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
  ...

1 files changed, 1263 insertions, 0 deletions

diff --git a/drivers/greybus/interface.c b/drivers/greybus/interface.c
new file mode 100644
index 000000000..4ee4bda4a
--- /dev/null
+++ b/drivers/greybus/interface.c
@@ -0,0 +1,1263 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Greybus interface code
+ *
+ * Copyright 2014 Google Inc.
+ * Copyright 2014 Linaro Ltd.
+ */
+
+#include <linux/delay.h>
+#include <linux/greybus.h>
+
+#include "greybus_trace.h"
+
+#define GB_INTERFACE_MODE_SWITCH_TIMEOUT	2000
+
+#define GB_INTERFACE_DEVICE_ID_BAD	0xff
+
+#define GB_INTERFACE_AUTOSUSPEND_MS			3000
+
+/* Time required for interface to enter standby before disabling REFCLK */
+#define GB_INTERFACE_SUSPEND_HIBERNATE_DELAY_MS			20
+
+/* Don't-care selector index */
+#define DME_SELECTOR_INDEX_NULL		0
+
+/* DME attributes */
+/* FIXME: remove ES2 support and DME_T_TST_SRC_INCREMENT */
+#define DME_T_TST_SRC_INCREMENT		0x4083
+
+#define DME_DDBL1_MANUFACTURERID	0x5003
+#define DME_DDBL1_PRODUCTID		0x5004
+
+#define DME_TOSHIBA_GMP_VID		0x6000
+#define DME_TOSHIBA_GMP_PID		0x6001
+#define DME_TOSHIBA_GMP_SN0		0x6002
+#define DME_TOSHIBA_GMP_SN1		0x6003
+#define DME_TOSHIBA_GMP_INIT_STATUS	0x6101
+
+/* DDBL1 Manufacturer and Product ids */
+#define TOSHIBA_DMID			0x0126
+#define TOSHIBA_ES2_BRIDGE_DPID		0x1000
+#define TOSHIBA_ES3_APBRIDGE_DPID	0x1001
+#define TOSHIBA_ES3_GBPHY_DPID	0x1002
+
+static int gb_interface_hibernate_link(struct gb_interface *intf);
+static int gb_interface_refclk_set(struct gb_interface *intf, bool enable);
+
+static int gb_interface_dme_attr_get(struct gb_interface *intf,
+				     u16 attr, u32 *val)
+{
+	return gb_svc_dme_peer_get(intf->hd->svc, intf->interface_id,
+					attr, DME_SELECTOR_INDEX_NULL, val);
+}
+
+static int gb_interface_read_ara_dme(struct gb_interface *intf)
+{
+	u32 sn0, sn1;
+	int ret;
+
+	/*
+	 * Unless this is a Toshiba bridge, bail out until we have defined
+	 * standard GMP attributes.
+	 */
+	if (intf->ddbl1_manufacturer_id != TOSHIBA_DMID) {
+		dev_err(&intf->dev, "unknown manufacturer %08x\n",
+			intf->ddbl1_manufacturer_id);
+		return -ENODEV;
+	}
+
+	ret = gb_interface_dme_attr_get(intf, DME_TOSHIBA_GMP_VID,
+					&intf->vendor_id);
+	if (ret)
+		return ret;
+
+	ret = gb_interface_dme_attr_get(intf, DME_TOSHIBA_GMP_PID,
+					&intf->product_id);
+	if (ret)
+		return ret;
+
+	ret = gb_interface_dme_attr_get(intf, DME_TOSHIBA_GMP_SN0, &sn0);
+	if (ret)
+		return ret;
+
+	ret = gb_interface_dme_attr_get(intf, DME_TOSHIBA_GMP_SN1, &sn1);
+	if (ret)
+		return ret;
+
+	intf->serial_number = (u64)sn1 << 32 | sn0;
+
+	return 0;
+}
+
+static int gb_interface_read_dme(struct gb_interface *intf)
+{
+	int ret;
+
+	/* DME attributes have already been read */
+	if (intf->dme_read)
+		return 0;
+
+	ret = gb_interface_dme_attr_get(intf, DME_DDBL1_MANUFACTURERID,
+					&intf->ddbl1_manufacturer_id);
+	if (ret)
+		return ret;
+
+	ret = gb_interface_dme_attr_get(intf, DME_DDBL1_PRODUCTID,
+					&intf->ddbl1_product_id);
+	if (ret)
+		return ret;
+
+	if (intf->ddbl1_manufacturer_id == TOSHIBA_DMID &&
+	    intf->ddbl1_product_id == TOSHIBA_ES2_BRIDGE_DPID) {
+		intf->quirks |= GB_INTERFACE_QUIRK_NO_GMP_IDS;
+		intf->quirks |= GB_INTERFACE_QUIRK_NO_INIT_STATUS;
+	}
+
+	ret = gb_interface_read_ara_dme(intf);
+	if (ret)
+		return ret;
+
+	intf->dme_read = true;
+
+	return 0;
+}
+
+static int gb_interface_route_create(struct gb_interface *intf)
+{
+	struct gb_svc *svc = intf->hd->svc;
+	u8 intf_id = intf->interface_id;
+	u8 device_id;
+	int ret;
+
+	/* Allocate an interface device id. */
+	ret = ida_alloc_range(&svc->device_id_map, GB_SVC_DEVICE_ID_MIN,
+			      GB_SVC_DEVICE_ID_MAX, GFP_KERNEL);
+	if (ret < 0) {
+		dev_err(&intf->dev, "failed to allocate device id: %d\n", ret);
+		return ret;
+	}
+	device_id = ret;
+
+	ret = gb_svc_intf_device_id(svc, intf_id, device_id);
+	if (ret) {
+		dev_err(&intf->dev, "failed to set device id %u: %d\n",
+			device_id, ret);
+		goto err_ida_remove;
+	}
+
+	/* FIXME: Hard-coded AP device id. */
+	ret = gb_svc_route_create(svc, svc->ap_intf_id, GB_SVC_DEVICE_ID_AP,
+				  intf_id, device_id);
+	if (ret) {
+		dev_err(&intf->dev, "failed to create route: %d\n", ret);
+		goto err_svc_id_free;
+	}
+
+	intf->device_id = device_id;
+
+	return 0;
+
+err_svc_id_free:
+	/*
+	 * XXX Should we tell SVC that this id doesn't belong to interface
+	 * XXX anymore.
