The Linux Kernel Mailing List
 help / color / mirror / Atom feed
* [PATCH 1/2] Eliminate references to new_client in i2c documentation
@ 2007-05-10 21:04 Grant Likely
  2007-05-10 21:06 ` [PATCH 2/2] Add driver for Dallas DS1682 elapsed time recorder Grant Likely
  2007-05-11  8:01 ` [PATCH 1/2] Eliminate references to new_client in i2c documentation Jean Delvare
  0 siblings, 2 replies; 4+ messages in thread
From: Grant Likely @ 2007-05-10 21:04 UTC (permalink / raw)
  To: Jean Delvare, i2c, linux-kernel

The use of 'new_client' in i2c device allocation has been refered to as
a 'disease'.  Replace all occurances of 'struct i2c_client *new_client'
in documentation examples with 'struct i2c_client *client'.

Also remove unnecessary zero initializtion of .flags since kzalloc is used.

Signed-off-by: Grant Likely <grant.likely@secretlab.ca>
---

 Documentation/i2c/writing-clients |   29 ++++++++++++++---------------
 1 files changed, 14 insertions(+), 15 deletions(-)

diff --git a/Documentation/i2c/writing-clients b/Documentation/i2c/writing-clients
index fbcff96..bd61480 100644
--- a/Documentation/i2c/writing-clients
+++ b/Documentation/i2c/writing-clients
@@ -260,7 +260,7 @@ For now, you can ignore the `flags' parameter. It is there for future use.
   {
     int err = 0;
     int i;
-    struct i2c_client *new_client;
+    struct i2c_client *client;
     struct foo_data *data;
     const char *client_name = ""; /* For non-`sensors' drivers, put the real
                                      name here! */
@@ -323,13 +323,12 @@ For now, you can ignore the `flags' parameter. It is there for future use.
       goto ERROR0;
     }
 
-    new_client = &data->client;
-    i2c_set_clientdata(new_client, data);
+    client = &data->client;
+    i2c_set_clientdata(client, data);
 
-    new_client->addr = address;
-    new_client->adapter = adapter;
-    new_client->driver = &foo_driver;
-    new_client->flags = 0;
+    client->addr = address;
+    client->adapter = adapter;
+    client->driver = &foo_driver;
 
     /* Now, we do the remaining detection. If no `force' parameter is used. */
 
@@ -337,7 +336,7 @@ For now, you can ignore the `flags' parameter. It is there for future use.
        parameter was used. */
     if (kind < 0) {
       /* The below is of course bogus */
-      if (foo_read(new_client,FOO_REG_GENERIC) != FOO_GENERIC_VALUE)
+      if (foo_read(client,FOO_REG_GENERIC) != FOO_GENERIC_VALUE)
          goto ERROR1;
     }
 
@@ -349,7 +348,7 @@ For now, you can ignore the `flags' parameter. It is there for future use.
     /* Determine the chip type. Not needed if a `force_CHIPTYPE' parameter
        was used. */
     if (kind <= 0) {
-      i = foo_read(new_client,FOO_REG_CHIPTYPE);
+      i = foo_read(client,FOO_REG_CHIPTYPE);
       if (i == FOO_TYPE_1) 
         kind = chip1; /* As defined in the enum */
       else if (i == FOO_TYPE_2)
@@ -377,7 +376,7 @@ For now, you can ignore the `flags' parameter. It is there for future use.
     /* SENSORS ONLY END */
 
     /* Fill in the remaining client fields. */
-    strcpy(new_client->name,client_name);
+    strcpy(client->name,client_name);
 
     /* SENSORS ONLY BEGIN */
     data->type = kind;
@@ -389,14 +388,14 @@ For now, you can ignore the `flags' parameter. It is there for future use.
     /* Any other initializations in data must be done here too. */
 
     /* Tell the i2c layer a new client has arrived */
-    if ((err = i2c_attach_client(new_client)))
+    if ((err = i2c_attach_client(client)))
       goto ERROR3;
 
     /* SENSORS ONLY BEGIN */
     /* Register a new directory entry with module sensors. See below for
        the `template' structure. */
-    if ((i = i2c_register_entry(new_client, type_name,
-                                    foo_dir_table_template,THIS_MODULE)) < 0) {
+    if ((i = i2c_register_entry(client, type_name,
+                                foo_dir_table_template,THIS_MODULE)) < 0) {
       err = i;
       goto ERROR4;
     }
@@ -406,14 +405,14 @@ For now, you can ignore the `flags' parameter. It is there for future use.
 
