster);
+	size_t i;
+	size_t cycles = max(setbuf_len, getbuf_len);
+
+	/* reset data and clock */
+	gpiod_set_value_cansleep(ddata->dout, 0);
+	gpiod_set_value_cansleep(ddata->dclk, 0);
+
+	gpiod_set_value_cansleep(ddata->dld, 1);
+	ndelay(siox_loadhigh_ns);
+	gpiod_set_value_cansleep(ddata->dld, 0);
+
+	for (i = 0; i < cycles; ++i) {
+		u8 set = 0, get = 0;
+		size_t j;
+
+		if (i >= cycles - setbuf_len)
+			set = setbuf[i - (cycles - setbuf_len)];
+
+		for (j = 0; j < 8; ++j) {
+			get <<= 1;
+			if (gpiod_get_value_cansleep(ddata->din))
+				get |= 1;
+
+			/* DOUT is logically inverted */
+			gpiod_set_value_cansleep(ddata->dout, !(set & 0x80));
+			set <<= 1;
+
+			gpiod_set_value_cansleep(ddata->dclk, 1);
+			ndelay(siox_clkhigh_ns);
+			gpiod_set_value_cansleep(ddata->dclk, 0);
+		}
+
+		if (i < getbuf_len)
+			getbuf[i] = get;
+
+		ndelay(siox_bytegap_ns);
+	}
+
+	gpiod_set_value_cansleep(ddata->dld, 1);
+	ndelay(siox_loadhigh_ns);
+	gpiod_set_value_cansleep(ddata->dld, 0);
+
+	/*
+	 * Resetting dout isn't necessary protocol wise, but it makes the
+	 * signals more pretty because the dout level is deterministic between
+	 * cycles. Note that this only affects dout between the master and the
+	 * first siox device. dout for the later devices depend on the output of
+	 * the previous siox device.
+	 */
+	gpiod_set_value_cansleep(ddata->dout, 0);
+
+	return 0;
+}
+
+static int siox_gpio_probe(struct platform_device *pdev)
+{
+	struct device *dev = &pdev->dev;
+	struct siox_gpio_ddata *ddata;
+	int ret;
+	struct siox_master *smaster;
+
+	smaster = devm_siox_master_alloc(dev, sizeof(*ddata));
+	if (!smaster)
+		return -ENOMEM;
+
+	platform_set_drvdata(pdev, smaster);
+	ddata = siox_master_get_devdata(smaster);
+
+	ddata->din = devm_gpiod_get(dev, "din", GPIOD_IN);
+	if (IS_ERR(ddata->din))
+		return dev_err_probe(dev, PTR_ERR(ddata->din),
+				     "Failed to get din GPIO\n");
+
+	ddata->dout = devm_gpiod_get(dev, "dout", GPIOD_OUT_LOW);
+	if (IS_ERR(ddata->dout))
+		return dev_err_probe(dev, PTR_ERR(ddata->dout),
+				     "Failed to get dout GPIO\n");
+
+	ddata->dclk = devm_gpiod_get(dev, "dclk", GPIOD_OUT_LOW);
+	if (IS_ERR(ddata->dclk))
+		return dev_err_probe(dev, PTR_ERR(ddata->dclk),
+				     "Failed to get dclk GPIO\n");
+
+	ddata->dld = devm_gpiod_get(dev, "dld", GPIOD_OUT_LOW);
+	if (IS_ERR(ddata->dld))
+		return dev_err_probe(dev, PTR_ERR(ddata->dld),
+				     "Failed to get dld GPIO\n");
+
+	smaster->pushpull = siox_gpio_pushpull;
+	/* XXX: determine automatically like spi does */
+	smaster->busno = 0;
+
+	ret = devm_siox_master_register(dev, smaster);
+	if (ret)
+		return dev_err_probe(dev, ret,
+				     "Failed to register siox master\n");
+
+	return 0;
+}
+
+static const struct of_device_id siox_gpio_dt_ids[] = {
+	{ .compatible = "eckelmann,siox-gpio", },
+	{ /* sentinel */ }
+};
+MODULE_DEVICE_TABLE(of, siox_gpio_dt_ids);
+
+static struct platform_driver siox_gpio_driver = {
+	.probe = siox_gpio_probe,
+
+	.driver = {
+		.name = DRIVER_NAME,
+		.of_match_table = siox_gpio_dt_ids,
+	},
+};
+module_platform_driver(siox_gpio_driver);
+
+MODULE_AUTHOR("Uwe Kleine-Koenig <u.kleine-koenig@pengutronix.de>");
+MODULE_DESCRIPTION("SIOX GPIO bus driver");
+MODULE_LICENSE("GPL v2");
+MODULE_ALIAS("platform:" DRIVER_NAME);
diff --git a/drivers/siox/siox-core.c b/drivers/siox/siox-core.c
new file mode 100644
index 000000000..3e8f3b6a4
--- /dev/null
+++ b/drivers/siox/siox-core.c
@@ -0,0 +1,980 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2015-2017 Pengutronix, Uwe Kleine-König <kernel@pengutronix.de>
+ */
+#include <linux/kernel.h>
+#include <linux/device.h>
+#include <linux/module.h>
+#include <linux/slab.h>
+#include <linux/sysfs.h>
+
+#include "siox.h"
+
+/*
+ * The lowest bit in the SIOX status word signals if the in-device watchdog is
+ * ok. If the bit is set, the device is functional.