+	 */
+err_ida_remove:
+	ida_free(&svc->device_id_map, device_id);
+
+	return ret;
+}
+
+static void gb_interface_route_destroy(struct gb_interface *intf)
+{
+	struct gb_svc *svc = intf->hd->svc;
+
+	if (intf->device_id == GB_INTERFACE_DEVICE_ID_BAD)
+		return;
+
+	gb_svc_route_destroy(svc, svc->ap_intf_id, intf->interface_id);
+	ida_free(&svc->device_id_map, intf->device_id);
+	intf->device_id = GB_INTERFACE_DEVICE_ID_BAD;
+}
+
+/* Locking: Caller holds the interface mutex. */
+static int gb_interface_legacy_mode_switch(struct gb_interface *intf)
+{
+	int ret;
+
+	dev_info(&intf->dev, "legacy mode switch detected\n");
+
+	/* Mark as disconnected to prevent I/O during disable. */
+	intf->disconnected = true;
+	gb_interface_disable(intf);
+	intf->disconnected = false;
+
+	ret = gb_interface_enable(intf);
+	if (ret) {
+		dev_err(&intf->dev, "failed to re-enable interface: %d\n", ret);
+		gb_interface_deactivate(intf);
+	}
+
+	return ret;
+}
+
+void gb_interface_mailbox_event(struct gb_interface *intf, u16 result,
+				u32 mailbox)
+{
+	mutex_lock(&intf->mutex);
+
+	if (result) {
+		dev_warn(&intf->dev,
+			 "mailbox event with UniPro error: 0x%04x\n",
+			 result);
+		goto err_disable;
+	}
+
+	if (mailbox != GB_SVC_INTF_MAILBOX_GREYBUS) {
+		dev_warn(&intf->dev,
+			 "mailbox event with unexpected value: 0x%08x\n",
+			 mailbox);
+		goto err_disable;
+	}
+
+	if (intf->quirks & GB_INTERFACE_QUIRK_LEGACY_MODE_SWITCH) {
+		gb_interface_legacy_mode_switch(intf);
+		goto out_unlock;
+	}
+
+	if (!intf->mode_switch) {
+		dev_warn(&intf->dev, "unexpected mailbox event: 0x%08x\n",
+			 mailbox);
+		goto err_disable;
+	}
+
+	dev_info(&intf->dev, "mode switch detected\n");
+
+	complete(&intf->mode_switch_completion);
+
+out_unlock:
+	mutex_unlock(&intf->mutex);
+
+	return;
+
+err_disable:
+	gb_interface_disable(intf);
+	gb_interface_deactivate(intf);
+	mutex_unlock(&intf->mutex);
+}
+
+static void gb_interface_mode_switch_work(struct work_struct *work)
+{
+	struct gb_interface *intf;
+	struct gb_control *control;
+	unsigned long timeout;
+	int ret;
+
+	intf = container_of(work, struct gb_interface, mode_switch_work);
+
+	mutex_lock(&intf->mutex);
+	/* Make sure interface is still enabled. */
+	if (!intf->enabled) {
+		dev_dbg(&intf->dev, "mode switch aborted\n");
+		intf->mode_switch = false;
+		mutex_unlock(&intf->mutex);
+		goto out_interface_put;
+	}
+
+	/*
+	 * Prepare the control device for mode switch and make sure to get an
+	 * extra reference before it goes away during interface disable.