     /* This function can write default values to the client registers, if
        needed. */
-    foo_init_client(new_client);
+    foo_init_client(client);
     return 0;
 
     /* OK, this is not exactly good programming practice, usually. But it is
        very code-efficient in this case. */
 
     ERROR4:
-      i2c_detach_client(new_client);
+      i2c_detach_client(client);
     ERROR3:
     ERROR2:
     /* SENSORS ONLY START */


^ permalink raw reply related	[flat|nested] 4+ messages in thread

* [PATCH 2/2] Add driver for Dallas DS1682 elapsed time recorder
  2007-05-10 21:04 [PATCH 1/2] Eliminate references to new_client in i2c documentation Grant Likely
@ 2007-05-10 21:06 ` Grant Likely
  2007-05-11  8:01 ` [PATCH 1/2] Eliminate references to new_client in i2c documentation Jean Delvare
  1 sibling, 0 replies; 4+ messages in thread
From: Grant Likely @ 2007-05-10 21:06 UTC (permalink / raw)
  To: Jean Delvare, i2c, linux-kernel

Signed-off-by: Grant Likely <grant.likely@secretlab.ca>
---

Updated patch based on comments received from mailing list.  I think
I've addressed all the comments except the idea of moving this to a
new style i2c driver.

 drivers/i2c/chips/ds1682.c |  315 ++++++++++++++++++++++++++++++++++++++++++++
 1 files changed, 315 insertions(+), 0 deletions(-)