+ *
+ * On writing the watchdog timer is reset when this bit toggles.
+ */
+#define SIOX_STATUS_WDG			0x01
+
+/*
+ * Bits 1 to 3 of the status word read as the bitwise negation of what was
+ * clocked in before. The value clocked in is changed in each cycle and so
+ * allows to detect transmit/receive problems.
+ */
+#define SIOX_STATUS_COUNTER		0x0e
+
+/*
+ * Each Siox-Device has a 4 bit type number that is neither 0 nor 15. This is
+ * available in the upper nibble of the read status.
+ *
+ * On write these bits are DC.
+ */
+#define SIOX_STATUS_TYPE		0xf0
+
+#define CREATE_TRACE_POINTS
+#include <trace/events/siox.h>
+
+static bool siox_is_registered;
+
+static void siox_master_lock(struct siox_master *smaster)
+{
+	mutex_lock(&smaster->lock);
+}
+
+static void siox_master_unlock(struct siox_master *smaster)
+{
+	mutex_unlock(&smaster->lock);
+}
+
+static inline u8 siox_status_clean(u8 status_read, u8 status_written)
+{
+	/*
+	 * bits 3:1 of status sample the respective bit in the status
+	 * byte written in the previous cycle but inverted. So if you wrote the
+	 * status word as 0xa before (counter = 0b101), it is expected to get
+	 * back the counter bits as 0b010.
+	 *
+	 * So given the last status written this function toggles the there
+	 * unset counter bits in the read value such that the counter bits in
+	 * the return value are all zero iff the bits were read as expected to
+	 * simplify error detection.
+	 */
+
+	return status_read ^ (~status_written & 0xe);
+}
+
+static bool siox_device_counter_error(struct siox_device *sdevice,
+				      u8 status_clean)
+{
+	return (status_clean & SIOX_STATUS_COUNTER) != 0;
+}
+
+static bool siox_device_type_error(struct siox_device *sdevice, u8 status_clean)
+{
+	u8 statustype = (status_clean & SIOX_STATUS_TYPE) >> 4;
+
+	/*
+	 * If the device knows which value the type bits should have, check
+	 * against this value otherwise just rule out the invalid values 0b0000
+	 * and 0b1111.
+	 */
+	if (sdevice->statustype) {
+		if (statustype != sdevice->statustype)
+			return true;
+	} else {
+		switch (statustype) {
+		case 0:
+		case 0xf:
+			return true;
+		}
+	}
+
+	return false;
+}
+
+static bool siox_device_wdg_error(struct siox_device *sdevice, u8 status_clean)
+{
+	return (status_clean & SIOX_STATUS_WDG) == 0;
+}
+
+/*
+ * If there is a type or counter error the device is called "unsynced".
+ */
+bool siox_device_synced(struct siox_device *sdevice)
+{
+	if (siox_device_type_error(sdevice, sdevice->status_read_clean))
+		return false;
+
+	return !siox_device_counter_error(sdevice, sdevice->status_read_clean);
+
+}
+EXPORT_SYMBOL_GPL(siox_device_synced);
+
+/*
+ * A device is called "connected" if it is synced and the watchdog is not
+ * asserted.
+ */
+bool siox_device_connected(struct siox_device *sdevice)
+{
+	if (!siox_device_synced(sdevice))
+		return false;
+
+	return !siox_device_wdg_error(sdevice, sdevice->status_read_clean);
+}
+EXPORT_SYMBOL_GPL(siox_device_connected);
+
+static void siox_poll(struct siox_master *smaster)
+{
+	struct siox_device *sdevice;
+	size_t i = smaster->setbuf_len;
+	unsigned int devno = 0;
+	int unsync_error = 0;
+
+	smaster->last_poll = jiffies;
+
+	/*
+	 * The counter bits change in each second cycle, the watchdog bit
+	 * toggles each time.