+	 */
+	control = gb_control_get(intf->control);
+	gb_control_mode_switch_prepare(control);
+	gb_interface_disable(intf);
+	mutex_unlock(&intf->mutex);
+
+	timeout = msecs_to_jiffies(GB_INTERFACE_MODE_SWITCH_TIMEOUT);
+	ret = wait_for_completion_interruptible_timeout(
+			&intf->mode_switch_completion, timeout);
+
+	/* Finalise control-connection mode switch. */
+	gb_control_mode_switch_complete(control);
+	gb_control_put(control);
+
+	if (ret < 0) {
+		dev_err(&intf->dev, "mode switch interrupted\n");
+		goto err_deactivate;
+	} else if (ret == 0) {
+		dev_err(&intf->dev, "mode switch timed out\n");
+		goto err_deactivate;
+	}
+
+	/* Re-enable (re-enumerate) interface if still active. */
+	mutex_lock(&intf->mutex);
+	intf->mode_switch = false;
+	if (intf->active) {
+		ret = gb_interface_enable(intf);
+		if (ret) {
+			dev_err(&intf->dev, "failed to re-enable interface: %d\n",
+				ret);
+			gb_interface_deactivate(intf);
+		}
+	}
+	mutex_unlock(&intf->mutex);
+
+out_interface_put:
+	gb_interface_put(intf);
+
+	return;
+
+err_deactivate:
+	mutex_lock(&intf->mutex);
+	intf->mode_switch = false;
+	gb_interface_deactivate(intf);
+	mutex_unlock(&intf->mutex);
+
+	gb_interface_put(intf);
+}
+
+int gb_interface_request_mode_switch(struct gb_interface *intf)
+{
+	int ret = 0;
+
+	mutex_lock(&intf->mutex);
+	if (intf->mode_switch) {
+		ret = -EBUSY;
+		goto out_unlock;
+	}
+
+	intf->mode_switch = true;
+	reinit_completion(&intf->mode_switch_completion);
+
+	/*
+	 * Get a reference to the interface device, which will be put once the
+	 * mode switch is complete.
+	 */
+	get_device(&intf->dev);
+
+	if (!queue_work(system_long_wq, &intf->mode_switch_work)) {
+		put_device(&intf->dev);
+		ret = -EBUSY;
+		goto out_unlock;
+	}
+
+out_unlock:
+	mutex_unlock(&intf->mutex);
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(gb_interface_request_mode_switch);
+
+/*
+ * T_TstSrcIncrement is written by the module on ES2 as a stand-in for the
+ * init-status attribute DME_TOSHIBA_INIT_STATUS. The AP needs to read and
+ * clear it after reading a non-zero value from it.
+ *
+ * FIXME: This is module-hardware dependent and needs to be extended for every
+ * type of module we want to support.
+ */
+static int gb_interface_read_and_clear_init_status(struct gb_interface *intf)
+{
+	struct gb_host_device *hd = intf->hd;
+	unsigned long bootrom_quirks;
+	unsigned long s2l_quirks;
+	int ret;
+	u32 value;
+	u16 attr;
+	u8 init_status;
+
+	/*
+	 * ES2 bridges use T_TstSrcIncrement for the init status.
+	 *
+	 * FIXME: Remove ES2 support
+	 */
+	if (intf->quirks & GB_INTERFACE_QUIRK_NO_INIT_STATUS)
+		attr = DME_T_TST_SRC_INCREMENT;
+	else
+		attr = DME_TOSHIBA_GMP_INIT_STATUS;
+
+	ret = gb_svc_dme_peer_get(hd->svc, intf->interface_id, attr,
+				  DME_SELECTOR_INDEX_NULL, &value);
+	if (ret)
+		return ret;
+
+	/*
+	 * A nonzero init status indicates the module has finished
+	 * initializing.
+	 */
+	if (!value) {
+		dev_err(&intf->dev, "invalid init status\n");
+		return -ENODEV;
+	}
+
+	/*
+	 * Extract the init status.
+	 *
+	 * For ES2: We need to check lowest 8 bits of 'value'.
+	 * For ES3: We need to check highest 8 bits out of 32 of 'value'.
+	 *
+	 * FIXME: Remove ES2 support
+	 */
+	if (intf->quirks & GB_INTERFACE_QUIRK_NO_INIT_STATUS)
+		init_status = value & 0xff;
+	else
+		init_status = value >> 24;
+
+	/*
+	 * Check if the interface is executing the quirky ES3 bootrom that,
+	 * for example, requires E2EFC, CSD and CSV to be disabled.