diff --git a/drivers/i2c/chips/ds1682.c b/drivers/i2c/chips/ds1682.c
new file mode 100644
index 0000000..b7f33e5
--- /dev/null
+++ b/drivers/i2c/chips/ds1682.c
@@ -0,0 +1,315 @@
+/*
+ * Dallas Semiconductor DS1682 Elapsed Time Recorder device driver
+ *
+ * Written by: Grant Likely <grant.likely@secretlab.ca>
+ *
+ * Copyright (C) 2007 Secret Lab Technologies Ltd.
+ * Copyright (C) 2005 James Chapman <jchapman@katalix.com>
+ * Copyright (C) 2000 Russell King
+ *
+ * Derived from ds1337 real time clock device driver
+ *
+ * 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.
+ */
+
+/*
+ * The DS1682 elapsed timer recorder is a simple device that implements
+ * one elapsed time counter, one event counter, an alarm signal and 10
+ * bytes of general purpose EEPROM.
+ *
+ * This driver provides access to the DS1682 counters and user data via
+ * the sysfs.  The following attributes are added to the device node:
+ *     elapsed_time (u32): Total elapsed event time in ms resolution
+ *     alarm_time (u32): When elapsed time exceeds the value in alarm_time,
+ *                       then the alarm pin is asserted.
+ *     event_count (u16): number of times the event pin has gone low.
+ *     eeprom (u8[10]): general purpose EEPROM
+ *
+ * Counter registers and user data are both read/write unless the device
+ * has been write protected.  This driver does not support turning off write
+ * protection.  Once write protection is turned on, it is impossible to
+ * turn it off again, so I have left the feature out of this driver to avoid
+ * accidental enabling, but it is trivial to add write protect support.
+ *
+ */
+
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/slab.h>
+#include <linux/i2c.h>
+#include <linux/string.h>
+#include <linux/list.h>
+#include <linux/sysfs.h>
+#include <linux/ctype.h>
+#include <linux/hwmon-sysfs.h>
+
+/* Device registers */
+#define DS1682_REG_CONFIG		0x00
+#define DS1682_REG_ALARM		0x01
+#define DS1682_REG_ELAPSED		0x05
+#define DS1682_REG_EVT_CNTR		0x09
+#define DS1682_REG_EEPROM		0x0b
+#define DS1682_REG_RESET		0x1d
+#define DS1682_REG_WRITE_DISABLE	0x1e
+#define DS1682_REG_WRITE_MEM_DISABLE	0x1f
+
+#define DS1682_EEPROM_SIZE		10
+
+/*
+ * Device address
+ */
+static unsigned short normal_i2c[] = { 0x6B, I2C_CLIENT_END };
+I2C_CLIENT_INSMOD;
+
+/*
+ * Low level chip access functions
+ */
+static int ds1682_read(struct i2c_client *client, int reg, void *buf, int len)
+{
+	u8 *bytes = buf;
+	int val;
+
+	dev_dbg(&client->dev, "ds1682_read(reg=%i, buf=%p, len=%i)\n",
+		reg, buf, len);
+
+	while (len--) {
+		val = i2c_smbus_read_byte_data(client, reg++);
+		if (val < 0)
+			return -EIO;
+		*bytes++ = val;
+	}
+	return 0;
+}
+
+/*
+ * Generic counter attributes
+ */
+static ssize_t ds1682_show(struct device *dev, struct device_attribute *attr,
+			   char *buf)
+{
+	struct sensor_device_attribute_2 *sattr = to_sensor_dev_attr_2(attr);
+	__le32 val = 0;
+
+	dev_dbg(dev, "ds1682_show() called on %s\n", attr->attr.name);
+
+	/* Read the register */
+	if (ds1682_read(to_i2c_client(dev), sattr->index, &val, sattr->nr))
+		return -EIO;
+
+	/* Special case: the 32 bit regs are time values with 1/4s
+	 * resolution, scale them up to milliseconds */
+	if (sattr->nr == 4)
+		return sprintf(buf, "%lli\n", ((u64) le32_to_cpu(val)) * 250);
+
+	/* Format the output string and return # of bytes */
+	return sprintf(buf, "%li\n", (long)le32_to_cpu(val));
+}
+
+static ssize_t ds1682_store(struct device *dev, struct device_attribute *attr,
+			    const char *buf, size_t count)
+{
+	struct sensor_device_attribute_2 *sattr = to_sensor_dev_attr_2(attr);
+	struct i2c_client *client = to_i2c_client(dev);
+	char *endp;
+	u64 val;
+	__le32 val_le;
+	int rc;
+
+	dev_dbg(dev, "ds1682_store() called on %s\n", attr->attr.name);
+
+	/* Decode input */
+	val = simple_strtoull(buf, &endp, 0);
+	if (buf == endp) {
+		dev_dbg(dev, "input string not a number\n");
+		return -EIO;
+	}
+
+	/* Special case: the 32 bit regs are time values with 1/4s
+	 * resolution, scale input down to quarter-seconds */
+	if (sattr->nr == 4)
+		do_div(val, 250);
+
+	/* write out the value */
+	val_le = cpu_to_le32(val);
+	rc = i2c_smbus_write_i2c_block_data(client, sattr->index, sattr->nr,
+					    (u8 *) & val);
+	if (rc < 0) {
+		dev_err(dev, "register write failed; reg=0x%x, size=%i\n",
+			sattr->index, sattr->nr);
