[PATCH v2 0/5] drivers/fsi: Add SBEFIFO and OCC drivers

[PATCH v2 0/5] drivers/fsi: Add SBEFIFO and OCC drivers

[Eddie James]

From: "Edward A. James" <eajames@us.ibm.com>

This series adds two FSI-based device drivers. The OCC driver is dependent on
the SBEFIFO driver, as a user of it's in-kernel API. The in-kernel API provided
by the OCC driver will be used by a hwmon driver (on the lkml as "Add On-Chip
Controller (OCC) hwmon driver").

Changes since v1:
 * Split bindings into separate patch and added SBEFIFO device binding
 * Fixed #includes
 * Fix SBEFIFO race condition between write() and poll_timer().
 * Followed Rob's suggestion to just create one platform device for hwmon
   driver, instead of using the device tree.
 * Also check for "command in progress" response from OCC and try a while

Edward A. James (5):
  drivers/fsi: Add SBEFIFO FSI client device driver
  drivers/fsi/sbefifo: Add in-kernel API
  drivers/fsi: Add On-Chip Controller (OCC) driver
  drivers/fsi/occ: Add in-kernel API
  Documentation/devicetree/bindings: Add FSI device documentation

 .../devicetree/bindings/fsi/ibm,p9-occ.txt         |  15 +
 .../devicetree/bindings/fsi/ibm,p9-sbefifo.txt     |  20 +
 drivers/fsi/Kconfig                                |  17 +
 drivers/fsi/Makefile                               |   2 +
 drivers/fsi/fsi-sbefifo.c                          | 932 +++++++++++++++++++++
 drivers/fsi/occ.c                                  | 803 ++++++++++++++++++
 include/linux/fsi-sbefifo.h                        |  30 +
 include/linux/occ.h                                |  41 +
 8 files changed, 1860 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/fsi/ibm,p9-occ.txt
 create mode 100644 Documentation/devicetree/bindings/fsi/ibm,p9-sbefifo.txt
 create mode 100644 drivers/fsi/fsi-sbefifo.c
 create mode 100644 drivers/fsi/occ.c
 create mode 100644 include/linux/fsi-sbefifo.h
 create mode 100644 include/linux/occ.h

-- 
1.8.3.1

[PATCH v2 1/5] drivers/fsi: Add SBEFIFO FSI client device driver

[Eddie James]

From: "Edward A. James" <eajames@us.ibm.com>

IBM POWER9 processors contain some embedded hardware and software bits
collectively referred to as the self boot engine (SBE).  One role of
the SBE is to act as a proxy that provides access to the registers of
the POWER chip from other (embedded) systems.

The POWER9 chip contains a hardware frontend for communicating with
the SBE from remote systems called the SBEFIFO.  The SBEFIFO logic
is contained within an FSI CFAM  and as such the driver implements an
FSI bus device.

The SBE expects to communicate using a defined wire protocol; however,
the driver knows nothing of the protocol and only provides raw access
to the fifo device to userspace applications wishing to communicate with
the SBE using the wire protocol.

The SBEFIFO consists of two hardware fifos.  The upstream fifo is used
by the driver to transfer data to the SBE on the POWER chip, from the
system hosting the driver.  The downstream fifo is used by the driver to
transfer data from the SBE on the power chip to the system hosting the
driver.

The driver also provides an in-kernel API for SBEFIFO client device
drivers. In this way, applications and client drivers all have safe
access to the hardware.

Signed-off-by: Brad Bishop <bradleyb@fuzziesquirrel.com>
Signed-off-by: Edward A. James <eajames@us.ibm.com>
---
 drivers/fsi/Kconfig       |   7 +
 drivers/fsi/Makefile      |   1 +
 drivers/fsi/fsi-sbefifo.c | 832 ++++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 840 insertions(+)
 create mode 100644 drivers/fsi/fsi-sbefifo.c

diff --git a/drivers/fsi/Kconfig b/drivers/fsi/Kconfig
index 6821ed0..8c4d903 100644
--- a/drivers/fsi/Kconfig
+++ b/drivers/fsi/Kconfig
@@ -33,6 +33,13 @@ config FSI_SCOM
 	---help---
 	This option enables an FSI based SCOM device driver.
 
+config FSI_SBEFIFO
+	tristate "SBEFIFO FSI client device driver"
+	---help---
+	This option enables an FSI based SBEFIFO device driver. The SBEFIFO is
+	a pipe-like FSI device for communicating with the self boot engine
+	(SBE) on POWER processors.
+
 endif
 