+	 */
+	bootrom_quirks = GB_INTERFACE_QUIRK_NO_CPORT_FEATURES |
+				GB_INTERFACE_QUIRK_FORCED_DISABLE |
+				GB_INTERFACE_QUIRK_LEGACY_MODE_SWITCH |
+				GB_INTERFACE_QUIRK_NO_BUNDLE_ACTIVATE;
+
+	s2l_quirks = GB_INTERFACE_QUIRK_NO_PM;
+
+	switch (init_status) {
+	case GB_INIT_BOOTROM_UNIPRO_BOOT_STARTED:
+	case GB_INIT_BOOTROM_FALLBACK_UNIPRO_BOOT_STARTED:
+		intf->quirks |= bootrom_quirks;
+		break;
+	case GB_INIT_S2_LOADER_BOOT_STARTED:
+		/* S2 Loader doesn't support runtime PM */
+		intf->quirks &= ~bootrom_quirks;
+		intf->quirks |= s2l_quirks;
+		break;
+	default:
+		intf->quirks &= ~bootrom_quirks;
+		intf->quirks &= ~s2l_quirks;
+	}
+
+	/* Clear the init status. */
+	return gb_svc_dme_peer_set(hd->svc, intf->interface_id, attr,
+				   DME_SELECTOR_INDEX_NULL, 0);
+}
+
+/* interface sysfs attributes */
+#define gb_interface_attr(field, type)					\
+static ssize_t field##_show(struct device *dev,				\
+			    struct device_attribute *attr,		\
+			    char *buf)					\
+{									\
+	struct gb_interface *intf = to_gb_interface(dev);		\
+	return scnprintf(buf, PAGE_SIZE, type"\n", intf->field);	\
+}									\
+static DEVICE_ATTR_RO(field)
+
+gb_interface_attr(ddbl1_manufacturer_id, "0x%08x");
+gb_interface_attr(ddbl1_product_id, "0x%08x");
+gb_interface_attr(interface_id, "%u");
+gb_interface_attr(vendor_id, "0x%08x");
+gb_interface_attr(product_id, "0x%08x");
+gb_interface_attr(serial_number, "0x%016llx");
+
+static ssize_t voltage_now_show(struct device *dev,
+				struct device_attribute *attr, char *buf)
+{
+	struct gb_interface *intf = to_gb_interface(dev);
+	int ret;
+	u32 measurement;
+
+	ret = gb_svc_pwrmon_intf_sample_get(intf->hd->svc, intf->interface_id,
+					    GB_SVC_PWRMON_TYPE_VOL,
+					    &measurement);
+	if (ret) {
+		dev_err(&intf->dev, "failed to get voltage sample (%d)\n", ret);
+		return ret;
+	}
+
+	return sprintf(buf, "%u\n", measurement);
+}
+static DEVICE_ATTR_RO(voltage_now);
+
+static ssize_t current_now_show(struct device *dev,
+				struct device_attribute *attr, char *buf)
+{
+	struct gb_interface *intf = to_gb_interface(dev);
+	int ret;
+	u32 measurement;
+
+	ret = gb_svc_pwrmon_intf_sample_get(intf->hd->svc, intf->interface_id,
+					    GB_SVC_PWRMON_TYPE_CURR,
+					    &measurement);
+	if (ret) {
+		dev_err(&intf->dev, "failed to get current sample (%d)\n", ret);
+		return ret;
+	}
+
+	return sprintf(buf, "%u\n", measurement);
+}
+static DEVICE_ATTR_RO(current_now);
+
+static ssize_t power_now_show(struct device *dev,
+			      struct device_attribute *attr, char *buf)
+{
+	struct gb_interface *intf = to_gb_interface(dev);
+	int ret;
+	u32 measurement;
+
+	ret = gb_svc_pwrmon_intf_sample_get(intf->hd->svc, intf->interface_id,
+					    GB_SVC_PWRMON_TYPE_PWR,
+					    &measurement);
+	if (ret) {
+		dev_err(&intf->dev, "failed to get power sample (%d)\n", ret);
+		return ret;
+	}
+
+	return sprintf(buf, "%u\n", measurement);
+}
+static DEVICE_ATTR_RO(power_now);
+
+static ssize_t power_state_show(struct device *dev,
+				struct device_attribute *attr, char *buf)
+{
+	struct gb_interface *intf = to_gb_interface(dev);
+
+	if (intf->active)
+		return scnprintf(buf, PAGE_SIZE, "on\n");
+	else
+		return scnprintf(buf, PAGE_SIZE, "off\n");
+}
+
+static ssize_t power_state_store(struct device *dev,
+				 struct device_attribute *attr, const char *buf,
+				 size_t len)
+{
+	struct gb_interface *intf = to_gb_interface(dev);
+	bool activate;
+	int ret = 0;
+
+	if (kstrtobool(buf, &activate))
+		return -EINVAL;
+
+	mutex_lock(&intf->mutex);
+
+	if (activate == intf->active)
+		goto unlock;
+
+	if (activate) {
+		ret = gb_interface_activate(intf);
+		if (ret) {
+			dev_err(&intf->dev,
+				"failed to activate interface: %d\n", ret);
+			goto unlock;
+		}
+
+		ret = gb_interface_enable(intf);
+		if (ret) {
+			dev_err(&intf->dev,
+				"failed to enable interface: %d\n", ret);
+			gb_interface_deactivate(intf);
+			goto unlock;
+		}
+	} else {
+		gb_interface_disable(intf);
+		gb_interface_deactivate(intf);
+	}
+
+unlock:
+	mutex_unlock(&intf->mutex);
+
+	if (ret)
+		return ret;
+
+	return len;
+}
+static DEVICE_ATTR_RW(power_state);
+
+static const char *gb_interface_type_string(struct gb_interface *intf)
+{
+	static const char * const types[] = {
+		[GB_INTERFACE_TYPE_INVALID] = "invalid",
+		[GB_INTERFACE_TYPE_UNKNOWN] = "unknown",
+		[GB_INTERFACE_TYPE_DUMMY] = "dummy",
+		[GB_INTERFACE_TYPE_UNIPRO] = "unipro",
+		[GB_INTERFACE_TYPE_GREYBUS] = "greybus",
+	};
+
+	return types[intf->type];
+}
+
+static ssize_t interface_type_show(struct device *dev,