+		return -EIO;
+	}
+
+	return count;
+}
+
+/*
+ * Simple register attributes
+ */
+SENSOR_DEVICE_ATTR_2(config, S_IRUGO, ds1682_show, NULL, DS1682_REG_CONFIG, 1);
+SENSOR_DEVICE_ATTR_2(elapsed_time, S_IRUGO | S_IWUSR, ds1682_show, ds1682_store,
+		     4, DS1682_REG_ELAPSED);
+SENSOR_DEVICE_ATTR_2(alarm_time, S_IRUGO | S_IWUSR, ds1682_show, ds1682_store,
+		     4, DS1682_REG_ALARM);
+SENSOR_DEVICE_ATTR_2(event_count, S_IRUGO | S_IWUSR, ds1682_show, ds1682_store,
+		     2, DS1682_REG_EVT_CNTR);
+
+static const struct attribute_group ds1682_group = {
+	.attrs = (struct attribute *[]) {
+		&sensor_dev_attr_config.dev_attr.attr,
+		&sensor_dev_attr_elapsed_time.dev_attr.attr,
+		&sensor_dev_attr_alarm_time.dev_attr.attr,
+		&sensor_dev_attr_event_count.dev_attr.attr,
+		NULL,
+	},
+};
+
+/*
+ * User data attribute
+ */
+static ssize_t ds1682_eeprom_read(struct kobject *kobj, char *buf, loff_t off,
+				  size_t count)
+{
+	struct i2c_client *client = kobj_to_i2c_client(kobj);
+
+	dev_dbg(&client->dev, "ds1682_eeprom_read(buf=%p, off=%i, count=%i)\n",
+		buf, (int)off, (int)count);
+
+	if (off > DS1682_EEPROM_SIZE)
+		return 0;
+
+	if (off + count > DS1682_EEPROM_SIZE)
+		count = DS1682_EEPROM_SIZE - off;
+
+	if (ds1682_read(client, DS1682_REG_EEPROM + off, buf, count))
+		return -EIO;
+
+	return count;
+}
+
+static ssize_t ds1682_eeprom_write(struct kobject *kobj, char *buf, loff_t off,
+				   size_t count)
+{
+	struct i2c_client *client = kobj_to_i2c_client(kobj);
+
+	dev_dbg(&client->dev, "ds1682_eeprom_write(buf=%p, off=%i, count=%i)\n",
+		buf, (int)off, (int)count);
+
+	if (off >= DS1682_EEPROM_SIZE)
+		return -ENOSPC;
+
+	if (off + count > DS1682_EEPROM_SIZE)
+		count = DS1682_EEPROM_SIZE - off;
+
+	/* Write out to the device */
+	if (i2c_smbus_write_i2c_block_data(client, DS1682_REG_EEPROM + off,
+					   count, buf) < 0)
+		return -EIO;
+
+	return count;
+}
+
+static struct bin_attribute ds1682_eeprom_attr = {
+	.attr = {
+		.name = "eeprom",
+		.mode = S_IRUGO | S_IWUSR,
+		.owner = THIS_MODULE,
+	},
+	.size = DS1682_EEPROM_SIZE,
+	.read = ds1682_eeprom_read,
+	.write = ds1682_eeprom_write,
+};
+
+/*
+ * Called when a device is found at the ds1682 address
+ */
+static struct i2c_driver ds1682_driver;
+static int ds1682_detect(struct i2c_adapter *adapter, int address, int kind)
+{
+	struct i2c_client *client;
+	int err = 0;
+
+	if (!i2c_check_functionality(adapter, I2C_FUNC_SMBUS_READ_BYTE_DATA |
+				     I2C_FUNC_SMBUS_WRITE_I2C_BLOCK))
+		goto exit;
+
+	if (!(client = kzalloc(sizeof(struct i2c_client), GFP_KERNEL))) {
+		err = -ENOMEM;
+		goto exit;
+	}
+
+	client->addr = address;
+	client->adapter = adapter;
+	client->driver = &ds1682_driver;
+
+	/* Note: Ignoring 'kind' value as the ds1682 uses a fixed address */
+
+	/* We can fill in the remaining client fields */
+	strlcpy(client->name, "ds1682", I2C_NAME_SIZE);
+
+	/* Tell the I2C layer a new client has arrived */
+	if ((err = i2c_attach_client(client)))
+		goto exit_free;
+
+	if (sysfs_create_group(&client->dev.kobj, &ds1682_group))
+		goto exit_group_attr;
+
+	if (sysfs_create_bin_file(&client->dev.kobj, &ds1682_eeprom_attr))
+		goto exit_bin_attr;
+
+	return 0;
+
+      exit_bin_attr:
+	sysfs_remove_group(&client->dev.kobj, &ds1682_group);
+      exit_group_attr:
+	i2c_detach_client(client);
+      exit_free:
+	kfree(client);
+      exit:
+	return err;
+}
+
+static int ds1682_attach_adapter(struct i2c_adapter *adapter)
+{
+	return i2c_probe(adapter, &addr_data, ds1682_detect);
+}
+
+static int ds1682_detach_client(struct i2c_client *client)
+{
+	int err;
+
+	sysfs_remove_bin_file(&client->dev.kobj, &ds1682_eeprom_attr);
+	sysfs_remove_group(&client->dev.kobj, &ds1682_group);
+
+	if ((err = i2c_detach_client(client)))
+		return err;
+
+	kfree(client);
+	return 0;
+}
+
+static struct i2c_driver ds1682_driver = {
+	.driver = {
+		.name = "ds1682",
+	},
+	.attach_adapter = ds1682_attach_adapter,
+	.detach_client = ds1682_detach_client,
+};
+
+static int __init ds1682_init(void)
+{
+	return i2c_add_driver(&ds1682_driver);
+}
+
+static void __exit ds1682_exit(void)
+{
+	i2c_del_driver(&ds1682_driver);
+}
+
+MODULE_AUTHOR("Grant Likely <grant.likely@secretlab.ca>");
+MODULE_DESCRIPTION("DS1682 Elapsed Time Indicator driver");
+MODULE_LICENSE("GPL");
+
+module_init(ds1682_init);
+module_exit(ds1682_exit);