 endmenu
diff --git a/drivers/fsi/Makefile b/drivers/fsi/Makefile
index 65eb99d..851182e 100644
--- a/drivers/fsi/Makefile
+++ b/drivers/fsi/Makefile
@@ -3,3 +3,4 @@ obj-$(CONFIG_FSI) += fsi-core.o
 obj-$(CONFIG_FSI_MASTER_HUB) += fsi-master-hub.o
 obj-$(CONFIG_FSI_MASTER_GPIO) += fsi-master-gpio.o
 obj-$(CONFIG_FSI_SCOM) += fsi-scom.o
+obj-$(CONFIG_FSI_SBEFIFO) += fsi-sbefifo.o
diff --git a/drivers/fsi/fsi-sbefifo.c b/drivers/fsi/fsi-sbefifo.c
new file mode 100644
index 0000000..6915f5e
--- /dev/null
+++ b/drivers/fsi/fsi-sbefifo.c
@@ -0,0 +1,832 @@
+/*
+ * Copyright (C) IBM Corporation 2017
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERGCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#include <linux/device.h>
+#include <linux/errno.h>
+#include <linux/fs.h>
+#include <linux/fsi.h>
+#include <linux/idr.h>
+#include <linux/kernel.h>
+#include <linux/kref.h>
+#include <linux/list.h>
+#include <linux/miscdevice.h>
+#include <linux/module.h>
+#include <linux/poll.h>
+#include <linux/sched.h>
+#include <linux/slab.h>
+#include <linux/spinlock.h>
+#include <linux/timer.h>
+#include <linux/uaccess.h>
+#include <linux/wait.h>
+
+/*
+ * The SBEFIFO is a pipe-like FSI device for communicating with
+ * the self boot engine on POWER processors.
+ */
+
+#define DEVICE_NAME		"sbefifo"
+#define FSI_ENGID_SBE		0x22
+#define SBEFIFO_BUF_CNT		32
+
+#define SBEFIFO_UP		0x00	/* Up register offset */
+#define SBEFIFO_DWN		0x40	/* Down register offset */
+
+#define SBEFIFO_STS		0x04
+#define   SBEFIFO_EMPTY			BIT(20)
+#define SBEFIFO_EOT_RAISE	0x08
+#define   SBEFIFO_EOT_MAGIC		0xffffffff
+#define SBEFIFO_EOT_ACK		0x14
+
+struct sbefifo {
+	struct timer_list poll_timer;
+	struct fsi_device *fsi_dev;	/* parent fsi device */
+	struct miscdevice mdev;		/* /dev/ entry */
+	wait_queue_head_t wait;
+	struct list_head link;
+	struct list_head xfrs;
+	struct kref kref;		/* reference counter */
+	spinlock_t lock;
+	char name[32];
+	int idx;
+	int rc;
+};
+
+struct sbefifo_buf {
+	u32 buf[SBEFIFO_BUF_CNT];
+	unsigned long flags;
+#define SBEFIFO_BUF_FULL		1
+	u32 *rpos;
+	u32 *wpos;
+};
+
+struct sbefifo_xfr {
+	struct sbefifo_buf *rbuf;
+	struct sbefifo_buf *wbuf;
+	struct list_head client;
+	struct list_head xfrs;
+	unsigned long flags;
+#define SBEFIFO_XFR_WRITE_DONE		1
+#define SBEFIFO_XFR_RESP_PENDING	2
+#define SBEFIFO_XFR_COMPLETE		3
+#define SBEFIFO_XFR_CANCEL		4
+};
+
+struct sbefifo_client {
+	struct sbefifo_buf rbuf;
+	struct sbefifo_buf wbuf;
+	struct list_head xfrs;
+	struct sbefifo *dev;
+	struct kref kref;
+};
+
+static struct list_head sbefifo_fifos;
+
+static DEFINE_IDA(sbefifo_ida);
+
+static int sbefifo_inw(struct sbefifo *sbefifo, int reg, u32 *word)
+{
+	int rc;
+	__be32 raw_word;
+
+	rc = fsi_device_read(sbefifo->fsi_dev, reg, &raw_word,
+			     sizeof(raw_word));
+	if (rc)
+		return rc;
+
+	*word = be32_to_cpu(raw_word);
+
+	return 0;
+}
+
+static int sbefifo_outw(struct sbefifo *sbefifo, int reg, u32 word)
+{
+	__be32 raw_word = cpu_to_be32(word);
+
+	return fsi_device_write(sbefifo->fsi_dev, reg, &raw_word,
+				sizeof(raw_word));
+}
+
+/* Don't flip endianness of data to/from FIFO, just pass through. */
+static int sbefifo_readw(struct sbefifo *sbefifo, u32 *word)
+{
+	return fsi_device_read(sbefifo->fsi_dev, SBEFIFO_DWN, word,
+			       sizeof(*word));
+}
+
+static int sbefifo_writew(struct sbefifo *sbefifo, u32 word)
+{
+	return fsi_device_write(sbefifo->fsi_dev, SBEFIFO_UP, &word,
+				sizeof(word));
+}
+
+static int sbefifo_ack_eot(struct sbefifo *sbefifo)
+{
+	u32 discard;
+	int ret;
+
+	 /* Discard the EOT word. */
+	ret = sbefifo_readw(sbefifo, &discard);
+	if (ret)
+		return ret;
+
+	return sbefifo_outw(sbefifo, SBEFIFO_DWN | SBEFIFO_EOT_ACK,
+			    SBEFIFO_EOT_MAGIC);
+}
+
+static size_t sbefifo_dev_nwreadable(u32 sts)
+{
+	static const u32 FIFO_NTRY_CNT_MSK = 0x000f0000;
+	static const unsigned int FIFO_NTRY_CNT_SHIFT = 16;
+
+	return (sts & FIFO_NTRY_CNT_MSK) >> FIFO_NTRY_CNT_SHIFT;
+}
+
+static size_t sbefifo_dev_nwwriteable(u32 sts)
+{
+	static const size_t FIFO_DEPTH = 8;
+
+	return FIFO_DEPTH - sbefifo_dev_nwreadable(sts);
+}
+
+static void sbefifo_buf_init(struct sbefifo_buf *buf)
+{
+	WRITE_ONCE(buf->flags, 0);
+	WRITE_ONCE(buf->rpos, buf->buf);
+	WRITE_ONCE(buf->wpos, buf->buf);
+}
+
+static size_t sbefifo_buf_nbreadable(struct sbefifo_buf *buf)
+{
+	size_t n;
+	u32 *rpos = READ_ONCE(buf->rpos);
+	u32 *wpos = READ_ONCE(buf->wpos);
+
+	if (test_bit(SBEFIFO_BUF_FULL, &buf->flags))
+		n = SBEFIFO_BUF_CNT;
+	else if (rpos <= wpos)
+		n = wpos - rpos;
+	else
+		n = (buf->buf + SBEFIFO_BUF_CNT) - rpos;
+
+	return n << 2;
+}
+
+static size_t sbefifo_buf_nbwriteable(struct sbefifo_buf *buf)
+{
+	size_t n;
+	u32 *rpos = READ_ONCE(buf->rpos);
+	u32 *wpos = READ_ONCE(buf->wpos);
+
+	if (test_bit(SBEFIFO_BUF_FULL, &buf->flags))
+		n = 0;
+	else if (wpos < rpos)
+		n = rpos - wpos;
+	else
+		n = (buf->buf + SBEFIFO_BUF_CNT) - wpos;
+
+	return n << 2;
+}
+
+/*
+ * Update pointers and flags after doing a buffer read.  Return true if the
+ * buffer is now empty;
+ */
+static bool sbefifo_buf_readnb(struct sbefifo_buf *buf, size_t n)
+{
+	u32 *rpos = READ_ONCE(buf->rpos);
+	u32 *wpos = READ_ONCE(buf->wpos);
+
+	if (n)
+		clear_bit(SBEFIFO_BUF_FULL, &buf->flags);
+
+	rpos += (n >> 2);
+	if (rpos == buf->buf + SBEFIFO_BUF_CNT)
+		rpos = buf->buf;
+
+	WRITE_ONCE(buf->rpos, rpos);
+
+	return rpos == wpos;
+}
+
+/*
+ * Update pointers and flags after doing a buffer write.  Return true if the
+ * buffer is now full.
+ */
+static bool sbefifo_buf_wrotenb(struct sbefifo_buf *buf, size_t n)
+{
+	u32 *rpos = READ_ONCE(buf->rpos);
+	u32 *wpos = READ_ONCE(buf->wpos);
+
+	wpos += (n >> 2);
+	if (wpos == buf->buf + SBEFIFO_BUF_CNT)
+		wpos = buf->buf;
+	if (wpos == rpos)
+		set_bit(SBEFIFO_BUF_FULL, &buf->flags);
+
+	WRITE_ONCE(buf->wpos, wpos);
+
+	return rpos == wpos;
+}
+
+static void sbefifo_free(struct kref *kref)
+{
+	struct sbefifo *sbefifo = container_of(kref, struct sbefifo, kref);
+
+	kfree(sbefifo);
+}
+
+static void sbefifo_get(struct sbefifo *sbefifo)
+{
+	kref_get(&sbefifo->kref);
+}
+
+static void sbefifo_put(struct sbefifo *sbefifo)
+{
+	kref_put(&sbefifo->kref, sbefifo_free);
+}
+
+static struct sbefifo_xfr *sbefifo_enq_xfr(struct sbefifo_client *client)
+{
+	struct sbefifo *sbefifo = client->dev;
+	struct sbefifo_xfr *xfr;
+
+	xfr = kzalloc(sizeof(*xfr), GFP_KERNEL);
+	if (!xfr)
+		return NULL;
+
+	xfr->rbuf = &client->rbuf;
+	xfr->wbuf = &client->wbuf;
+	list_add_tail(&xfr->xfrs, &sbefifo->xfrs);
+	list_add_tail(&xfr->client, &client->xfrs);
+
+	return xfr;
+}
+
+static struct sbefifo_xfr *sbefifo_client_next_xfr(
+		struct sbefifo_client *client)
+{
+	if (list_empty(&client->xfrs))
+		return NULL;
+
+	return container_of(client->xfrs.next, struct sbefifo_xfr, client);
+}
+
+static bool sbefifo_xfr_rsp_pending(struct sbefifo_client *client)
+{
+	struct sbefifo_xfr *xfr;
+
+	xfr = sbefifo_client_next_xfr(client);
+	if (xfr && test_bit(SBEFIFO_XFR_RESP_PENDING, &xfr->flags))
+		return true;
+
+	return false;
+}
+
+static struct sbefifo_client *sbefifo_new_client(struct sbefifo *sbefifo)
+{
+	struct sbefifo_client *client;
+
+	client = kzalloc(sizeof(*client), GFP_KERNEL);
+	if (!client)
+		return NULL;
+
+	kref_init(&client->kref);
+	client->dev = sbefifo;
+	sbefifo_buf_init(&client->rbuf);
+	sbefifo_buf_init(&client->wbuf);
+	INIT_LIST_HEAD(&client->xfrs);
+
+	sbefifo_get(sbefifo);
+
+	return client;
+}
+
+static void sbefifo_client_release(struct kref *kref)
+{
+	struct sbefifo_xfr *xfr;
+	struct sbefifo_client *client = container_of(kref,
+						     struct sbefifo_client,
+						     kref);
+
+	list_for_each_entry(xfr, &client->xfrs, client) {
+		/*
+		 * The client left with pending or running xfrs.
+		 * Cancel them.
+		 */
+		set_bit(SBEFIFO_XFR_CANCEL, &xfr->flags);
+		sbefifo_get(client->dev);
+		if (mod_timer(&client->dev->poll_timer, jiffies))
+			sbefifo_put(client->dev);
+	}
+
+	sbefifo_put(client->dev);
+	kfree(client);
+}
+
+static void sbefifo_get_client(struct sbefifo_client *client)
+{
+	kref_get(&client->kref);
+}
+
+static void sbefifo_put_client(struct sbefifo_client *client)
+{
+	kref_put(&client->kref, sbefifo_client_release);
+}
+
+static struct sbefifo_xfr *sbefifo_next_xfr(struct sbefifo *sbefifo)
+{
+	struct sbefifo_xfr *xfr, *tmp;
+
+	list_for_each_entry_safe(xfr, tmp, &sbefifo->xfrs, xfrs) {
+		if (unlikely(test_bit(SBEFIFO_XFR_CANCEL, &xfr->flags))) {
+			/* Discard cancelled transfers. */
+			list_del(&xfr->xfrs);
+			kfree(xfr);
+			continue;
+		}
+
+		return xfr;
+	}
+
+	return NULL;
+}
+
+static void sbefifo_poll_timer(unsigned long data)
+{
+	static const unsigned long EOT_MASK = 0x000000ff;
+	struct sbefifo *sbefifo = (void *)data;
+	struct sbefifo_buf *rbuf, *wbuf;
+	struct sbefifo_xfr *xfr = NULL;
+	struct sbefifo_buf drain;
+	size_t devn, bufn;
+	int eot = 0;
+	int ret = 0;
+	u32 sts;
+	int i;
+
+	spin_lock(&sbefifo->lock);
+	xfr = list_first_entry_or_null(&sbefifo->xfrs, struct sbefifo_xfr,
+				       xfrs);
+	if (!xfr)
+		goto out_unlock;
+
+	rbuf = xfr->rbuf;
+	wbuf = xfr->wbuf;
+
+	if (unlikely(test_bit(SBEFIFO_XFR_CANCEL, &xfr->flags))) {
+		/* The client left. */
+		rbuf = &drain;
+		wbuf = &drain;
+		sbefifo_buf_init(&drain);
+		if (!test_bit(SBEFIFO_XFR_RESP_PENDING, &xfr->flags))
+			set_bit(SBEFIFO_XFR_WRITE_DONE, &xfr->flags);
+	}
+
+	 /* Drain the write buffer. */
+	while ((bufn = sbefifo_buf_nbreadable(wbuf))) {
+		ret = sbefifo_inw(sbefifo, SBEFIFO_UP | SBEFIFO_STS, &sts);
+		if (ret)
+			goto out;
+
+		devn = sbefifo_dev_nwwriteable(sts);
+		if (devn == 0) {
+			/* No open slot for write.  Reschedule. */
+			sbefifo->poll_timer.expires = jiffies +
+				msecs_to_jiffies(500);
+			add_timer(&sbefifo->poll_timer);
+			goto out_unlock;
+		}
+
+		devn = min_t(size_t, devn, bufn >> 2);
+		for (i = 0; i < devn; i++) {