+				   struct device_attribute *attr, char *buf)
+{
+	struct gb_interface *intf = to_gb_interface(dev);
+
+	return sprintf(buf, "%s\n", gb_interface_type_string(intf));
+}
+static DEVICE_ATTR_RO(interface_type);
+
+static struct attribute *interface_unipro_attrs[] = {
+	&dev_attr_ddbl1_manufacturer_id.attr,
+	&dev_attr_ddbl1_product_id.attr,
+	NULL
+};
+
+static struct attribute *interface_greybus_attrs[] = {
+	&dev_attr_vendor_id.attr,
+	&dev_attr_product_id.attr,
+	&dev_attr_serial_number.attr,
+	NULL
+};
+
+static struct attribute *interface_power_attrs[] = {
+	&dev_attr_voltage_now.attr,
+	&dev_attr_current_now.attr,
+	&dev_attr_power_now.attr,
+	&dev_attr_power_state.attr,
+	NULL
+};
+
+static struct attribute *interface_common_attrs[] = {
+	&dev_attr_interface_id.attr,
+	&dev_attr_interface_type.attr,
+	NULL
+};
+
+static umode_t interface_unipro_is_visible(struct kobject *kobj,
+					   struct attribute *attr, int n)
+{
+	struct device *dev = kobj_to_dev(kobj);
+	struct gb_interface *intf = to_gb_interface(dev);
+
+	switch (intf->type) {
+	case GB_INTERFACE_TYPE_UNIPRO:
+	case GB_INTERFACE_TYPE_GREYBUS:
+		return attr->mode;
+	default:
+		return 0;
+	}
+}
+
+static umode_t interface_greybus_is_visible(struct kobject *kobj,
+					    struct attribute *attr, int n)
+{
+	struct device *dev = kobj_to_dev(kobj);
+	struct gb_interface *intf = to_gb_interface(dev);
+
+	switch (intf->type) {
+	case GB_INTERFACE_TYPE_GREYBUS:
+		return attr->mode;
+	default:
+		return 0;
+	}
+}
+
+static umode_t interface_power_is_visible(struct kobject *kobj,
+					  struct attribute *attr, int n)
+{
+	struct device *dev = kobj_to_dev(kobj);
+	struct gb_interface *intf = to_gb_interface(dev);
+
+	switch (intf->type) {
+	case GB_INTERFACE_TYPE_UNIPRO:
+	case GB_INTERFACE_TYPE_GREYBUS:
+		return attr->mode;
+	default:
+		return 0;
+	}
+}
+
+static const struct attribute_group interface_unipro_group = {
+	.is_visible	= interface_unipro_is_visible,
+	.attrs		= interface_unipro_attrs,
+};
+
+static const struct attribute_group interface_greybus_group = {
+	.is_visible	= interface_greybus_is_visible,
+	.attrs		= interface_greybus_attrs,
+};
+
+static const struct attribute_group interface_power_group = {
+	.is_visible	= interface_power_is_visible,
+	.attrs		= interface_power_attrs,
+};
+
+static const struct attribute_group interface_common_group = {
+	.attrs		= interface_common_attrs,
+};
+
+static const struct attribute_group *interface_groups[] = {
+	&interface_unipro_group,
+	&interface_greybus_group,
+	&interface_power_group,
+	&interface_common_group,
+	NULL
+};
+
+static void gb_interface_release(struct device *dev)
+{
+	struct gb_interface *intf = to_gb_interface(dev);
+
+	trace_gb_interface_release(intf);
+
+	cancel_work_sync(&intf->mode_switch_work);
+	kfree(intf);
+}
+
+#ifdef CONFIG_PM
+static int gb_interface_suspend(struct device *dev)
+{
+	struct gb_interface *intf = to_gb_interface(dev);
+	int ret;
+
+	ret = gb_control_interface_suspend_prepare(intf->control);
+	if (ret)
+		return ret;
+
+	ret = gb_control_suspend(intf->control);
+	if (ret)
+		goto err_hibernate_abort;
+
+	ret = gb_interface_hibernate_link(intf);
+	if (ret)
+		return ret;
+
+	/* Delay to allow interface to enter standby before disabling refclk */
+	msleep(GB_INTERFACE_SUSPEND_HIBERNATE_DELAY_MS);
+
+	ret = gb_interface_refclk_set(intf, false);
+	if (ret)
+		return ret;
+
+	return 0;
+
+err_hibernate_abort:
+	gb_control_interface_hibernate_abort(intf->control);
+
+	return ret;
+}
+
+static int gb_interface_resume(struct device *dev)
+{
+	struct gb_interface *intf = to_gb_interface(dev);
+	struct gb_svc *svc = intf->hd->svc;
+	int ret;
+
+	ret = gb_interface_refclk_set(intf, true);
+	if (ret)
+		return ret;
+
+	ret = gb_svc_intf_resume(svc, intf->interface_id);
+	if (ret)
+		return ret;
+
+	ret = gb_control_resume(intf->control);
+	if (ret)
+		return ret;
+
+	return 0;
+}
+
+static int gb_interface_runtime_idle(struct device *dev)
+{
+	pm_runtime_mark_last_busy(dev);
+	pm_request_autosuspend(dev);
+
+	return 0;
+}
+#endif
+
+static const struct dev_pm_ops gb_interface_pm_ops = {
+	SET_RUNTIME_PM_OPS(gb_interface_suspend, gb_interface_resume,
+			   gb_interface_runtime_idle)
+};
+
+const struct device_type greybus_interface_type = {
+	.name =		"greybus_interface",
+	.release =	gb_interface_release,
+	.pm =		&gb_interface_pm_ops,
+};
+
+/*
+ * A Greybus module represents a user-replaceable component on a GMP
+ * phone.  An interface is the physical connection on that module.  A
+ * module may have more than one interface.