+	 * The counter bits hold values from [0, 6]. 7 would be possible
+	 * theoretically but the protocol designer considered that a bad idea
+	 * for reasons unknown today. (Maybe that's because then the status read
+	 * back has only zeros in the counter bits then which might be confused
+	 * with a stuck-at-0 error. But for the same reason (with s/0/1/) 0
+	 * could be skipped.)
+	 */
+	if (++smaster->status > 0x0d)
+		smaster->status = 0;
+
+	memset(smaster->buf, 0, smaster->setbuf_len);
+
+	/* prepare data pushed out to devices in buf[0..setbuf_len) */
+	list_for_each_entry(sdevice, &smaster->devices, node) {
+		struct siox_driver *sdriver =
+			to_siox_driver(sdevice->dev.driver);
+		sdevice->status_written = smaster->status;
+
+		i -= sdevice->inbytes;
+
+		/*
+		 * If the device or a previous one is unsynced, don't pet the
+		 * watchdog. This is done to ensure that the device is kept in
+		 * reset when something is wrong.
+		 */
+		if (!siox_device_synced(sdevice))
+			unsync_error = 1;
+
+		if (sdriver && !unsync_error)
+			sdriver->set_data(sdevice, sdevice->status_written,
+					  &smaster->buf[i + 1]);
+		else
+			/*
+			 * Don't trigger watchdog if there is no driver or a
+			 * sync problem
+			 */
+			sdevice->status_written &= ~SIOX_STATUS_WDG;
+
+		smaster->buf[i] = sdevice->status_written;
+
+		trace_siox_set_data(smaster, sdevice, devno, i);
+
+		devno++;
+	}
+
+	smaster->pushpull(smaster, smaster->setbuf_len, smaster->buf,
+			  smaster->getbuf_len,
+			  smaster->buf + smaster->setbuf_len);
+
+	unsync_error = 0;
+
+	/* interpret data pulled in from devices in buf[setbuf_len..] */
+	devno = 0;
+	i = smaster->setbuf_len;
+	list_for_each_entry(sdevice, &smaster->devices, node) {
+		struct siox_driver *sdriver =
+			to_siox_driver(sdevice->dev.driver);
+		u8 status = smaster->buf[i + sdevice->outbytes - 1];
+		u8 status_clean;
+		u8 prev_status_clean = sdevice->status_read_clean;
+		bool synced = true;
+		bool connected = true;
+
+		if (!siox_device_synced(sdevice))
+			unsync_error = 1;
+
+		/*
+		 * If the watchdog bit wasn't toggled in this cycle, report the
+		 * watchdog as active to give a consistent view for drivers and
+		 * sysfs consumers.
+		 */
+		if (!sdriver || unsync_error)
+			status &= ~SIOX_STATUS_WDG;
+
+		status_clean =
+			siox_status_clean(status,
+					  sdevice->status_written_lastcycle);
+
+		/* Check counter and type bits */
+		if (siox_device_counter_error(sdevice, status_clean) ||
+		    siox_device_type_error(sdevice, status_clean)) {
+			bool prev_error;
+
+			synced = false;
+
+			/* only report a new error if the last cycle was ok */
+			prev_error =
+				siox_device_counter_error(sdevice,
+							  prev_status_clean) ||
+				siox_device_type_error(sdevice,
+						       prev_status_clean);
+
+			if (!prev_error) {
+				sdevice->status_errors++;
+				sysfs_notify_dirent(sdevice->status_errors_kn);
+			}
+		}
+
+		/* If the device is unsynced report the watchdog as active */
+		if (!synced) {
+			status &= ~SIOX_STATUS_WDG;
+			status_clean &= ~SIOX_STATUS_WDG;
+		}
+
+		if (siox_device_wdg_error(sdevice, status_clean))
+			connected = false;
+
+		/* The watchdog state changed just now */
+		if ((status_clean ^ prev_status_clean) & SIOX_STATUS_WDG) {
+			sysfs_notify_dirent(sdevice->watchdog_kn);
+
+			if (siox_device_wdg_error(sdevice, status_clean)) {
+				struct kernfs_node *wd_errs =
+					sdevice->watchdog_errors_kn;
+
+				sdevice->watchdog_errors++;
+				sysfs_notify_dirent(wd_errs);
+			}
+		}
+
+		if (connected != sdevice->connected)
+			sysfs_notify_dirent(sdevice->connected_kn);
+
+		sdevice->status_read_clean = status_clean;
+		sdevice->status_written_lastcycle = sdevice->status_written;
+		sdevice->connected = connected;
+
+		trace_siox_get_data(smaster, sdevice, devno, status_clean, i);
+
+		/* only give data read to driver if the device is connected */
+		if (sdriver && connected)
+			sdriver->get_data(sdevice, &smaster->buf[i]);
+
+		devno++;
+		i += sdevice->outbytes;
+	}
+}
+
+static int siox_poll_thread(void *data)
+{
+	struct siox_master *smaster = data;
+	signed long timeout = 0;
+
+	get_device(&smaster->dev);
+
+	for (;;) {
+		if (kthread_should_stop()) {
+			put_device(&smaster->dev);
+			return 0;
+		}
+
+		siox_master_lock(smaster);
+
+		if (smaster->active) {
+			unsigned long next_poll =
+				smaster->last_poll + smaster->poll_interval;
+			if (time_is_before_eq_jiffies(next_poll))
+				siox_poll(smaster);
+
+			timeout = smaster->poll_interval -
+				(jiffies - smaster->last_poll);
+		} else {
+			timeout = MAX_SCHEDULE_TIMEOUT;
+		}
+
+		/*
+		 * Set the task to idle while holding the lock. This makes sure
+		 * that we don't sleep too long when the bus is reenabled before
+		 * schedule_timeout is reached.