+			ret = sbefifo_writew(sbefifo, *wbuf->rpos);
+			if (ret)
+				goto out;
+
+			sbefifo_buf_readnb(wbuf, 1 << 2);
+		}
+	}
+
+	 /* Send EOT if the writer is finished. */
+	if (test_and_clear_bit(SBEFIFO_XFR_WRITE_DONE, &xfr->flags)) {
+		ret = sbefifo_outw(sbefifo, SBEFIFO_UP | SBEFIFO_EOT_RAISE,
+				   SBEFIFO_EOT_MAGIC);
+		if (ret)
+			goto out;
+
+		/* Inform reschedules that the writer is finished. */
+		set_bit(SBEFIFO_XFR_RESP_PENDING, &xfr->flags);
+	}
+
+	/* Nothing left to do if the writer is not finished. */
+	if (!test_bit(SBEFIFO_XFR_RESP_PENDING, &xfr->flags))
+		goto out;
+
+	 /* Fill the read buffer. */
+	while ((bufn = sbefifo_buf_nbwriteable(rbuf))) {
+		ret = sbefifo_inw(sbefifo, SBEFIFO_DWN | SBEFIFO_STS, &sts);
+		if (ret)
+			goto out;
+
+		devn = sbefifo_dev_nwreadable(sts);
+		if (devn == 0) {
+			/* No data yet.  Reschedule. */
+			sbefifo->poll_timer.expires = jiffies +
+				msecs_to_jiffies(500);
+			add_timer(&sbefifo->poll_timer);
+			goto out_unlock;
+		}
+
+		/* Fill.  The EOT word is discarded.  */
+		devn = min_t(size_t, devn, bufn >> 2);
+		eot = (sts & EOT_MASK) != 0;
+		if (eot)
+			devn--;
+
+		for (i = 0; i < devn; i++) {
+			ret = sbefifo_readw(sbefifo, rbuf->wpos);
+			if (ret)
+				goto out;
+
+			if (likely(!test_bit(SBEFIFO_XFR_CANCEL, &xfr->flags)))
+				sbefifo_buf_wrotenb(rbuf, 1 << 2);
+		}
+
+		if (eot) {
+			ret = sbefifo_ack_eot(sbefifo);
+			if (ret)
+				goto out;
+
+			set_bit(SBEFIFO_XFR_COMPLETE, &xfr->flags);
+			list_del(&xfr->xfrs);
+			if (unlikely(test_bit(SBEFIFO_XFR_CANCEL,
+					      &xfr->flags)))
+				kfree(xfr);
+			break;
+		}
+	}
+
+out:
+	if (unlikely(ret)) {
+		sbefifo->rc = ret;
+		dev_err(&sbefifo->fsi_dev->dev,
+			"Fatal bus access failure: %d\n", ret);
+		list_for_each_entry(xfr, &sbefifo->xfrs, xfrs)
+			kfree(xfr);
+		INIT_LIST_HEAD(&sbefifo->xfrs);
+
+	} else if (eot && sbefifo_next_xfr(sbefifo)) {
+		sbefifo_get(sbefifo);
+		sbefifo->poll_timer.expires = jiffies;
+		add_timer(&sbefifo->poll_timer);
+	}
+
+	sbefifo_put(sbefifo);
+	wake_up(&sbefifo->wait);
+
+out_unlock:
+	spin_unlock(&sbefifo->lock);
+}
+
+static int sbefifo_open(struct inode *inode, struct file *file)
+{
+	struct sbefifo *sbefifo = container_of(file->private_data,
+					       struct sbefifo, mdev);
+	struct sbefifo_client *client;
+	int ret;
+
+	ret = READ_ONCE(sbefifo->rc);
+	if (ret)
+		return ret;
+
+	client = sbefifo_new_client(sbefifo);
+	if (!client)
+		return -ENOMEM;
+
+	file->private_data = client;
+
+	return 0;
+}
+
+static unsigned int sbefifo_poll(struct file *file, poll_table *wait)
+{
+	struct sbefifo_client *client = file->private_data;
+	struct sbefifo *sbefifo = client->dev;
+	unsigned int mask = 0;
+
+	poll_wait(file, &sbefifo->wait, wait);
+
+	if (READ_ONCE(sbefifo->rc))
+		mask |= POLLERR;
+
+	if (sbefifo_buf_nbreadable(&client->rbuf))
+		mask |= POLLIN;
+
+	if (sbefifo_buf_nbwriteable(&client->wbuf))
+		mask |= POLLOUT;
+
+	return mask;
+}
+
+static ssize_t sbefifo_read(struct file *file, char __user *buf, size_t len,
+			    loff_t *offset)
+{
+	struct sbefifo_client *client = file->private_data;
+	struct sbefifo *sbefifo = client->dev;
+	struct sbefifo_xfr *xfr;
+	ssize_t ret = 0;
+	size_t n;
+
+	if ((len >> 2) << 2 != len)
+		return -EINVAL;
+
+	if ((file->f_flags & O_NONBLOCK) && !sbefifo_xfr_rsp_pending(client))
+		return -EAGAIN;
+
+	sbefifo_get_client(client);
+	if (wait_event_interruptible(sbefifo->wait,
+				(ret = READ_ONCE(sbefifo->rc)) ||
+				(n = sbefifo_buf_nbreadable(&client->rbuf)))) {
+		sbefifo_put_client(client);
+		return -ERESTARTSYS;
+	}
+
+	if (ret) {
+		INIT_LIST_HEAD(&client->xfrs);
+		sbefifo_put_client(client);
+		return ret;
+	}
+
+	n = min_t(size_t, n, len);
+
+	if (copy_to_user(buf, READ_ONCE(client->rbuf.rpos), n)) {
+		sbefifo_put_client(client);
+		return -EFAULT;
+	}
+
+	if (sbefifo_buf_readnb(&client->rbuf, n)) {
+		xfr = sbefifo_client_next_xfr(client);
+		if (!test_bit(SBEFIFO_XFR_COMPLETE, &xfr->flags)) {
+			/*
+			 * Fill the read buffer back up.
+			 */
+			sbefifo_get(sbefifo);
+			if (mod_timer(&client->dev->poll_timer, jiffies))
+				sbefifo_put(sbefifo);
+		} else {
+			kfree(xfr);
+			list_del(&xfr->client);
+			wake_up(&sbefifo->wait);
+		}
+	}
+
+	sbefifo_put_client(client);
+
+	return n;
+}
+
+static ssize_t sbefifo_write(struct file *file, const char __user *buf,
+			     size_t len, loff_t *offset)
+{
+	struct sbefifo_client *client = file->private_data;
+	struct sbefifo *sbefifo = client->dev;
+	struct sbefifo_buf *wbuf = &client->wbuf;
+	struct sbefifo_xfr *xfr;
+	ssize_t ret = 0;
+	size_t n;
+
+	if ((len >> 2) << 2 != len)
+		return -EINVAL;
+
+	if (!len)
+		return 0;
+
+	n = sbefifo_buf_nbwriteable(wbuf);
+
+	spin_lock_irq(&sbefifo->lock);
+	xfr = sbefifo_next_xfr(sbefifo);
+
+	if ((file->f_flags & O_NONBLOCK) && xfr && n < len) {
+		spin_unlock_irq(&sbefifo->lock);
+		return -EAGAIN;
+	}
+
+	xfr = sbefifo_enq_xfr(client);
+	if (!xfr) {
+		spin_unlock_irq(&sbefifo->lock);
+		return -ENOMEM;
+	}
+	spin_unlock_irq(&sbefifo->lock);
+
+	sbefifo_get_client(client);
+
+	/*
+	 * Partial writes are not really allowed in that EOT is sent exactly
+	 * once per write.
+	 */
+	while (len) {
+		if (wait_event_interruptible(sbefifo->wait,
+				READ_ONCE(sbefifo->rc) ||
+				(sbefifo_client_next_xfr(client) == xfr &&
+				(n = sbefifo_buf_nbwriteable(wbuf))))) {
+			set_bit(SBEFIFO_XFR_CANCEL, &xfr->flags);
+			sbefifo_get(sbefifo);
+			if (mod_timer(&sbefifo->poll_timer, jiffies))
+				sbefifo_put(sbefifo);
+			sbefifo_put_client(client);
+			return -ERESTARTSYS;
+		}
+
+		if (sbefifo->rc) {
+			INIT_LIST_HEAD(&client->xfrs);
+			sbefifo_put_client(client);
+			return sbefifo->rc;
+		}
+
+		n = min_t(size_t, n, len);
+
+		if (copy_from_user(READ_ONCE(wbuf->wpos), buf, n)) {
+			set_bit(SBEFIFO_XFR_CANCEL, &xfr->flags);
+			sbefifo_get(sbefifo);
+			if (mod_timer(&sbefifo->poll_timer, jiffies))
+				sbefifo_put(sbefifo);
+			sbefifo_put_client(client);
+			return -EFAULT;
+		}
+
+		sbefifo_buf_wrotenb(wbuf, n);
+		len -= n;
+		buf += n;
+		ret += n;
+
+		/* Set this before starting timer to avoid race condition on
+		 * this flag with the timer function writer.
+		 */
+		if (!len)
+			set_bit(SBEFIFO_XFR_WRITE_DONE, &xfr->flags);
+
+		/*
+		 * Drain the write buffer.
+		 */
+		sbefifo_get(sbefifo);
+		if (mod_timer(&client->dev->poll_timer, jiffies))
+			sbefifo_put(sbefifo);
+	}
+
+	sbefifo_put_client(client);
+
+	return ret;
+}
+
+static int sbefifo_release(struct inode *inode, struct file *file)
+{
+	struct sbefifo_client *client = file->private_data;
+	struct sbefifo *sbefifo = client->dev;
+
+	sbefifo_put_client(client);
+
+	return READ_ONCE(sbefifo->rc);
+}
+
+static const struct file_operations sbefifo_fops = {
+	.owner		= THIS_MODULE,
+	.open		= sbefifo_open,
+	.read		= sbefifo_read,
+	.write		= sbefifo_write,
+	.poll		= sbefifo_poll,
+	.release	= sbefifo_release,
+};
+
+static int sbefifo_probe(struct device *dev)
+{
+	struct fsi_device *fsi_dev = to_fsi_dev(dev);
+	struct sbefifo *sbefifo;
+	u32 sts;
+	int ret;
+
+	sbefifo = kzalloc(sizeof(*sbefifo), GFP_KERNEL);
+	if (!sbefifo)
+		return -ENOMEM;
+
+	sbefifo->fsi_dev = fsi_dev;
+
+	ret = sbefifo_inw(sbefifo, SBEFIFO_UP | SBEFIFO_STS, &sts);
+	if (ret)
+		return ret;
+
+	if (!(sts & SBEFIFO_EMPTY)) {
+		dev_err(&sbefifo->fsi_dev->dev,
+			"Found data in upstream fifo\n");
+		return -EIO;
+	}
+
+	ret = sbefifo_inw(sbefifo, SBEFIFO_DWN | SBEFIFO_STS, &sts);
+	if (ret)
+		return ret;
+
+	if (!(sts & SBEFIFO_EMPTY)) {
+		dev_err(&sbefifo->fsi_dev->dev,
+			"Found data in downstream fifo\n");
+		return -EIO;
+	}
+
+	sbefifo->mdev.minor = MISC_DYNAMIC_MINOR;
+	sbefifo->mdev.fops = &sbefifo_fops;
+	sbefifo->mdev.name = sbefifo->name;
+	sbefifo->mdev.parent = dev;
+	spin_lock_init(&sbefifo->lock);
+	kref_init(&sbefifo->kref);
+
+	sbefifo->idx = ida_simple_get(&sbefifo_ida, 1, INT_MAX, GFP_KERNEL);
+	snprintf(sbefifo->name, sizeof(sbefifo->name), "sbefifo%d",
+		 sbefifo->idx);
+	init_waitqueue_head(&sbefifo->wait);
+	INIT_LIST_HEAD(&sbefifo->xfrs);
+
+	/* This bit of silicon doesn't offer any interrupts... */
+	setup_timer(&sbefifo->poll_timer, sbefifo_poll_timer,
+		    (unsigned long)sbefifo);
+
+	list_add(&sbefifo->link, &sbefifo_fifos);
+
+	return misc_register(&sbefifo->mdev);
+}
+
+static int sbefifo_remove(struct device *dev)
+{
+	struct fsi_device *fsi_dev = to_fsi_dev(dev);
+	struct sbefifo *sbefifo, *sbefifo_tmp;
+	struct sbefifo_xfr *xfr;
+
+	list_for_each_entry_safe(sbefifo, sbefifo_tmp, &sbefifo_fifos, link) {
+		if (sbefifo->fsi_dev != fsi_dev)
+			continue;
+
+		misc_deregister(&sbefifo->mdev);
+		list_del(&sbefifo->link);
+		ida_simple_remove(&sbefifo_ida, sbefifo->idx);
+
+		if (del_timer_sync(&sbefifo->poll_timer))
+			sbefifo_put(sbefifo);
+
+		spin_lock(&sbefifo->lock);
+		list_for_each_entry(xfr, &sbefifo->xfrs, xfrs)
+			kfree(xfr);
+		spin_unlock(&sbefifo->lock);
+
+		WRITE_ONCE(sbefifo->rc, -ENODEV);
+
+		wake_up(&sbefifo->wait);
+		sbefifo_put(sbefifo);
+	}
+
+	return 0;
+}
+
+static struct fsi_device_id sbefifo_ids[] = {
+	{
+		.engine_type = FSI_ENGID_SBE,
+		.version = FSI_VERSION_ANY,
+	},
+	{ 0 }
+};
+
+static struct fsi_driver sbefifo_drv = {
+	.id_table = sbefifo_ids,
+	.drv = {
+		.name = DEVICE_NAME,
+		.bus = &fsi_bus_type,
+		.probe = sbefifo_probe,
+		.remove = sbefifo_remove,
+	}
+};
+
+static int sbefifo_init(void)
+{
+	INIT_LIST_HEAD(&sbefifo_fifos);
+	return fsi_driver_register(&sbefifo_drv);
+}
+
+static void sbefifo_exit(void)
+{
+	fsi_driver_unregister(&sbefifo_drv);
+
+	ida_destroy(&sbefifo_ida);
+}
+
+module_init(sbefifo_init);
+module_exit(sbefifo_exit);
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Brad Bishop <bradleyb@fuzziesquirrel.com>");
+MODULE_AUTHOR("Eddie James <eajames@linux.vnet.ibm.com>");
+MODULE_DESCRIPTION("Linux device interface to the POWER self boot engine");
-- 
1.8.3.1