+ *
+ * Create a gb_interface structure to represent a discovered interface.
+ * The position of interface within the Endo is encoded in "interface_id"
+ * argument.
+ *
+ * Returns a pointer to the new interface or a null pointer if a
+ * failure occurs due to memory exhaustion.
+ */
+struct gb_interface *gb_interface_create(struct gb_module *module,
+					 u8 interface_id)
+{
+	struct gb_host_device *hd = module->hd;
+	struct gb_interface *intf;
+
+	intf = kzalloc_obj(*intf);
+	if (!intf)
+		return NULL;
+
+	intf->hd = hd;		/* XXX refcount? */
+	intf->module = module;
+	intf->interface_id = interface_id;
+	INIT_LIST_HEAD(&intf->bundles);
+	INIT_LIST_HEAD(&intf->manifest_descs);
+	mutex_init(&intf->mutex);
+	INIT_WORK(&intf->mode_switch_work, gb_interface_mode_switch_work);
+	init_completion(&intf->mode_switch_completion);
+
+	/* Invalid device id to start with */
+	intf->device_id = GB_INTERFACE_DEVICE_ID_BAD;
+
+	intf->dev.parent = &module->dev;
+	intf->dev.bus = &greybus_bus_type;
+	intf->dev.type = &greybus_interface_type;
+	intf->dev.groups = interface_groups;
+	intf->dev.dma_mask = module->dev.dma_mask;
+	device_initialize(&intf->dev);
+	dev_set_name(&intf->dev, "%s.%u", dev_name(&module->dev),
+		     interface_id);
+
+	pm_runtime_set_autosuspend_delay(&intf->dev,
+					 GB_INTERFACE_AUTOSUSPEND_MS);
+
+	trace_gb_interface_create(intf);
+
+	return intf;
+}
+
+static int gb_interface_vsys_set(struct gb_interface *intf, bool enable)
+{
+	struct gb_svc *svc = intf->hd->svc;
+	int ret;
+
+	dev_dbg(&intf->dev, "%s - %d\n", __func__, enable);
+
+	ret = gb_svc_intf_vsys_set(svc, intf->interface_id, enable);
+	if (ret) {
+		dev_err(&intf->dev, "failed to set v_sys: %d\n", ret);
+		return ret;
+	}
+
+	return 0;
+}
+
+static int gb_interface_refclk_set(struct gb_interface *intf, bool enable)
+{
+	struct gb_svc *svc = intf->hd->svc;
+	int ret;
+
+	dev_dbg(&intf->dev, "%s - %d\n", __func__, enable);
+
+	ret = gb_svc_intf_refclk_set(svc, intf->interface_id, enable);
+	if (ret) {
+		dev_err(&intf->dev, "failed to set refclk: %d\n", ret);
+		return ret;
+	}
+
+	return 0;
+}
+
+static int gb_interface_unipro_set(struct gb_interface *intf, bool enable)
+{
+	struct gb_svc *svc = intf->hd->svc;
+	int ret;
+
+	dev_dbg(&intf->dev, "%s - %d\n", __func__, enable);
+
+	ret = gb_svc_intf_unipro_set(svc, intf->interface_id, enable);
+	if (ret) {
+		dev_err(&intf->dev, "failed to set UniPro: %d\n", ret);
+		return ret;
+	}
+
+	return 0;
+}
+
+static int gb_interface_activate_operation(struct gb_interface *intf,
+					   enum gb_interface_type *intf_type)
+{
+	struct gb_svc *svc = intf->hd->svc;
+	u8 type;
+	int ret;
+
+	dev_dbg(&intf->dev, "%s\n", __func__);
+
+	ret = gb_svc_intf_activate(svc, intf->interface_id, &type);
+	if (ret) {
+		dev_err(&intf->dev, "failed to activate: %d\n", ret);
+		return ret;
+	}
+
+	switch (type) {
+	case GB_SVC_INTF_TYPE_DUMMY:
+		*intf_type = GB_INTERFACE_TYPE_DUMMY;
+		/* FIXME: handle as an error for now */
+		return -ENODEV;
+	case GB_SVC_INTF_TYPE_UNIPRO:
+		*intf_type = GB_INTERFACE_TYPE_UNIPRO;
+		dev_err(&intf->dev, "interface type UniPro not supported\n");
+		/* FIXME: handle as an error for now */
+		return -ENODEV;
+	case GB_SVC_INTF_TYPE_GREYBUS:
+		*intf_type = GB_INTERFACE_TYPE_GREYBUS;
+		break;
+	default:
+		dev_err(&intf->dev, "unknown interface type: %u\n", type);
+		*intf_type = GB_INTERFACE_TYPE_UNKNOWN;
+		return -ENODEV;
+	}
+
+	return 0;
+}
+
+static int gb_interface_hibernate_link(struct gb_interface *intf)
+{
+	struct gb_svc *svc = intf->hd->svc;
+
+	return gb_svc_intf_set_power_mode_hibernate(svc, intf->interface_id);
+}
+
+static int _gb_interface_activate(struct gb_interface *intf,
+				  enum gb_interface_type *type)
+{
+	int ret;
+
+	*type = GB_INTERFACE_TYPE_UNKNOWN;
+
+	if (intf->ejected || intf->removed)
+		return -ENODEV;
+
+	ret = gb_interface_vsys_set(intf, true);
+	if (ret)
+		return ret;
+
+	ret = gb_interface_refclk_set(intf, true);
+	if (ret)
+		goto err_vsys_disable;
+
+	ret = gb_interface_unipro_set(intf, true);
+	if (ret)
+		goto err_refclk_disable;
+
+	ret = gb_interface_activate_operation(intf, type);
+	if (ret) {