+		 */
+		if (timeout > 0)
+			set_current_state(TASK_IDLE);
+
+		siox_master_unlock(smaster);
+
+		if (timeout > 0)
+			schedule_timeout(timeout);
+
+		/*
+		 * I'm not clear if/why it is important to set the state to
+		 * RUNNING again, but it fixes a "do not call blocking ops when
+		 * !TASK_RUNNING;"-warning.
+		 */
+		set_current_state(TASK_RUNNING);
+	}
+}
+
+static int __siox_start(struct siox_master *smaster)
+{
+	if (!(smaster->setbuf_len + smaster->getbuf_len))
+		return -ENODEV;
+
+	if (!smaster->buf)
+		return -ENOMEM;
+
+	if (smaster->active)
+		return 0;
+
+	smaster->active = 1;
+	wake_up_process(smaster->poll_thread);
+
+	return 1;
+}
+
+static int siox_start(struct siox_master *smaster)
+{
+	int ret;
+
+	siox_master_lock(smaster);
+	ret = __siox_start(smaster);
+	siox_master_unlock(smaster);
+
+	return ret;
+}
+
+static int __siox_stop(struct siox_master *smaster)
+{
+	if (smaster->active) {
+		struct siox_device *sdevice;
+
+		smaster->active = 0;
+
+		list_for_each_entry(sdevice, &smaster->devices, node) {
+			if (sdevice->connected)
+				sysfs_notify_dirent(sdevice->connected_kn);
+			sdevice->connected = false;
+		}
+
+		return 1;
+	}
+	return 0;
+}
+
+static int siox_stop(struct siox_master *smaster)
+{
+	int ret;
+
+	siox_master_lock(smaster);
+	ret = __siox_stop(smaster);
+	siox_master_unlock(smaster);
+
+	return ret;
+}
+
+static ssize_t type_show(struct device *dev,
+			 struct device_attribute *attr, char *buf)
+{
+	struct siox_device *sdev = to_siox_device(dev);
+
+	return sprintf(buf, "%s\n", sdev->type);
+}
+
+static DEVICE_ATTR_RO(type);
+
+static ssize_t inbytes_show(struct device *dev,
+			    struct device_attribute *attr, char *buf)
+{
+	struct siox_device *sdev = to_siox_device(dev);
+
+	return sprintf(buf, "%zu\n", sdev->inbytes);
+}
+
+static DEVICE_ATTR_RO(inbytes);
+
+static ssize_t outbytes_show(struct device *dev,
+			     struct device_attribute *attr, char *buf)
+{
+	struct siox_device *sdev = to_siox_device(dev);
+
+	return sprintf(buf, "%zu\n", sdev->outbytes);
+}
+
+static DEVICE_ATTR_RO(outbytes);
+
+static ssize_t status_errors_show(struct device *dev,
+				  struct device_attribute *attr, char *buf)
+{
+	struct siox_device *sdev = to_siox_device(dev);
+	unsigned int status_errors;
+
+	siox_master_lock(sdev->smaster);
+
+	status_errors = sdev->status_errors;
+
+	siox_master_unlock(sdev->smaster);
+
+	return sprintf(buf, "%u\n", status_errors);
+}
+
+static DEVICE_ATTR_RO(status_errors);
+
+static ssize_t connected_show(struct device *dev,
+			      struct device_attribute *attr, char *buf)
+{
+	struct siox_device *sdev = to_siox_device(dev);
+	bool connected;
+
+	siox_master_lock(sdev->smaster);
+
+	connected = sdev->connected;
+
+	siox_master_unlock(sdev->smaster);
+
+	return sprintf(buf, "%u\n", connected);
+}
+
+static DEVICE_ATTR_RO(connected);
+
+static ssize_t watchdog_show(struct device *dev,
+			     struct device_attribute *attr, char *buf)
+{
+	struct siox_device *sdev = to_siox_device(dev);
+	u8 status;
+
+	siox_master_lock(sdev->smaster);
+
+	status = sdev->status_read_clean;
+
+	siox_master_unlock(sdev->smaster);
+
+	return sprintf(buf, "%d\n", status & SIOX_STATUS_WDG);
+}
+
+static DEVICE_ATTR_RO(watchdog);
+
+static ssize_t watchdog_errors_show(struct device *dev,
+				    struct device_attribute *attr, char *buf)
+{
+	struct siox_device *sdev = to_siox_device(dev);
+	unsigned int watchdog_errors;
+
+	siox_master_lock(sdev->smaster);
+