[PATCH v2 2/5] drivers/fsi/sbefifo: Add in-kernel API

[Eddie James]

From: "Edward A. James" <eajames@us.ibm.com>

Refactor the user interface of the SBEFIFO driver to allow for an
in-kernel read/write API. Add exported functions for other drivers to
call, and add an include file with those functions. Also parse the
device tree for child nodes and create child platform devices
accordingly.

Signed-off-by: Edward A. James <eajames@us.ibm.com>
---
 drivers/fsi/fsi-sbefifo.c   | 142 +++++++++++++++++++++++++++++++++++++-------
 include/linux/fsi-sbefifo.h |  30 ++++++++++
 2 files changed, 151 insertions(+), 21 deletions(-)
 create mode 100644 include/linux/fsi-sbefifo.h

diff --git a/drivers/fsi/fsi-sbefifo.c b/drivers/fsi/fsi-sbefifo.c
index 6915f5e..e0ddc5d 100644
--- a/drivers/fsi/fsi-sbefifo.c
+++ b/drivers/fsi/fsi-sbefifo.c
@@ -15,12 +15,16 @@
 #include <linux/errno.h>
 #include <linux/fs.h>
 #include <linux/fsi.h>
+#include <linux/fsi-sbefifo.h>
 #include <linux/idr.h>
 #include <linux/kernel.h>
 #include <linux/kref.h>
 #include <linux/list.h>
 #include <linux/miscdevice.h>
 #include <linux/module.h>
+#include <linux/of.h>
+#include <linux/of_device.h>
+#include <linux/of_platform.h>
 #include <linux/poll.h>
 #include <linux/sched.h>
 #include <linux/slab.h>
@@ -87,6 +91,7 @@ struct sbefifo_client {
 	struct list_head xfrs;
 	struct sbefifo *dev;
 	struct kref kref;
+	unsigned long f_flags;
 };
 
 static struct list_head sbefifo_fifos;
@@ -514,6 +519,7 @@ static int sbefifo_open(struct inode *inode, struct file *file)
 		return -ENOMEM;
 
 	file->private_data = client;
+	client->f_flags = file->f_flags;
 
 	return 0;
 }
@@ -538,19 +544,18 @@ static unsigned int sbefifo_poll(struct file *file, poll_table *wait)
 	return mask;
 }
 
-static ssize_t sbefifo_read(struct file *file, char __user *buf, size_t len,
-			    loff_t *offset)
+static ssize_t sbefifo_read_common(struct sbefifo_client *client,
+				   char __user *ubuf, char *kbuf, size_t len)
 {
-	struct sbefifo_client *client = file->private_data;
 	struct sbefifo *sbefifo = client->dev;
 	struct sbefifo_xfr *xfr;
-	ssize_t ret = 0;
 	size_t n;
+	ssize_t ret = 0;
 
 	if ((len >> 2) << 2 != len)
 		return -EINVAL;
 
-	if ((file->f_flags & O_NONBLOCK) && !sbefifo_xfr_rsp_pending(client))
+	if ((client->f_flags & O_NONBLOCK) && !sbefifo_xfr_rsp_pending(client))
 		return -EAGAIN;
 
 	sbefifo_get_client(client);
@@ -569,10 +574,13 @@ static ssize_t sbefifo_read(struct file *file, char __user *buf, size_t len,
 
 	n = min_t(size_t, n, len);
 
-	if (copy_to_user(buf, READ_ONCE(client->rbuf.rpos), n)) {
-		sbefifo_put_client(client);
-		return -EFAULT;
-	}
+	if (ubuf) {
+		if (copy_to_user(ubuf, READ_ONCE(client->rbuf.rpos), n)) {
+			sbefifo_put_client(client);
+			return -EFAULT;
+		}
+	} else
+		memcpy(kbuf, READ_ONCE(client->rbuf.rpos), n);
 
 	if (sbefifo_buf_readnb(&client->rbuf, n)) {
 		xfr = sbefifo_client_next_xfr(client);
@@ -595,10 +603,18 @@ static ssize_t sbefifo_read(struct file *file, char __user *buf, size_t len,
 	return n;
 }
 
-static ssize_t sbefifo_write(struct file *file, const char __user *buf,
-			     size_t len, loff_t *offset)
+static ssize_t sbefifo_read(struct file *file, char __user *buf, size_t len,
+			    loff_t *offset)
 {
 	struct sbefifo_client *client = file->private_data;
+
+	return sbefifo_read_common(client, buf, NULL, len);
+}
+
+static ssize_t sbefifo_write_common(struct sbefifo_client *client,
+				    const char __user *ubuf, const char *kbuf,
+				    size_t len)
+{
 	struct sbefifo *sbefifo = client->dev;
 	struct sbefifo_buf *wbuf = &client->wbuf;
 	struct sbefifo_xfr *xfr;
@@ -616,7 +632,7 @@ static ssize_t sbefifo_write(struct file *file, const char __user *buf,
 	spin_lock_irq(&sbefifo->lock);
 	xfr = sbefifo_next_xfr(sbefifo);
 
-	if ((file->f_flags & O_NONBLOCK) && xfr && n < len) {
+	if ((client->f_flags & O_NONBLOCK) && xfr && n < len) {
 		spin_unlock_irq(&sbefifo->lock);
 		return -EAGAIN;
 	}
@@ -655,18 +671,24 @@ static ssize_t sbefifo_write(struct file *file, const char __user *buf,
 
 		n = min_t(size_t, n, len);
 
-		if (copy_from_user(READ_ONCE(wbuf->wpos), buf, n)) {
-			set_bit(SBEFIFO_XFR_CANCEL, &xfr->flags);
-			sbefifo_get(sbefifo);
-			if (mod_timer(&sbefifo->poll_timer, jiffies))
-				sbefifo_put(sbefifo);
-			sbefifo_put_client(client);
-			return -EFAULT;
+		if (ubuf) {
+			if (copy_from_user(READ_ONCE(wbuf->wpos), ubuf, n)) {
+				set_bit(SBEFIFO_XFR_CANCEL, &xfr->flags);
+				sbefifo_get(sbefifo);
+				if (mod_timer(&sbefifo->poll_timer, jiffies))
+					sbefifo_put(sbefifo);
+				sbefifo_put_client(client);
+				return -EFAULT;
+			}
+
+			ubuf += n;
+		} else {
+			memcpy(READ_ONCE(wbuf->wpos), kbuf, n);
+			kbuf += n;
 		}
 
 		sbefifo_buf_wrotenb(wbuf, n);
 		len -= n;
-		buf += n;
 		ret += n;
 
 		/* Set this before starting timer to avoid race condition on
@@ -688,6 +710,14 @@ static ssize_t sbefifo_write(struct file *file, const char __user *buf,
 	return ret;
 }
 
+static ssize_t sbefifo_write(struct file *file, const char __user *buf,
+			     size_t len, loff_t *offset)
+{
+	struct sbefifo_client *client = file->private_data;
+
+	return sbefifo_write_common(client, buf, NULL, len);
+}
+
 static int sbefifo_release(struct inode *inode, struct file *file)
 {
 	struct sbefifo_client *client = file->private_data;
@@ -707,12 +737,68 @@ static int sbefifo_release(struct inode *inode, struct file *file)
 	.release	= sbefifo_release,
 };
 
+struct sbefifo_client *sbefifo_drv_open(struct device *dev,
+					unsigned long flags)
+{
+	struct sbefifo_client *client = NULL;
+	struct sbefifo *sbefifo;
+	struct fsi_device *fsi_dev = to_fsi_dev(dev);
+
+	list_for_each_entry(sbefifo, &sbefifo_fifos, link) {
+		if (sbefifo->fsi_dev != fsi_dev)
+			continue;
+
+		client = sbefifo_new_client(sbefifo);
+		if (client)
+			client->f_flags = flags;
+	}
+
+	return client;
+}
+EXPORT_SYMBOL_GPL(sbefifo_drv_open);
+
+int sbefifo_drv_read(struct sbefifo_client *client, char *buf, size_t len)
+{
+	return sbefifo_read_common(client, NULL, buf, len);
+}
+EXPORT_SYMBOL_GPL(sbefifo_drv_read);
+
+int sbefifo_drv_write(struct sbefifo_client *client, const char *buf,
+		      size_t len)
+{
+	return sbefifo_write_common(client, NULL, buf, len);
+}
+EXPORT_SYMBOL_GPL(sbefifo_drv_write);
+
+void sbefifo_drv_release(struct sbefifo_client *client)
+{
+	if (!client)
+		return;
+
+	sbefifo_put_client(client);
+}
+EXPORT_SYMBOL_GPL(sbefifo_drv_release);
+
+static int sbefifo_unregister_child(struct device *dev, void *data)
+{
+	struct platform_device *child = to_platform_device(dev);
+
+	of_device_unregister(child);
+	if (dev->of_node)
+		of_node_clear_flag(dev->of_node, OF_POPULATED);
+
+	return 0;
+}
+
 static int sbefifo_probe(struct device *dev)
 {
 	struct fsi_device *fsi_dev = to_fsi_dev(dev);
 	struct sbefifo *sbefifo;
+	struct device_node *np;
+	struct platform_device *child;
+	char child_name[32];
 	u32 sts;
-	int ret;
+	int ret, child_idx = 0;
 
 	sbefifo = kzalloc(sizeof(*sbefifo), GFP_KERNEL);
 	if (!sbefifo)
@@ -759,6 +845,18 @@ static int sbefifo_probe(struct device *dev)
 
 	list_add(&sbefifo->link, &sbefifo_fifos);
 
+	if (dev->of_node) {
+		/* create platform devs for dts child nodes (occ, etc) */
+		for_each_child_of_node(dev->of_node, np) {
+			snprintf(child_name, sizeof(child_name), "%s-dev%d",
+				 sbefifo->name, child_idx++);
+			child = of_platform_device_create(np, child_name, dev);
+			if (!child)
+				dev_warn(&sbefifo->fsi_dev->dev,
+					 "failed to create child node dev\n");
+		}
+	}
+
 	return misc_register(&sbefifo->mdev);
 }
 
@@ -772,6 +870,8 @@ static int sbefifo_remove(struct device *dev)
 		if (sbefifo->fsi_dev != fsi_dev)
 			continue;
 
+		device_for_each_child(dev, NULL, sbefifo_unregister_child);
+
 		misc_deregister(&sbefifo->mdev);
 		list_del(&sbefifo->link);
 		ida_simple_remove(&sbefifo_ida, sbefifo->idx);
diff --git a/include/linux/fsi-sbefifo.h b/include/linux/fsi-sbefifo.h
new file mode 100644
index 0000000..8e55891
--- /dev/null
+++ b/include/linux/fsi-sbefifo.h
@@ -0,0 +1,30 @@
+/*
+ * SBEFIFO FSI Client device driver
+ *
+ * Copyright (C) IBM Corporation 2017
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERGCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#ifndef LINUX_FSI_SBEFIFO_H
+#define LINUX_FSI_SBEFIFO_H
+
+struct device;
+struct sbefifo_client;
+
+extern struct sbefifo_client *sbefifo_drv_open(struct device *dev,
+					       unsigned long flags);
+extern int sbefifo_drv_read(struct sbefifo_client *client, char *buf,
+			    size_t len);
+extern int sbefifo_drv_write(struct sbefifo_client *client, const char *buf,
+			     size_t len);
+extern void sbefifo_drv_release(struct sbefifo_client *client);
+
+#endif /* LINUX_FSI_SBEFIFO_H */
-- 
1.8.3.1

[PATCH v2 3/5] drivers/fsi: Add On-Chip Controller (OCC) driver

[Eddie James]

From: "Edward A. James" <eajames@us.ibm.com>

The OCC is a device embedded on a POWER processor that collects and
aggregates sensor data from the processor and system. The OCC can
provide the raw sensor data as well as perform thermal and power
management on the system.

This driver provides an atomic communications channel between a service
processor (e.g. a BMC) and the OCC. The driver is dependent on the FSI
SBEFIFO driver to get hardware access through the SBE to the OCC SRAM.
Commands are issued to the SBE to send or fetch data to the SRAM.

The format of communications to the OCC is writing a command to SRAM,
followed by a sending a "doorbell" or attention to the OCC, followed by
reading the response from OCC. All of this takes place atomically so
that multiple users don't collide in the SRAM.

Signed-off-by: Edward A. James <eajames@us.ibm.com>
---
 drivers/fsi/Kconfig  |  10 +
 drivers/fsi/Makefile |   1 +
 drivers/fsi/occ.c    | 712 +++++++++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 723 insertions(+)
 create mode 100644 drivers/fsi/occ.c

diff --git a/drivers/fsi/Kconfig b/drivers/fsi/Kconfig
index 8c4d903..9d59276 100644
--- a/drivers/fsi/Kconfig
+++ b/drivers/fsi/Kconfig
@@ -40,6 +40,16 @@ config FSI_SBEFIFO
 	a pipe-like FSI device for communicating with the self boot engine
 	(SBE) on POWER processors.
 