+		switch (*type) {
+		case GB_INTERFACE_TYPE_UNIPRO:
+		case GB_INTERFACE_TYPE_GREYBUS:
+			goto err_hibernate_link;
+		default:
+			goto err_unipro_disable;
+		}
+	}
+
+	ret = gb_interface_read_dme(intf);
+	if (ret)
+		goto err_hibernate_link;
+
+	ret = gb_interface_route_create(intf);
+	if (ret)
+		goto err_hibernate_link;
+
+	intf->active = true;
+
+	trace_gb_interface_activate(intf);
+
+	return 0;
+
+err_hibernate_link:
+	gb_interface_hibernate_link(intf);
+err_unipro_disable:
+	gb_interface_unipro_set(intf, false);
+err_refclk_disable:
+	gb_interface_refclk_set(intf, false);
+err_vsys_disable:
+	gb_interface_vsys_set(intf, false);
+
+	return ret;
+}
+
+/*
+ * At present, we assume a UniPro-only module to be a Greybus module that
+ * failed to send its mailbox poke. There is some reason to believe that this
+ * is because of a bug in the ES3 bootrom.
+ *
+ * FIXME: Check if this is a Toshiba bridge before retrying?
+ */
+static int _gb_interface_activate_es3_hack(struct gb_interface *intf,
+					   enum gb_interface_type *type)
+{
+	int retries = 3;
+	int ret;
+
+	while (retries--) {
+		ret = _gb_interface_activate(intf, type);
+		if (ret == -ENODEV && *type == GB_INTERFACE_TYPE_UNIPRO)
+			continue;
+
+		break;
+	}
+
+	return ret;
+}
+
+/*
+ * Activate an interface.
+ *
+ * Locking: Caller holds the interface mutex.
+ */
+int gb_interface_activate(struct gb_interface *intf)
+{
+	enum gb_interface_type type;
+	int ret;
+
+	switch (intf->type) {
+	case GB_INTERFACE_TYPE_INVALID:
+	case GB_INTERFACE_TYPE_GREYBUS:
+		ret = _gb_interface_activate_es3_hack(intf, &type);
+		break;
+	default:
+		ret = _gb_interface_activate(intf, &type);
+	}
+
+	/* Make sure type is detected correctly during reactivation. */
+	if (intf->type != GB_INTERFACE_TYPE_INVALID) {
+		if (type != intf->type) {
+			dev_err(&intf->dev, "failed to detect interface type\n");
+
+			if (!ret)
+				gb_interface_deactivate(intf);
+
+			return -EIO;
+		}
+	} else {
+		intf->type = type;
+	}
+
+	return ret;
+}
+
+/*
+ * Deactivate an interface.
+ *
+ * Locking: Caller holds the interface mutex.
+ */
+void gb_interface_deactivate(struct gb_interface *intf)
+{
+	if (!intf->active)
+		return;
+
+	trace_gb_interface_deactivate(intf);
+
+	/* Abort any ongoing mode switch. */
+	if (intf->mode_switch)
+		complete(&intf->mode_switch_completion);
+
+	gb_interface_route_destroy(intf);
+	gb_interface_hibernate_link(intf);
+	gb_interface_unipro_set(intf, false);
+	gb_interface_refclk_set(intf, false);
+	gb_interface_vsys_set(intf, false);
+
+	intf->active = false;
+}
+
+/*
+ * Enable an interface by enabling its control connection, fetching the
+ * manifest and other information over it, and finally registering its child
+ * devices.
+ *
+ * Locking: Caller holds the interface mutex.
+ */
+int gb_interface_enable(struct gb_interface *intf)
+{
+	struct gb_control *control;
+	struct gb_bundle *bundle, *tmp;
+	int ret, size;
+	void *manifest;
+
+	ret = gb_interface_read_and_clear_init_status(intf);
+	if (ret) {
+		dev_err(&intf->dev, "failed to clear init status: %d\n", ret);
+		return ret;
+	}
+
+	/* Establish control connection */
+	control = gb_control_create(intf);
+	if (IS_ERR(control)) {
+		dev_err(&intf->dev, "failed to create control device: %ld\n",
+			PTR_ERR(control));
+		return PTR_ERR(control);
+	}
+	intf->control = control;
+
+	ret = gb_control_enable(intf->control);
+	if (ret)
+		goto err_put_control;
+
+	/* Get manifest size using control protocol on CPort */
+	size = gb_control_get_manifest_size_operation(intf);
+	if (size <= 0) {
+		dev_err(&intf->dev, "failed to get manifest size: %d\n", size);
+
+		if (size)
+			ret = size;
+		else
+			ret =  -EINVAL;
+
+		goto err_disable_control;
+	}
+
+	manifest = kmalloc(size, GFP_KERNEL);
+	if (!manifest) {
+		ret = -ENOMEM;
+		goto err_disable_control;
+	}
+
+	/* Get manifest using control protocol on CPort */
+	ret = gb_control_get_manifest_operation(intf, manifest, size);
+	if (ret) {
+		dev_err(&intf->dev, "failed to get manifest: %d\n", ret);
+		goto err_free_manifest;
+	}
+
+	/*
+	 * Parse the manifest and build up our data structures representing
+	 * what's in it.