+	watchdog_errors = sdev->watchdog_errors;
+
+	siox_master_unlock(sdev->smaster);
+
+	return sprintf(buf, "%u\n", watchdog_errors);
+}
+
+static DEVICE_ATTR_RO(watchdog_errors);
+
+static struct attribute *siox_device_attrs[] = {
+	&dev_attr_type.attr,
+	&dev_attr_inbytes.attr,
+	&dev_attr_outbytes.attr,
+	&dev_attr_status_errors.attr,
+	&dev_attr_connected.attr,
+	&dev_attr_watchdog.attr,
+	&dev_attr_watchdog_errors.attr,
+	NULL
+};
+ATTRIBUTE_GROUPS(siox_device);
+
+static void siox_device_release(struct device *dev)
+{
+	struct siox_device *sdevice = to_siox_device(dev);
+
+	kfree(sdevice);
+}
+
+static const struct device_type siox_device_type = {
+	.groups = siox_device_groups,
+	.release = siox_device_release,
+};
+
+static int siox_match(struct device *dev, const struct device_driver *drv)
+{
+	if (dev->type != &siox_device_type)
+		return 0;
+
+	/* up to now there is only a single driver so keeping this simple */
+	return 1;
+}
+
+static int siox_probe(struct device *dev)
+{
+	struct siox_driver *sdriver = to_siox_driver(dev->driver);
+	struct siox_device *sdevice = to_siox_device(dev);
+
+	return sdriver->probe(sdevice);
+}
+
+static void siox_remove(struct device *dev)
+{
+	struct siox_driver *sdriver =
+		container_of(dev->driver, struct siox_driver, driver);
+	struct siox_device *sdevice = to_siox_device(dev);
+
+	if (sdriver->remove)
+		sdriver->remove(sdevice);
+}
+
+static void siox_shutdown(struct device *dev)
+{
+	struct siox_device *sdevice = to_siox_device(dev);
+	struct siox_driver *sdriver;
+
+	if (!dev->driver)
+		return;
+
+	sdriver = container_of(dev->driver, struct siox_driver, driver);
+	if (sdriver->shutdown)
+		sdriver->shutdown(sdevice);
+}
+
+static const struct bus_type siox_bus_type = {
+	.name = "siox",
+	.match = siox_match,
+	.probe = siox_probe,
+	.remove = siox_remove,
+	.shutdown = siox_shutdown,
+};
+
+static ssize_t active_show(struct device *dev,
+			   struct device_attribute *attr, char *buf)
+{
+	struct siox_master *smaster = to_siox_master(dev);
+
+	return sprintf(buf, "%d\n", smaster->active);
+}
+
+static ssize_t active_store(struct device *dev,
+			    struct device_attribute *attr,
+			    const char *buf, size_t count)
+{
+	struct siox_master *smaster = to_siox_master(dev);
+	int ret;
+	int active;
+
+	ret = kstrtoint(buf, 0, &active);
+	if (ret < 0)
+		return ret;
+
+	if (active)
+		ret = siox_start(smaster);
+	else
+		ret = siox_stop(smaster);
+
+	if (ret < 0)
+		return ret;
+
+	return count;
+}
+
+static DEVICE_ATTR_RW(active);
+
+static struct siox_device *siox_device_add(struct siox_master *smaster,
+					   const char *type, size_t inbytes,
+					   size_t outbytes, u8 statustype);
+
+static ssize_t device_add_store(struct device *dev,
+				struct device_attribute *attr,
+				const char *buf, size_t count)
+{
+	struct siox_master *smaster = to_siox_master(dev);
+	int ret;
+	char type[20] = "";
+	size_t inbytes = 0, outbytes = 0;
+	u8 statustype = 0;
+
+	ret = sscanf(buf, "%19s %zu %zu %hhu", type, &inbytes,
+		     &outbytes, &statustype);
+	if (ret != 3 && ret != 4)
+		return -EINVAL;
+
+	if (strcmp(type, "siox-12x8") || inbytes != 2 || outbytes != 4)
+		return -EINVAL;
+
+	siox_device_add(smaster, "siox-12x8", inbytes, outbytes, statustype);
+
+	return count;
+}
+
+static DEVICE_ATTR_WO(device_add);
+
+static void siox_device_remove(struct siox_master *smaster);
+
+static ssize_t device_remove_store(struct device *dev,
+				   struct device_attribute *attr,
+				   const char *buf, size_t count)