+config OCCFIFO
+	tristate "OCC SBEFIFO client device driver"
+	depends on FSI_SBEFIFO
+	---help---
+	This option enables an SBEFIFO based On-Chip Controller (OCC) device
+	driver. The OCC is a device embedded on a POWER processor that collects
+	and aggregates sensor data from the processor and system. The OCC can
+	provide the raw sensor data as well as perform thermal and power
+	management on the system.
+
 endif
 
 endmenu
diff --git a/drivers/fsi/Makefile b/drivers/fsi/Makefile
index 851182e..7f5ca61 100644
--- a/drivers/fsi/Makefile
+++ b/drivers/fsi/Makefile
@@ -4,3 +4,4 @@ obj-$(CONFIG_FSI_MASTER_HUB) += fsi-master-hub.o
 obj-$(CONFIG_FSI_MASTER_GPIO) += fsi-master-gpio.o
 obj-$(CONFIG_FSI_SCOM) += fsi-scom.o
 obj-$(CONFIG_FSI_SBEFIFO) += fsi-sbefifo.o
+obj-$(CONFIG_OCCFIFO) += occ.o
diff --git a/drivers/fsi/occ.c b/drivers/fsi/occ.c
new file mode 100644
index 0000000..df13dcb
--- /dev/null
+++ b/drivers/fsi/occ.c
@@ -0,0 +1,712 @@
+/*
+ * Copyright 2017 IBM Corp.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#include <asm/unaligned.h>
+#include <linux/device.h>
+#include <linux/err.h>
+#include <linux/errno.h>
+#include <linux/fs.h>
+#include <linux/fsi-sbefifo.h>
+#include <linux/idr.h>
+#include <linux/kernel.h>
+#include <linux/list.h>
+#include <linux/miscdevice.h>
+#include <linux/module.h>
+#include <linux/mutex.h>
+#include <linux/of.h>
+#include <linux/platform_device.h>
+#include <linux/sched.h>
+#include <linux/slab.h>
+#include <linux/spinlock.h>
+#include <linux/uaccess.h>
+#include <linux/wait.h>
+#include <linux/workqueue.h>
+
+#define OCC_SRAM_BYTES		4096
+#define OCC_CMD_DATA_BYTES	4090
+#define OCC_RESP_DATA_BYTES	4089
+
+#define OCC_RESP_CMD_IN_PRG	0xFF
+
+#define OCC_TIMEOUT_MS		1000
+#define OCC_CMD_IN_PRG_WAIT_MS	50
+
+struct occ {
+	struct device *sbefifo;
+	char name[32];
+	int idx;
+	struct miscdevice mdev;
+	struct list_head xfrs;
+	spinlock_t list_lock;
+	struct mutex occ_lock;
+	struct work_struct work;
+};
+
+#define to_occ(x)	container_of((x), struct occ, mdev)
+
+struct occ_response {
+	u8 seq_no;
+	u8 cmd_type;
+	u8 return_status;
+	__be16 data_length;
+	u8 data[OCC_RESP_DATA_BYTES];
+	__be16 checksum;
+} __packed;
+
+/*
+ * transfer flags are NOT mutually exclusive
+ *
+ * Initial flags are none; transfer is created and queued from write(). All
+ *  flags are cleared when the transfer is completed by closing the file or
+ *  reading all of the available response data.
+ * XFR_IN_PROGRESS is set when a transfer is started from occ_worker_putsram,
+ *  and cleared if the transfer fails or occ_worker_getsram completes.
+ * XFR_COMPLETE is set when a transfer fails or finishes occ_worker_getsram.
+ * XFR_CANCELED is set when the transfer's client is released.
+ * XFR_WAITING is set from read() if the transfer isn't complete and
+ *  O_NONBLOCK wasn't specified. Cleared in read() when transfer completes or
+ *  fails.
+ */
+enum {
+	XFR_IN_PROGRESS,
+	XFR_COMPLETE,
+	XFR_CANCELED,
+	XFR_WAITING,
+};
+
+struct occ_xfr {
+	struct list_head link;
+	int rc;
+	u8 buf[OCC_SRAM_BYTES];
+	size_t cmd_data_length;
+	size_t resp_data_length;
+	unsigned long flags;
+};
+
+/*
+ * client flags
+ *
+ * CLIENT_NONBLOCKING is set during open() if the file was opened with the
+ *  O_NONBLOCK flag.
+ * CLIENT_XFR_PENDING is set during write() and cleared when all data has been
+ *  read.
+ */
+enum {
+	CLIENT_NONBLOCKING,
+	CLIENT_XFR_PENDING,
+};
+
+struct occ_client {
+	struct occ *occ;
+	struct occ_xfr xfr;
+	spinlock_t lock;
+	wait_queue_head_t wait;
+	size_t read_offset;
+	unsigned long flags;
+};
+
+#define to_client(x)	container_of((x), struct occ_client, xfr)
+
+static struct workqueue_struct *occ_wq;
+
+static DEFINE_IDA(occ_ida);
+
+static void occ_enqueue_xfr(struct occ_xfr *xfr)
+{
+	int empty;
+	struct occ_client *client = to_client(xfr);
+	struct occ *occ = client->occ;
+
+	spin_lock_irq(&occ->list_lock);
+	empty = list_empty(&occ->xfrs);
+	list_add_tail(&xfr->link, &occ->xfrs);
+	spin_unlock(&occ->list_lock);
+
+	if (empty)
+		queue_work(occ_wq, &occ->work);
+}
+
+static int occ_open(struct inode *inode, struct file *file)
+{
+	struct miscdevice *mdev = file->private_data;
+	struct occ *occ = to_occ(mdev);
+	struct occ_client *client = kzalloc(sizeof(*client), GFP_KERNEL);
+
+	if (!client)
+		return -ENOMEM;
+
+	client->occ = occ;
+	spin_lock_init(&client->lock);
+	init_waitqueue_head(&client->wait);
+
+	if (file->f_flags & O_NONBLOCK)
+		set_bit(CLIENT_NONBLOCKING, &client->flags);
+
+	file->private_data = client;
+
+	return 0;
+}
+
+static ssize_t occ_read(struct file *file, char __user *buf, size_t len,
+			loff_t *offset)
+{
+	int rc;
+	size_t bytes;
+	struct occ_client *client = file->private_data;
+	struct occ_xfr *xfr = &client->xfr;
+
+	if (len > OCC_SRAM_BYTES)
+		return -EINVAL;
+
+	spin_lock_irq(&client->lock);
+	if (!test_bit(CLIENT_XFR_PENDING, &client->flags)) {
+		/* we just finished reading all data, return 0 */
+		if (client->read_offset) {
+			rc = 0;
+			client->read_offset = 0;
+		} else
+			rc = -ENOMSG;
+
+		goto done;
+	}
+
+	if (!test_bit(XFR_COMPLETE, &xfr->flags)) {
+		if (test_bit(CLIENT_NONBLOCKING, &client->flags)) {
+			rc = -EAGAIN;
+			goto done;
+		}
+
+		set_bit(XFR_WAITING, &xfr->flags);
+		spin_unlock(&client->lock);
+
+		rc = wait_event_interruptible(client->wait,
+			test_bit(XFR_COMPLETE, &xfr->flags) ||
+			test_bit(XFR_CANCELED, &xfr->flags));
+
+		spin_lock_irq(&client->lock);
+		if (test_bit(XFR_CANCELED, &xfr->flags)) {
+			spin_unlock(&client->lock);
+			kfree(client);
+			return -EBADFD;
+		}
+
+		clear_bit(XFR_WAITING, &xfr->flags);
+		if (!test_bit(XFR_COMPLETE, &xfr->flags)) {
+			rc = -EINTR;
+			goto done;
+		}
+	}
+
+	if (xfr->rc) {
+		rc = xfr->rc;
+		goto done;
+	}
+
+	if (copy_to_user(buf, &xfr->buf[client->read_offset], bytes)) {
+		rc = -EFAULT;
+		goto done;
+	}
+
+	client->read_offset += bytes;
+
+	/* xfr done */
+	if (client->read_offset == xfr->resp_data_length)
+		clear_bit(CLIENT_XFR_PENDING, &client->flags);
+
+	rc = bytes;
+
+done:
+	spin_unlock(&client->lock);
+	return rc;
+}
+
+static ssize_t occ_write(struct file *file, const char __user *buf,
+			 size_t len, loff_t *offset)
+{
+	int rc;
+	unsigned int i;
+	u16 data_length, checksum = 0;
+	struct occ_client *client = file->private_data;
+	struct occ_xfr *xfr = &client->xfr;
+
+	if (len > (OCC_CMD_DATA_BYTES + 3) || len < 3)
+		return -EINVAL;
+
+	spin_lock_irq(&client->lock);
+
+	if (test_bit(CLIENT_XFR_PENDING, &client->flags)) {
+		rc = -EBUSY;
+		goto done;
+	}
+
+	memset(xfr, 0, sizeof(*xfr));	/* clear out the transfer */
+	xfr->buf[0] = 1;		/* occ sequence number */
+
+	/* Assume user data follows the occ command format.
+	 * byte 0: command type
+	 * bytes 1-2: data length (msb first)
+	 * bytes 3-n: data
+	 */
+	if (copy_from_user(&xfr->buf[1], buf, len)) {
+		rc = -EFAULT;
+		goto done;
+	}
+
+	data_length = (xfr->buf[2] << 8) + xfr->buf[3];
+	if (data_length > OCC_CMD_DATA_BYTES) {
+		rc = -EINVAL;
+		goto done;
+	}
+
+	for (i = 0; i < data_length + 4; ++i)
+		checksum += xfr->buf[i];
+
+	xfr->buf[data_length + 4] = checksum >> 8;
+	xfr->buf[data_length + 5] = checksum & 0xFF;
+
+	xfr->cmd_data_length = data_length + 6;
+	client->read_offset = 0;
+	set_bit(CLIENT_XFR_PENDING, &client->flags);
+	occ_enqueue_xfr(xfr);
+
+	rc = len;
+
+done:
+	spin_unlock(&client->lock);
+	return rc;
+}
+
+static int occ_release(struct inode *inode, struct file *file)
+{
+	struct occ_client *client = file->private_data;
+	struct occ_xfr *xfr = &client->xfr;
+	struct occ *occ = client->occ;
+
+	spin_lock_irq(&client->lock);
+	if (!test_bit(CLIENT_XFR_PENDING, &client->flags)) {
+		spin_unlock(&client->lock);
+		kfree(client);
+		return 0;
+	}
+
+	spin_lock_irq(&occ->list_lock);
+	set_bit(XFR_CANCELED, &xfr->flags);
+	if (!test_bit(XFR_IN_PROGRESS, &xfr->flags)) {
+		/* already deleted from list if complete */
+		if (!test_bit(XFR_COMPLETE, &xfr->flags))
+			list_del(&xfr->link);
+
+		spin_unlock(&occ->list_lock);
+
+		if (test_bit(XFR_WAITING, &xfr->flags)) {
+			/* blocking read; let reader clean up */
+			wake_up_interruptible(&client->wait);
+			spin_unlock(&client->lock);
+			return 0;
+		}
+
+		spin_unlock(&client->lock);
+		kfree(client);
+		return 0;
+	}
+
+	/* operation is in progress; let worker clean up*/
+	spin_unlock(&occ->list_lock);
+	spin_unlock(&client->lock);
+	return 0;
+}
+
+static const struct file_operations occ_fops = {
+	.owner = THIS_MODULE,
+	.open = occ_open,
+	.read = occ_read,
+	.write = occ_write,
+	.release = occ_release,
+};
+
+static int occ_write_sbefifo(struct sbefifo_client *client, const char *buf,
+			     ssize_t len)
+{
+	int rc;