+	 */
+	if (!gb_manifest_parse(intf, manifest, size)) {
+		dev_err(&intf->dev, "failed to parse manifest\n");
+		ret = -EINVAL;
+		goto err_destroy_bundles;
+	}
+
+	ret = gb_control_get_bundle_versions(intf->control);
+	if (ret)
+		goto err_destroy_bundles;
+
+	/* Register the control device and any bundles */
+	ret = gb_control_add(intf->control);
+	if (ret)
+		goto err_destroy_bundles;
+
+	pm_runtime_use_autosuspend(&intf->dev);
+	pm_runtime_get_noresume(&intf->dev);
+	pm_runtime_set_active(&intf->dev);
+	pm_runtime_enable(&intf->dev);
+
+	list_for_each_entry_safe_reverse(bundle, tmp, &intf->bundles, links) {
+		ret = gb_bundle_add(bundle);
+		if (ret) {
+			gb_bundle_destroy(bundle);
+			continue;
+		}
+	}
+
+	kfree(manifest);
+
+	intf->enabled = true;
+
+	pm_runtime_put(&intf->dev);
+
+	trace_gb_interface_enable(intf);
+
+	return 0;
+
+err_destroy_bundles:
+	list_for_each_entry_safe(bundle, tmp, &intf->bundles, links)
+		gb_bundle_destroy(bundle);
+err_free_manifest:
+	kfree(manifest);
+err_disable_control:
+	gb_control_disable(intf->control);
+err_put_control:
+	gb_control_put(intf->control);
+	intf->control = NULL;
+
+	return ret;
+}
+
+/*
+ * Disable an interface and destroy its bundles.
+ *
+ * Locking: Caller holds the interface mutex.
+ */
+void gb_interface_disable(struct gb_interface *intf)
+{
+	struct gb_bundle *bundle;
+	struct gb_bundle *next;
+
+	if (!intf->enabled)
+		return;
+
+	trace_gb_interface_disable(intf);
+
+	pm_runtime_get_sync(&intf->dev);
+
+	/* Set disconnected flag to avoid I/O during connection tear down. */
+	if (intf->quirks & GB_INTERFACE_QUIRK_FORCED_DISABLE)
+		intf->disconnected = true;
+
+	list_for_each_entry_safe(bundle, next, &intf->bundles, links)
+		gb_bundle_destroy(bundle);
+
+	if (!intf->mode_switch && !intf->disconnected)
+		gb_control_interface_deactivate_prepare(intf->control);
+
+	gb_control_del(intf->control);
+	gb_control_disable(intf->control);
+	gb_control_put(intf->control);
+	intf->control = NULL;
+
+	intf->enabled = false;
+
+	pm_runtime_disable(&intf->dev);
+	pm_runtime_set_suspended(&intf->dev);
+	pm_runtime_dont_use_autosuspend(&intf->dev);
+	pm_runtime_put_noidle(&intf->dev);
+}
+
+/* Register an interface. */
+int gb_interface_add(struct gb_interface *intf)
+{
+	int ret;
+
+	ret = device_add(&intf->dev);
+	if (ret) {
+		dev_err(&intf->dev, "failed to register interface: %d\n", ret);
+		return ret;
+	}
+
+	trace_gb_interface_add(intf);
+
+	dev_info(&intf->dev, "Interface added (%s)\n",
+		 gb_interface_type_string(intf));
+
+	switch (intf->type) {
+	case GB_INTERFACE_TYPE_GREYBUS:
+		dev_info(&intf->dev, "GMP VID=0x%08x, PID=0x%08x\n",
+			 intf->vendor_id, intf->product_id);
+		fallthrough;
+	case GB_INTERFACE_TYPE_UNIPRO:
+		dev_info(&intf->dev, "DDBL1 Manufacturer=0x%08x, Product=0x%08x\n",
+			 intf->ddbl1_manufacturer_id,
+			 intf->ddbl1_product_id);
+		break;
+	default:
+		break;
+	}
+
+	return 0;
+}
+
+/* Deregister an interface. */
+void gb_interface_del(struct gb_interface *intf)
+{
+	if (device_is_registered(&intf->dev)) {
+		trace_gb_interface_del(intf);
+
+		device_del(&intf->dev);
+		dev_info(&intf->dev, "Interface removed\n");
+	}
+}
+
+void gb_interface_put(struct gb_interface *intf)
+{
+	put_device(&intf->dev);
+}