+{
+	struct siox_master *smaster = to_siox_master(dev);
+
+	/* XXX? require to write <type> <inbytes> <outbytes> */
+	siox_device_remove(smaster);
+
+	return count;
+}
+
+static DEVICE_ATTR_WO(device_remove);
+
+static ssize_t poll_interval_ns_show(struct device *dev,
+				     struct device_attribute *attr, char *buf)
+{
+	struct siox_master *smaster = to_siox_master(dev);
+
+	return sprintf(buf, "%lld\n", jiffies_to_nsecs(smaster->poll_interval));
+}
+
+static ssize_t poll_interval_ns_store(struct device *dev,
+				      struct device_attribute *attr,
+				      const char *buf, size_t count)
+{
+	struct siox_master *smaster = to_siox_master(dev);
+	int ret;
+	u64 val;
+
+	ret = kstrtou64(buf, 0, &val);
+	if (ret < 0)
+		return ret;
+
+	siox_master_lock(smaster);
+
+	smaster->poll_interval = nsecs_to_jiffies(val);
+
+	siox_master_unlock(smaster);
+
+	return count;
+}
+
+static DEVICE_ATTR_RW(poll_interval_ns);
+
+static struct attribute *siox_master_attrs[] = {
+	&dev_attr_active.attr,
+	&dev_attr_device_add.attr,
+	&dev_attr_device_remove.attr,
+	&dev_attr_poll_interval_ns.attr,
+	NULL
+};
+ATTRIBUTE_GROUPS(siox_master);
+
+static void siox_master_release(struct device *dev)
+{
+	struct siox_master *smaster = to_siox_master(dev);
+
+	kfree(smaster);
+}
+
+static const struct device_type siox_master_type = {
+	.groups = siox_master_groups,
+	.release = siox_master_release,
+};
+
+struct siox_master *siox_master_alloc(struct device *dev,
+				      size_t size)
+{
+	struct siox_master *smaster;
+
+	if (!dev)
+		return NULL;
+
+	smaster = kzalloc(sizeof(*smaster) + size, GFP_KERNEL);
+	if (!smaster)
+		return NULL;
+
+	device_initialize(&smaster->dev);
+
+	smaster->busno = -1;
+	smaster->dev.bus = &siox_bus_type;
+	smaster->dev.type = &siox_master_type;
+	smaster->dev.parent = dev;
+	smaster->poll_interval = DIV_ROUND_UP(HZ, 40);
+
+	dev_set_drvdata(&smaster->dev, &smaster[1]);
+
+	return smaster;
+}
+EXPORT_SYMBOL_GPL(siox_master_alloc);
+
+static void devm_siox_master_put(void *data)
+{
+	struct siox_master *smaster = data;
+
+	siox_master_put(smaster);
+}
+
+struct siox_master *devm_siox_master_alloc(struct device *dev,
+					   size_t size)
+{
+	struct siox_master *smaster;
+	int ret;
+
+	smaster = siox_master_alloc(dev, size);
+	if (!smaster)
+		return NULL;
+
+	ret = devm_add_action_or_reset(dev, devm_siox_master_put, smaster);
+	if (ret)
+		return NULL;
+
+	return smaster;
+}
+EXPORT_SYMBOL_GPL(devm_siox_master_alloc);
+
+int siox_master_register(struct siox_master *smaster)
+{
+	int ret;
+
+	if (!siox_is_registered)
+		return -EPROBE_DEFER;
+
+	if (!smaster->pushpull)
+		return -EINVAL;
+
+	get_device(&smaster->dev);
+
+	dev_set_name(&smaster->dev, "siox-%d", smaster->busno);
+
+	mutex_init(&smaster->lock);
+	INIT_LIST_HEAD(&smaster->devices);
+
+	smaster->last_poll = jiffies;
+	smaster->poll_thread = kthread_run(siox_poll_thread, smaster,
+					   "siox-%d", smaster->busno);
+	if (IS_ERR(smaster->poll_thread)) {
+		smaster->active = 0;
+		return PTR_ERR(smaster->poll_thread);
+	}
+
+	ret = device_add(&smaster->dev);
+	if (ret)
+		kthread_stop(smaster->poll_thread);
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(siox_master_register);
+
+void siox_master_unregister(struct siox_master *smaster)
+{
+	/* remove device */
+	device_del(&smaster->dev);
+
+	siox_master_lock(smaster);
+
+	__siox_stop(smaster);
+
+	while (smaster->num_devices) {
+		struct siox_device *sdevice;
+
+		sdevice = container_of(smaster->devices.prev,