+	ssize_t total = 0;
+
+	do {
+		rc = sbefifo_drv_write(client, &buf[total], len - total);
+		if (rc < 0)
+			return rc;
+		else if (!rc)
+			break;
+
+		total += rc;
+	} while (total < len);
+
+	return (total == len) ? 0 : -ENODATA;
+}
+
+static int occ_read_sbefifo(struct sbefifo_client *client, char *buf,
+			    ssize_t len)
+{
+	int rc;
+	ssize_t total = 0;
+
+	do {
+		rc = sbefifo_drv_read(client, &buf[total], len - total);
+		if (rc < 0)
+			return rc;
+		else if (!rc)
+			break;
+
+		total += rc;
+	} while (total < len);
+
+	return (total == len) ? 0 : -ENODATA;
+}
+
+static int occ_getsram(struct device *sbefifo, u32 address, u8 *data,
+		       ssize_t len)
+{
+	int rc;
+	u8 *resp;
+	__be32 buf[5];
+	u32 data_len = ((len + 7) / 8) * 8;	/* must be multiples of 8 B */
+	struct sbefifo_client *client;
+
+	/* Magic sequence to do SBE getsram command. SBE will fetch data from
+	 * specified SRAM address.
+	 */
+	buf[0] = cpu_to_be32(0x5);
+	buf[1] = cpu_to_be32(0xa403);
+	buf[2] = cpu_to_be32(1);
+	buf[3] = cpu_to_be32(address);
+	buf[4] = cpu_to_be32(data_len);
+
+	client = sbefifo_drv_open(sbefifo, 0);
+	if (!client)
+		return -ENODEV;
+
+	rc = occ_write_sbefifo(client, (const char *)buf, sizeof(buf));
+	if (rc)
+		goto done;
+
+	resp = kzalloc(data_len, GFP_KERNEL);
+	if (!resp) {
+		rc = -ENOMEM;
+		goto done;
+	}
+
+	rc = occ_read_sbefifo(client, (char *)resp, data_len);
+	if (rc)
+		goto free;
+
+	/* check for good response */
+	rc = occ_read_sbefifo(client, (char *)buf, 8);
+	if (rc)
+		goto free;
+
+	if ((be32_to_cpu(buf[0]) == data_len) &&
+	    (be32_to_cpu(buf[1]) == 0xC0DEA403))
+		memcpy(data, resp, len);
+	else
+		rc = -EBADMSG;
+
+free:
+	kfree(resp);
+
+done:
+	sbefifo_drv_release(client);
+	return rc;
+}
+
+static int occ_putsram(struct device *sbefifo, u32 address, u8 *data,
+		       ssize_t len)
+{
+	int rc;
+	__be32 *buf;
+	u32 data_len = ((len + 7) / 8) * 8;	/* must be multiples of 8 B */
+	size_t cmd_len = data_len + 20;
+	struct sbefifo_client *client;
+
+	buf = kzalloc(cmd_len, GFP_KERNEL);
+	if (!buf)
+		return -ENOMEM;
+
+	/* Magic sequence to do SBE putsram command. SBE will transfer
+	 * data to specified SRAM address.
+	 */
+	buf[0] = cpu_to_be32(0x5 + (data_len / 4));
+	buf[1] = cpu_to_be32(0xa404);
+	buf[2] = cpu_to_be32(1);
+	buf[3] = cpu_to_be32(address);
+	buf[4] = cpu_to_be32(data_len);
+
+	memcpy(&buf[5], data, len);
+
+	client = sbefifo_drv_open(sbefifo, 0);
+	if (!client) {
+		rc = -ENODEV;
+		goto free;
+	}
+
+	rc = occ_write_sbefifo(client, (const char *)buf, cmd_len);
+	if (rc)
+		goto done;
+
+	rc = occ_read_sbefifo(client, (char *)buf, 8);
+	if (rc)
+		goto done;
+
+	/* check for good response */
+	if ((be32_to_cpu(buf[0]) != data_len) ||
+	    (be32_to_cpu(buf[1]) != 0xC0DEA404))
+		rc = -EBADMSG;
+
+done:
+	sbefifo_drv_release(client);
+free:
+	kfree(buf);
+	return rc;
+}
+
+static int occ_trigger_attn(struct device *sbefifo)
+{
+	int rc;
+	__be32 buf[6];
+	struct sbefifo_client *client;
+
+	/* Magic sequence to do SBE putscom command. SBE will write 8 bytes to
+	 * specified SCOM address.
+	 */
+	buf[0] = cpu_to_be32(0x6);
+	buf[1] = cpu_to_be32(0xa202);
+	buf[2] = 0;
+	buf[3] = cpu_to_be32(0x6D035);
+	buf[4] = cpu_to_be32(0x20010000);	/* trigger occ attention */
+	buf[5] = 0;
+
+	client = sbefifo_drv_open(sbefifo, 0);
+	if (!client)
+		return -ENODEV;
+
+	rc = occ_write_sbefifo(client, (const char *)buf, sizeof(buf));
+	if (rc)
+		goto done;
+
+	rc = occ_read_sbefifo(client, (char *)buf, 8);
+	if (rc)
+		goto done;
+
+	/* check for good response */
+	if ((be32_to_cpu(buf[0]) != 0xC0DEA202) ||
+	    (be32_to_cpu(buf[1]) & 0x0FFFFFFF))
+		rc = -EBADMSG;
+
+done:
+	sbefifo_drv_release(client);
+
+	return rc;
+}
+
+static void occ_worker(struct work_struct *work)
+{
+	int rc = 0, empty, waiting, canceled;
+	u16 resp_data_length;
+	unsigned long start;
+	const unsigned long timeout = msecs_to_jiffies(OCC_TIMEOUT_MS);
+	const long int wait_time = msecs_to_jiffies(OCC_CMD_IN_PRG_WAIT_MS);
+	struct occ_xfr *xfr;
+	struct occ_response *resp;
+	struct occ_client *client;
+	struct occ *occ = container_of(work, struct occ, work);
+	struct device *sbefifo = occ->sbefifo;
+
+again:
+	spin_lock_irq(&occ->list_lock);
+	xfr = list_first_entry(&occ->xfrs, struct occ_xfr, link);
+	if (!xfr) {
+		spin_unlock(&occ->list_lock);
+		return;
+	}
+
+	resp = (struct occ_response *)xfr->buf;
+	set_bit(XFR_IN_PROGRESS, &xfr->flags);
+
+	spin_unlock(&occ->list_lock);
+	mutex_lock(&occ->occ_lock);
+
+	/* write occ command */
+	start = jiffies;
+	rc = occ_putsram(sbefifo, 0xFFFBE000, xfr->buf,
+			 xfr->cmd_data_length);
+	if (rc)
+		goto done;
+
+	rc = occ_trigger_attn(sbefifo);
+	if (rc)
+		goto done;
+
+	/* read occ response */
+	do {
+		rc = occ_getsram(sbefifo, 0xFFFBF000, xfr->buf, 8);
+		if (rc)
+			goto done;
+
+		if (resp->return_status == OCC_RESP_CMD_IN_PRG) {
+			rc = -EALREADY;
+
+			if (time_after(jiffies, start + timeout))
+				break;
+
+			set_current_state(TASK_INTERRUPTIBLE);
+			schedule_timeout(wait_time);
+		}
+	} while (rc);
+
+	resp_data_length = get_unaligned_be16(&resp->data_length);
+	if (resp_data_length > OCC_RESP_DATA_BYTES) {
+		rc = -EMSGSIZE;
+		goto done;
+	}
+
+	if (resp_data_length > 1) {
+		/* already got 3 bytes resp, also need 2 bytes checksum */
+		rc = occ_getsram(sbefifo, 0xFFFBF008, &xfr->buf[8],
+				 resp_data_length - 1);
+		if (rc)
+			goto done;
+	}
+
+	xfr->resp_data_length = resp_data_length + 7;
+
+done:
+	mutex_unlock(&occ->occ_lock);
+
+	xfr->rc = rc;
+	client = to_client(xfr);
+
+	/* lock client to prevent race with read() */
+	spin_lock_irq(&client->lock);
+	set_bit(XFR_COMPLETE, &xfr->flags);
+	waiting = test_bit(XFR_WAITING, &xfr->flags);
+	spin_unlock(&client->lock);
+
+	spin_lock_irq(&occ->list_lock);
+	clear_bit(XFR_IN_PROGRESS, &xfr->flags);
+	list_del(&xfr->link);
+	empty = list_empty(&occ->xfrs);
+	canceled = test_bit(XFR_CANCELED, &xfr->flags);
+	spin_unlock(&occ->list_lock);
+
+	if (waiting)
+		wake_up_interruptible(&client->wait);
+	else if (canceled)
+		kfree(client);
+
+	if (!empty)
+		goto again;
+}
+
+static int occ_probe(struct platform_device *pdev)
+{
+	int rc;
+	u32 reg;
+	struct occ *occ;
+	struct device *dev = &pdev->dev;
+
+	occ = devm_kzalloc(dev, sizeof(*occ), GFP_KERNEL);
+	if (!occ)
+		return -ENOMEM;
+
+	occ->sbefifo = dev->parent;
+	INIT_LIST_HEAD(&occ->xfrs);
+	spin_lock_init(&occ->list_lock);
+	mutex_init(&occ->occ_lock);
+	INIT_WORK(&occ->work, occ_worker);
+
+	if (dev->of_node) {
+		rc = of_property_read_u32(dev->of_node, "reg", &reg);
+		if (!rc) {
+			/* make sure we don't have a duplicate from dts */
+			occ->idx = ida_simple_get(&occ_ida, reg, reg + 1,
+						  GFP_KERNEL);
+			if (occ->idx < 0)
+				occ->idx = ida_simple_get(&occ_ida, 1, INT_MAX,
+							  GFP_KERNEL);
+		} else
+			occ->idx = ida_simple_get(&occ_ida, 1, INT_MAX,
+						  GFP_KERNEL);
+	} else
+		occ->idx = ida_simple_get(&occ_ida, 1, INT_MAX, GFP_KERNEL);
+
+	snprintf(occ->name, sizeof(occ->name), "occ%d", occ->idx);
+	occ->mdev.fops = &occ_fops;
+	occ->mdev.minor = MISC_DYNAMIC_MINOR;
+	occ->mdev.name = occ->name;
+	occ->mdev.parent = dev;
+
+	rc = misc_register(&occ->mdev);
+	if (rc) {
+		dev_err(dev, "failed to register miscdevice\n");
+		ida_simple_remove(&occ_ida, occ->idx);
+		return rc;
+	}
+
+	platform_set_drvdata(pdev, occ);
+
+	return 0;
+}
+
+static int occ_remove(struct platform_device *pdev)
+{
+	struct occ *occ = platform_get_drvdata(pdev);
+
+	flush_work(&occ->work);
+	misc_deregister(&occ->mdev);
+	ida_simple_remove(&occ_ida, occ->idx);
+
+	return 0;
+}
+
+static const struct of_device_id occ_match[] = {
+	{ .compatible = "ibm,p9-occ" },
+	{ },
+};
+
+static struct platform_driver occ_driver = {
+	.driver = {
+		.name = "occ",
+		.of_match_table	= occ_match,
+	},
+	.probe	= occ_probe,
+	.remove = occ_remove,
+};
+
+static int occ_init(void)
+{
+	occ_wq = create_singlethread_workqueue("occ");
+	if (!occ_wq)
+		return -ENOMEM;
+
+	return platform_driver_register(&occ_driver);
+}
+
+static void occ_exit(void)
+{
+	destroy_workqueue(occ_wq);
+
+	platform_driver_unregister(&occ_driver);
+
+	ida_destroy(&occ_ida);
+}
+
+module_init(occ_init);
+module_exit(occ_exit);
+
+MODULE_AUTHOR("Eddie James <eajames@us.ibm.com>");
+MODULE_DESCRIPTION("BMC P9 OCC driver");
+MODULE_LICENSE("GPL");
-- 
1.8.3.1

[PATCH v2 4/5] drivers/fsi/occ: Add in-kernel API

[Eddie James]

From: "Edward A. James" <eajames@us.ibm.com>

Add an in-kernel read/write API with exported functions. This is
necessary for other drivers which want to directly interact with the
OCC. Also parse the OCC device tree node for child nodes and create
child platform devices accordingly.