+				       struct siox_device, node);
+		list_del(&sdevice->node);
+		smaster->num_devices--;
+
+		siox_master_unlock(smaster);
+
+		device_unregister(&sdevice->dev);
+
+		siox_master_lock(smaster);
+	}
+
+	siox_master_unlock(smaster);
+
+	put_device(&smaster->dev);
+}
+EXPORT_SYMBOL_GPL(siox_master_unregister);
+
+static void devm_siox_master_unregister(void *data)
+{
+	struct siox_master *smaster = data;
+
+	siox_master_unregister(smaster);
+}
+
+int devm_siox_master_register(struct device *dev, struct siox_master *smaster)
+{
+	int ret;
+
+	ret = siox_master_register(smaster);
+	if (ret)
+		return ret;
+
+	return devm_add_action_or_reset(dev, devm_siox_master_unregister, smaster);
+}
+EXPORT_SYMBOL_GPL(devm_siox_master_register);
+
+static struct siox_device *siox_device_add(struct siox_master *smaster,
+					   const char *type, size_t inbytes,
+					   size_t outbytes, u8 statustype)
+{
+	struct siox_device *sdevice;
+	int ret;
+	size_t buf_len;
+
+	sdevice = kzalloc_obj(*sdevice);
+	if (!sdevice)
+		return ERR_PTR(-ENOMEM);
+
+	sdevice->type = type;
+	sdevice->inbytes = inbytes;
+	sdevice->outbytes = outbytes;
+	sdevice->statustype = statustype;
+
+	sdevice->smaster = smaster;
+	sdevice->dev.parent = &smaster->dev;
+	sdevice->dev.bus = &siox_bus_type;
+	sdevice->dev.type = &siox_device_type;
+
+	siox_master_lock(smaster);
+
+	dev_set_name(&sdevice->dev, "siox-%d-%d",
+		     smaster->busno, smaster->num_devices);
+
+	buf_len = smaster->setbuf_len + inbytes +
+		smaster->getbuf_len + outbytes;
+	if (smaster->buf_len < buf_len) {
+		u8 *buf = krealloc(smaster->buf, buf_len, GFP_KERNEL);
+
+		if (!buf) {
+			dev_err(&smaster->dev,
+				"failed to realloc buffer to %zu\n", buf_len);
+			ret = -ENOMEM;
+			goto err_buf_alloc;
+		}
+
+		smaster->buf_len = buf_len;
+		smaster->buf = buf;
+	}
+
+	ret = device_register(&sdevice->dev);
+	if (ret) {
+		dev_err(&smaster->dev, "failed to register device: %d\n", ret);
+
+		goto err_device_register;
+	}
+
+	smaster->num_devices++;
+	list_add_tail(&sdevice->node, &smaster->devices);
+
+	smaster->setbuf_len += sdevice->inbytes;
+	smaster->getbuf_len += sdevice->outbytes;
+
+	sdevice->status_errors_kn = sysfs_get_dirent(sdevice->dev.kobj.sd,
+						     "status_errors");
+	sdevice->watchdog_kn = sysfs_get_dirent(sdevice->dev.kobj.sd,
+						"watchdog");
+	sdevice->watchdog_errors_kn = sysfs_get_dirent(sdevice->dev.kobj.sd,
+						       "watchdog_errors");
+	sdevice->connected_kn = sysfs_get_dirent(sdevice->dev.kobj.sd,
+						 "connected");
+
+	siox_master_unlock(smaster);
+
+	return sdevice;
+
+err_device_register:
+	/* don't care to make the buffer smaller again */
+	put_device(&sdevice->dev);
+	sdevice = NULL;
+
+err_buf_alloc:
+	siox_master_unlock(smaster);
+
+	kfree(sdevice);
+
+	return ERR_PTR(ret);
+}
+
+static void siox_device_remove(struct siox_master *smaster)
+{
+	struct siox_device *sdevice;
+
+	siox_master_lock(smaster);
+
+	if (!smaster->num_devices) {
+		siox_master_unlock(smaster);
+		return;
+	}
+
+	sdevice = container_of(smaster->devices.prev, struct siox_device, node);
+	list_del(&sdevice->node);
+	smaster->num_devices--;
+
+	smaster->setbuf_len -= sdevice->inbytes;
+	smaster->getbuf_len -= sdevice->outbytes;
+
+	if (!smaster->num_devices)
+		__siox_stop(smaster);
+
+	siox_master_unlock(smaster);
+
+	/*
+	 * This must be done without holding the master lock because we're
+	 * called from device_remove_store which also holds a sysfs mutex.