Signed-off-by: Edward A. James <eajames@us.ibm.com>
---
 drivers/fsi/occ.c   | 135 +++++++++++++++++++++++++++++++++++++++++++---------
 include/linux/occ.h |  41 ++++++++++++++++
 2 files changed, 154 insertions(+), 22 deletions(-)
 create mode 100644 include/linux/occ.h

diff --git a/drivers/fsi/occ.c b/drivers/fsi/occ.c
index df13dcb..4fea39b 100644
--- a/drivers/fsi/occ.c
+++ b/drivers/fsi/occ.c
@@ -19,6 +19,7 @@
 #include <linux/miscdevice.h>
 #include <linux/module.h>
 #include <linux/mutex.h>
+#include <linux/occ.h>
 #include <linux/of.h>
 #include <linux/platform_device.h>
 #include <linux/sched.h>
@@ -32,8 +33,6 @@
 #define OCC_CMD_DATA_BYTES	4090
 #define OCC_RESP_DATA_BYTES	4089
 
-#define OCC_RESP_CMD_IN_PRG	0xFF
-
 #define OCC_TIMEOUT_MS		1000
 #define OCC_CMD_IN_PRG_WAIT_MS	50
 
@@ -132,33 +131,43 @@ static void occ_enqueue_xfr(struct occ_xfr *xfr)
 		queue_work(occ_wq, &occ->work);
 }
 
-static int occ_open(struct inode *inode, struct file *file)
+static struct occ_client *occ_open_common(struct occ *occ, unsigned long flags)
 {
-	struct miscdevice *mdev = file->private_data;
-	struct occ *occ = to_occ(mdev);
 	struct occ_client *client = kzalloc(sizeof(*client), GFP_KERNEL);
 
 	if (!client)
-		return -ENOMEM;
+		return NULL;
 
 	client->occ = occ;
 	spin_lock_init(&client->lock);
 	init_waitqueue_head(&client->wait);
 
-	if (file->f_flags & O_NONBLOCK)
+	if (flags & O_NONBLOCK)
 		set_bit(CLIENT_NONBLOCKING, &client->flags);
 
+	return client;
+}
+
+static int occ_open(struct inode *inode, struct file *file)
+{
+	struct occ_client *client;
+	struct miscdevice *mdev = file->private_data;
+	struct occ *occ = to_occ(mdev);
+
+	client = occ_open_common(occ, file->f_flags);
+	if (!client)
+		return -ENOMEM;
+
 	file->private_data = client;
 
 	return 0;
 }
 
-static ssize_t occ_read(struct file *file, char __user *buf, size_t len,
-			loff_t *offset)
+static ssize_t occ_read_common(struct occ_client *client, char __user *ubuf,
+			       char *kbuf, size_t len)
 {
 	int rc;
 	size_t bytes;
-	struct occ_client *client = file->private_data;
 	struct occ_xfr *xfr = &client->xfr;
 
 	if (len > OCC_SRAM_BYTES)
@@ -208,10 +217,15 @@ static ssize_t occ_read(struct file *file, char __user *buf, size_t len,
 		goto done;
 	}
 
-	if (copy_to_user(buf, &xfr->buf[client->read_offset], bytes)) {
-		rc = -EFAULT;
-		goto done;
-	}
+	bytes = min(len, xfr->resp_data_length - client->read_offset);
+	if (ubuf) {
+		if (copy_to_user(ubuf, &xfr->buf[client->read_offset],
+				 bytes)) {
+			rc = -EFAULT;
+			goto done;
+		}
+	} else
+		memcpy(kbuf, &xfr->buf[client->read_offset], bytes);
 
 	client->read_offset += bytes;
 
@@ -226,13 +240,21 @@ static ssize_t occ_read(struct file *file, char __user *buf, size_t len,
 	return rc;
 }
 
-static ssize_t occ_write(struct file *file, const char __user *buf,
-			 size_t len, loff_t *offset)
+static ssize_t occ_read(struct file *file, char __user *buf, size_t len,
+			loff_t *offset)
+{
+	struct occ_client *client = file->private_data;
+
+	return occ_read_common(client, buf, NULL, len);
+}
+
+static ssize_t occ_write_common(struct occ_client *client,
+				const char __user *ubuf, const char *kbuf,
+				size_t len)
 {
 	int rc;
 	unsigned int i;
 	u16 data_length, checksum = 0;
-	struct occ_client *client = file->private_data;
 	struct occ_xfr *xfr = &client->xfr;
 
 	if (len > (OCC_CMD_DATA_BYTES + 3) || len < 3)
@@ -253,10 +275,13 @@ static ssize_t occ_write(struct file *file, const char __user *buf,
 	 * bytes 1-2: data length (msb first)
 	 * bytes 3-n: data
 	 */
-	if (copy_from_user(&xfr->buf[1], buf, len)) {
-		rc = -EFAULT;
-		goto done;
-	}
+	if (ubuf) {
+		if (copy_from_user(&xfr->buf[1], ubuf, len)) {
+			rc = -EFAULT;
+			goto done;
+		}
+	} else
+		memcpy(&xfr->buf[1], kbuf, len);
 
 	data_length = (xfr->buf[2] << 8) + xfr->buf[3];
 	if (data_length > OCC_CMD_DATA_BYTES) {
@@ -282,9 +307,16 @@ static ssize_t occ_write(struct file *file, const char __user *buf,
 	return rc;
 }
 
-static int occ_release(struct inode *inode, struct file *file)
+static ssize_t occ_write(struct file *file, const char __user *buf,
+			 size_t len, loff_t *offset)
 {
 	struct occ_client *client = file->private_data;
+
+	return occ_write_common(client, buf, NULL, len);
+}
+
+static int occ_release_common(struct occ_client *client)
+{
 	struct occ_xfr *xfr = &client->xfr;
 	struct occ *occ = client->occ;
 
@@ -322,6 +354,13 @@ static int occ_release(struct inode *inode, struct file *file)
 	return 0;
 }
 
+static int occ_release(struct inode *inode, struct file *file)
+{
+	struct occ_client *client = file->private_data;
+
+	return occ_release_common(client);
+}
+
 static const struct file_operations occ_fops = {
 	.owner = THIS_MODULE,
 	.open = occ_open,
@@ -611,11 +650,51 @@ static void occ_worker(struct work_struct *work)
 		goto again;
 }
 
+struct occ_client *occ_drv_open(struct device *dev, unsigned long flags)
+{
+	struct occ *occ = dev_get_drvdata(dev);
+
+	if (!occ)
+		return NULL;
+
+	return occ_open_common(occ, flags);
+}
+EXPORT_SYMBOL_GPL(occ_drv_open);
+
+int occ_drv_read(struct occ_client *client, char *buf, size_t len)
+{
+	return occ_read_common(client, NULL, buf, len);
+}
+EXPORT_SYMBOL_GPL(occ_drv_read);
+
+int occ_drv_write(struct occ_client *client, const char *buf, size_t len)
+{
+	return occ_write_common(client, NULL, buf, len);
+}
+EXPORT_SYMBOL_GPL(occ_drv_write);
+
+void occ_drv_release(struct occ_client *client)
+{
+	occ_release_common(client);
+}
+EXPORT_SYMBOL_GPL(occ_drv_release);
+
+static int occ_unregister_child(struct device *dev, void *data)
+{
+	struct platform_device *child = to_platform_device(dev);
+
+	platform_device_unregister(child);
+
+	return 0;
+}
+
 static int occ_probe(struct platform_device *pdev)
 {
 	int rc;
 	u32 reg;
 	struct occ *occ;
+	struct platform_device *child;
+	struct platform_device_info child_info;
 	struct device *dev = &pdev->dev;
 
 	occ = devm_kzalloc(dev, sizeof(*occ), GFP_KERNEL);
@@ -627,8 +706,11 @@ static int occ_probe(struct platform_device *pdev)
 	spin_lock_init(&occ->list_lock);
 	mutex_init(&occ->occ_lock);
 	INIT_WORK(&occ->work, occ_worker);
+	memset(&child_info, 0, sizeof(child_info));
 
 	if (dev->of_node) {
+		child_info.fwnode = &dev->of_node->fwnode;
+
 		rc = of_property_read_u32(dev->of_node, "reg", &reg);
 		if (!rc) {
 			/* make sure we don't have a duplicate from dts */
@@ -656,6 +738,14 @@ static int occ_probe(struct platform_device *pdev)
 		return rc;
 	}
 
+	/* create device for hwmon driver */
+	child_info.id = occ->idx;
+	child_info.parent = dev;
+	child_info.name = "occ-hwmon";
+	child = platform_device_register_full(&child_info);
+	if (!child)
+		dev_warn(dev, "failed to register %s dev\n", child_info.name);
+
 	platform_set_drvdata(pdev, occ);
 
 	return 0;
@@ -667,6 +757,7 @@ static int occ_remove(struct platform_device *pdev)
 
 	flush_work(&occ->work);
 	misc_deregister(&occ->mdev);
+	device_for_each_child(&pdev->dev, NULL, occ_unregister_child);
 	ida_simple_remove(&occ_ida, occ->idx);
 
 	return 0;
diff --git a/include/linux/occ.h b/include/linux/occ.h
new file mode 100644
index 0000000..1ba80de
--- /dev/null
+++ b/include/linux/occ.h
@@ -0,0 +1,41 @@
+/*
+ * Copyright (C) IBM Corporation 2017
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERGCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#ifndef LINUX_OCC_H
+#define LINUX_OCC_H
+
+struct device;
+struct occ_client;
+
+#define OCC_RESP_CMD_IN_PRG		0xFF
+#define OCC_RESP_SUCCESS		0
+#define OCC_RESP_CMD_INVAL		0x11
+#define OCC_RESP_CMD_LEN_INVAL		0x12
+#define OCC_RESP_DATA_INVAL		0x13
+#define OCC_RESP_CHKSUM_ERR		0x14
+#define OCC_RESP_INT_ERR		0x15
+#define OCC_RESP_BAD_STATE		0x16
+#define OCC_RESP_CRIT_EXCEPT		0xE0
+#define OCC_RESP_CRIT_INIT		0xE1
+#define OCC_RESP_CRIT_WATCHDOG		0xE2
+#define OCC_RESP_CRIT_OCB		0xE3
+#define OCC_RESP_CRIT_HW		0xE4
+
+extern struct occ_client *occ_drv_open(struct device *dev,
+				       unsigned long flags);
+extern int occ_drv_read(struct occ_client *client, char *buf, size_t len);
+extern int occ_drv_write(struct occ_client *client, const char *buf,
+			 size_t len);
+extern void occ_drv_release(struct occ_client *client);
+
+#endif /* LINUX_OCC_H */
-- 
1.8.3.1

[PATCH v2 5/5] Documentation/devicetree/bindings: Add FSI device documentation

[Eddie James]

From: "Edward A. James" <eajames@us.ibm.com>

Document the bindings for the SBE and OCC devices. SBE devices are
located on a CFAM on an FSI bus, and OCC devices are accessed through
the SBEFIFO.