+	 * device_unregister tries to aquire the same lock.
+	 */
+	device_unregister(&sdevice->dev);
+}
+
+int __siox_driver_register(struct siox_driver *sdriver, struct module *owner)
+{
+	int ret;
+
+	if (unlikely(!siox_is_registered))
+		return -EPROBE_DEFER;
+
+	if (!sdriver->probe ||
+	    (!sdriver->set_data && !sdriver->get_data)) {
+		pr_err("Driver %s doesn't provide needed callbacks\n",
+		       sdriver->driver.name);
+		return -EINVAL;
+	}
+
+	sdriver->driver.owner = owner;
+	sdriver->driver.bus = &siox_bus_type;
+
+	ret = driver_register(&sdriver->driver);
+	if (ret)
+		pr_err("Failed to register siox driver %s (%d)\n",
+		       sdriver->driver.name, ret);
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(__siox_driver_register);
+
+static int __init siox_init(void)
+{
+	int ret;
+
+	ret = bus_register(&siox_bus_type);
+	if (ret) {
+		pr_err("Registration of SIOX bus type failed: %d\n", ret);
+		return ret;
+	}
+
+	siox_is_registered = true;
+
+	return 0;
+}
+subsys_initcall(siox_init);
+
+static void __exit siox_exit(void)
+{
+	bus_unregister(&siox_bus_type);
+}
+module_exit(siox_exit);
+
+MODULE_AUTHOR("Uwe Kleine-Koenig <u.kleine-koenig@pengutronix.de>");
+MODULE_DESCRIPTION("Eckelmann SIOX driver core");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/siox/siox.h b/drivers/siox/siox.h
new file mode 100644
index 000000000..513f2c831
--- /dev/null
+++ b/drivers/siox/siox.h
@@ -0,0 +1,53 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2015-2017 Pengutronix, Uwe Kleine-König <kernel@pengutronix.de>
+ */
+#include <linux/kernel.h>
+#include <linux/kthread.h>
+#include <linux/siox.h>
+
+#define to_siox_master(_dev)	container_of((_dev), struct siox_master, dev)
+struct siox_master {
+	/* these fields should be initialized by the driver */
+	int busno;
+	int (*pushpull)(struct siox_master *smaster,
+			size_t setbuf_len, const u8 setbuf[],
+			size_t getbuf_len, u8 getbuf[]);
+
+	/* might be initialized by the driver, if 0 it is set to HZ / 40 */
+	unsigned long poll_interval; /* in jiffies */
+
+	/* framework private stuff */
+	struct mutex lock;
+	bool active;
+	struct module *owner;
+	struct device dev;
+	unsigned int num_devices;
+	struct list_head devices;
+
+	size_t setbuf_len, getbuf_len;
+	size_t buf_len;
+	u8 *buf;
+	u8 status;
+
+	unsigned long last_poll;
+	struct task_struct *poll_thread;
+};
+
+static inline void *siox_master_get_devdata(struct siox_master *smaster)
+{
+	return dev_get_drvdata(&smaster->dev);
+}
+
+struct siox_master *siox_master_alloc(struct device *dev, size_t size);
+static inline void siox_master_put(struct siox_master *smaster)
+{
+	put_device(&smaster->dev);
+}
+
+struct siox_master *devm_siox_master_alloc(struct device *dev, size_t size);
+
+int siox_master_register(struct siox_master *smaster);
+void siox_master_unregister(struct siox_master *smaster);
+
+int devm_siox_master_register(struct device *dev, struct siox_master *smaster);