Signed-off-by: Edward A. James <eajames@us.ibm.com>
---
 Documentation/devicetree/bindings/fsi/ibm,p9-occ.txt | 15 +++++++++++++++
 .../devicetree/bindings/fsi/ibm,p9-sbefifo.txt       | 20 ++++++++++++++++++++
 2 files changed, 35 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/fsi/ibm,p9-occ.txt
 create mode 100644 Documentation/devicetree/bindings/fsi/ibm,p9-sbefifo.txt

diff --git a/Documentation/devicetree/bindings/fsi/ibm,p9-occ.txt b/Documentation/devicetree/bindings/fsi/ibm,p9-occ.txt
new file mode 100644
index 0000000..766b706
--- /dev/null
+++ b/Documentation/devicetree/bindings/fsi/ibm,p9-occ.txt
@@ -0,0 +1,15 @@
+Device-tree bindings for FSI-based On-Chip Controller driver
+------------------------------------------------------------
+
+Required properties:
+ - compatible = "ibm,p9-occ";
+
+Optional properties:
+ - reg = <processor index>;	: Index for the processor this OCC is on.
+
+Examples:
+
+    occ@1 {
+        compatible = "ibm,p9-occ";
+        reg = <1>;
+    };
diff --git a/Documentation/devicetree/bindings/fsi/ibm,p9-sbefifo.txt b/Documentation/devicetree/bindings/fsi/ibm,p9-sbefifo.txt
new file mode 100644
index 0000000..d16ab9d
--- /dev/null
+++ b/Documentation/devicetree/bindings/fsi/ibm,p9-sbefifo.txt
@@ -0,0 +1,20 @@
+Device-tree bindings for FSI-based SBEFIFO driver
+-------------------------------------------------
+
+Required properties:
+ - reg = < address size >		: FSI CFAM address for the SBE engine
+					  and address space size.
+
+Optional properties:
+ - <child nodes>			: Devices that are accessible through
+					  the SBEFIFO.
+
+Examples:
+
+    sbefifo@2400 {
+        reg = < 0x2400 0x400 >;
+
+        occ@1 {
+            ...
+        };
+    };
-- 
1.8.3.1

Re: [PATCH v2 5/5] Documentation/devicetree/bindings: Add FSI device documentation

[Rob Herring]

On Wed, Jul 05, 2017 at 04:36:57PM -0500, Eddie James wrote:
> From: "Edward A. James" <eajames@us.ibm.com>

"dt-bindings: fsi: ..." for the subject.

> 
> Document the bindings for the SBE and OCC devices. SBE devices are
> located on a CFAM on an FSI bus, and OCC devices are accessed through
> the SBEFIFO.
> 
> Signed-off-by: Edward A. James <eajames@us.ibm.com>
> ---
>  Documentation/devicetree/bindings/fsi/ibm,p9-occ.txt | 15 +++++++++++++++
>  .../devicetree/bindings/fsi/ibm,p9-sbefifo.txt       | 20 ++++++++++++++++++++
>  2 files changed, 35 insertions(+)
>  create mode 100644 Documentation/devicetree/bindings/fsi/ibm,p9-occ.txt
>  create mode 100644 Documentation/devicetree/bindings/fsi/ibm,p9-sbefifo.txt
> 
> diff --git a/Documentation/devicetree/bindings/fsi/ibm,p9-occ.txt b/Documentation/devicetree/bindings/fsi/ibm,p9-occ.txt
> new file mode 100644
> index 0000000..766b706
> --- /dev/null
> +++ b/Documentation/devicetree/bindings/fsi/ibm,p9-occ.txt
> @@ -0,0 +1,15 @@
> +Device-tree bindings for FSI-based On-Chip Controller driver
> +------------------------------------------------------------

Bindings are for h/w, not drivers.

> +
> +Required properties:
> + - compatible = "ibm,p9-occ";
> +
> +Optional properties:
> + - reg = <processor index>;	: Index for the processor this OCC is on.
> +
> +Examples:
> +
> +    occ@1 {
> +        compatible = "ibm,p9-occ";
> +        reg = <1>;
> +    };
> diff --git a/Documentation/devicetree/bindings/fsi/ibm,p9-sbefifo.txt b/Documentation/devicetree/bindings/fsi/ibm,p9-sbefifo.txt
> new file mode 100644
> index 0000000..d16ab9d
> --- /dev/null
> +++ b/Documentation/devicetree/bindings/fsi/ibm,p9-sbefifo.txt
> @@ -0,0 +1,20 @@
> +Device-tree bindings for FSI-based SBEFIFO driver
> +-------------------------------------------------

Bindings are for h/w, not drivers.

> +
> +Required properties:
> + - reg = < address size >		: FSI CFAM address for the SBE engine
> +					  and address space size.
> +
> +Optional properties:
> + - <child nodes>			: Devices that are accessible through
> +					  the SBEFIFO.
> +
> +Examples:
> +
> +    sbefifo@2400 {

This should have a compatible.

> +        reg = < 0x2400 0x400 >;
> +
> +        occ@1 {
> +            ...
> +        };
> +    };
> -- 
> 1.8.3.1
> 

Re: [PATCH v2 2/5] drivers/fsi/sbefifo: Add in-kernel API

[kbuild test robot]

[-- Attachment #1: Type: text/plain, Size: 4969 bytes --]

Hi Edward,

[auto build test ERROR on linus/master]
[also build test ERROR on next-20170712]
[cannot apply to v4.12]
[if your patch is applied to the wrong git tree, please drop us a note to help improve the system]

url:    https://github.com/0day-ci/linux/commits/Eddie-James/drivers-fsi-Add-SBEFIFO-and-OCC-drivers/20170707-232144
config: um-allyesconfig (attached as .config)
compiler: gcc-6 (Debian 6.2.0-3) 6.2.0 20160901
reproduce:
        # save the attached .config to linux build tree
        make ARCH=um 

All errors (new ones prefixed by >>):

   arch/um/drivers/built-in.o: In function `vde_open_real':
   (.text+0xc9f1): warning: Using 'getgrnam' in statically linked applications requires at runtime the shared libraries from the glibc version used for linking
   arch/um/drivers/built-in.o: In function `vde_open_real':
   (.text+0xc83c): warning: Using 'getpwuid' in statically linked applications requires at runtime the shared libraries from the glibc version used for linking
   arch/um/drivers/built-in.o: In function `vde_open_real':
   (.text+0xcb55): warning: Using 'getaddrinfo' in statically linked applications requires at runtime the shared libraries from the glibc version used for linking
   arch/um/drivers/built-in.o: In function `pcap_nametoaddr':
   (.text+0x1d5e5): warning: Using 'gethostbyname' in statically linked applications requires at runtime the shared libraries from the glibc version used for linking
   arch/um/drivers/built-in.o: In function `pcap_nametonetaddr':
   (.text+0x1d685): warning: Using 'getnetbyname' in statically linked applications requires at runtime the shared libraries from the glibc version used for linking
   arch/um/drivers/built-in.o: In function `pcap_nametoproto':
   (.text+0x1d8a5): warning: Using 'getprotobyname' in statically linked applications requires at runtime the shared libraries from the glibc version used for linking
   arch/um/drivers/built-in.o: In function `pcap_nametoport':
   (.text+0x1d6d7): warning: Using 'getservbyname' in statically linked applications requires at runtime the shared libraries from the glibc version used for linking
   drivers/built-in.o: In function `img_ascii_lcd_probe':
   drivers/auxdisplay/img-ascii-lcd.c:386: undefined reference to `devm_ioremap_resource'
   drivers/built-in.o: In function `sbefifo_probe':
>> drivers/fsi/fsi-sbefifo.c:853: undefined reference to `of_platform_device_create'
   collect2: error: ld returned 1 exit status

vim +853 drivers/fsi/fsi-sbefifo.c

   792	
   793	static int sbefifo_probe(struct device *dev)
   794	{
   795		struct fsi_device *fsi_dev = to_fsi_dev(dev);
   796		struct sbefifo *sbefifo;
   797		struct device_node *np;
   798		struct platform_device *child;
   799		char child_name[32];
   800		u32 sts;
   801		int ret, child_idx = 0;
   802	
   803		sbefifo = kzalloc(sizeof(*sbefifo), GFP_KERNEL);
   804		if (!sbefifo)
   805			return -ENOMEM;
   806	
   807		sbefifo->fsi_dev = fsi_dev;
   808	
   809		ret = sbefifo_inw(sbefifo, SBEFIFO_UP | SBEFIFO_STS, &sts);
   810		if (ret)
   811			return ret;
   812	
   813		if (!(sts & SBEFIFO_EMPTY)) {
   814			dev_err(&sbefifo->fsi_dev->dev,
   815				"Found data in upstream fifo\n");
   816			return -EIO;
   817		}
   818	
   819		ret = sbefifo_inw(sbefifo, SBEFIFO_DWN | SBEFIFO_STS, &sts);
   820		if (ret)
   821			return ret;
   822	
   823		if (!(sts & SBEFIFO_EMPTY)) {
   824			dev_err(&sbefifo->fsi_dev->dev,
   825				"Found data in downstream fifo\n");
   826			return -EIO;
   827		}
   828	
   829		sbefifo->mdev.minor = MISC_DYNAMIC_MINOR;
   830		sbefifo->mdev.fops = &sbefifo_fops;
   831		sbefifo->mdev.name = sbefifo->name;
   832		sbefifo->mdev.parent = dev;
   833		spin_lock_init(&sbefifo->lock);
   834		kref_init(&sbefifo->kref);
   835	
   836		sbefifo->idx = ida_simple_get(&sbefifo_ida, 1, INT_MAX, GFP_KERNEL);
   837		snprintf(sbefifo->name, sizeof(sbefifo->name), "sbefifo%d",
   838			 sbefifo->idx);
   839		init_waitqueue_head(&sbefifo->wait);
   840		INIT_LIST_HEAD(&sbefifo->xfrs);
   841	
   842		/* This bit of silicon doesn't offer any interrupts... */
   843		setup_timer(&sbefifo->poll_timer, sbefifo_poll_timer,
   844			    (unsigned long)sbefifo);
   845	
   846		list_add(&sbefifo->link, &sbefifo_fifos);
   847	
   848		if (dev->of_node) {
   849			/* create platform devs for dts child nodes (occ, etc) */
   850			for_each_child_of_node(dev->of_node, np) {
   851				snprintf(child_name, sizeof(child_name), "%s-dev%d",
   852					 sbefifo->name, child_idx++);
 > 853				child = of_platform_device_create(np, child_name, dev);
   854				if (!child)
   855					dev_warn(&sbefifo->fsi_dev->dev,
   856						 "failed to create child node dev\n");
   857			}
   858		}
   859	
   860		return misc_register(&sbefifo->mdev);
   861	}
   862	

---
0-DAY kernel test infrastructure                Open Source Technology Center
https://lists.01.org/pipermail/kbuild-all                   Intel Corporation

[-- Attachment #2: .config.gz --]
[-- Type: application/gzip, Size: 19403 bytes --]