^ permalink raw reply related	[flat|nested] 4+ messages in thread

* Re: [PATCH 1/2] Eliminate references to new_client in i2c documentation
  2007-05-10 21:04 [PATCH 1/2] Eliminate references to new_client in i2c documentation Grant Likely
  2007-05-10 21:06 ` [PATCH 2/2] Add driver for Dallas DS1682 elapsed time recorder Grant Likely
@ 2007-05-11  8:01 ` Jean Delvare
  2007-05-11 13:32   ` Grant Likely
  1 sibling, 1 reply; 4+ messages in thread
From: Jean Delvare @ 2007-05-11  8:01 UTC (permalink / raw)
  To: Grant Likely; +Cc: i2c, linux-kernel

Hi Grant,

On Thu, 10 May 2007 15:04:16 -0600, Grant Likely wrote:
> The use of 'new_client' in i2c device allocation has been refered to as
> a 'disease'.  Replace all occurances of 'struct i2c_client *new_client'
> in documentation examples with 'struct i2c_client *client'.
> 
> Also remove unnecessary zero initializtion of .flags since kzalloc is used.
> 
> Signed-off-by: Grant Likely <grant.likely@secretlab.ca>

Thanks for your contribution, however I made a similar documentation
cleanup 3 weeks ago:
  http://lists.lm-sensors.org/pipermail/i2c/2007-April/001075.html
It's upstream already.

> ---
> 
>  Documentation/i2c/writing-clients |   29 ++++++++++++++---------------
>  1 files changed, 14 insertions(+), 15 deletions(-)
> 
> diff --git a/Documentation/i2c/writing-clients b/Documentation/i2c/writing-clients
> index fbcff96..bd61480 100644
> --- a/Documentation/i2c/writing-clients
> +++ b/Documentation/i2c/writing-clients
> @@ -260,7 +260,7 @@ For now, you can ignore the `flags' parameter. It is there for future use.
>    {
>      int err = 0;
>      int i;
> -    struct i2c_client *new_client;
> +    struct i2c_client *client;
>      struct foo_data *data;
>      const char *client_name = ""; /* For non-`sensors' drivers, put the real
>                                       name here! */
> @@ -323,13 +323,12 @@ For now, you can ignore the `flags' parameter. It is there for future use.
>        goto ERROR0;
>      }
>  
> -    new_client = &data->client;
> -    i2c_set_clientdata(new_client, data);
> +    client = &data->client;
> +    i2c_set_clientdata(client, data);
>  
> -    new_client->addr = address;
> -    new_client->adapter = adapter;
> -    new_client->driver = &foo_driver;
> -    new_client->flags = 0;
> +    client->addr = address;
> +    client->adapter = adapter;
> +    client->driver = &foo_driver;
>  
>      /* Now, we do the remaining detection. If no `force' parameter is used. */
>  
> @@ -337,7 +336,7 @@ For now, you can ignore the `flags' parameter. It is there for future use.
>         parameter was used. */
>      if (kind < 0) {
>        /* The below is of course bogus */
> -      if (foo_read(new_client,FOO_REG_GENERIC) != FOO_GENERIC_VALUE)
> +      if (foo_read(client,FOO_REG_GENERIC) != FOO_GENERIC_VALUE)
>           goto ERROR1;
>      }
>  
> @@ -349,7 +348,7 @@ For now, you can ignore the `flags' parameter. It is there for future use.
>      /* Determine the chip type. Not needed if a `force_CHIPTYPE' parameter
>         was used. */
>      if (kind <= 0) {
> -      i = foo_read(new_client,FOO_REG_CHIPTYPE);
> +      i = foo_read(client,FOO_REG_CHIPTYPE);
>        if (i == FOO_TYPE_1) 
>          kind = chip1; /* As defined in the enum */
>        else if (i == FOO_TYPE_2)
> @@ -377,7 +376,7 @@ For now, you can ignore the `flags' parameter. It is there for future use.
>      /* SENSORS ONLY END */
>  
>      /* Fill in the remaining client fields. */
> -    strcpy(new_client->name,client_name);
> +    strcpy(client->name,client_name);
>  
>      /* SENSORS ONLY BEGIN */
>      data->type = kind;
> @@ -389,14 +388,14 @@ For now, you can ignore the `flags' parameter. It is there for future use.
>      /* Any other initializations in data must be done here too. */
>  
>      /* Tell the i2c layer a new client has arrived */
> -    if ((err = i2c_attach_client(new_client)))
> +    if ((err = i2c_attach_client(client)))
>        goto ERROR3;
>  
>      /* SENSORS ONLY BEGIN */
>      /* Register a new directory entry with module sensors. See below for
>         the `template' structure. */
> -    if ((i = i2c_register_entry(new_client, type_name,
> -                                    foo_dir_table_template,THIS_MODULE)) < 0) {
> +    if ((i = i2c_register_entry(client, type_name,
> +                                foo_dir_table_template,THIS_MODULE)) < 0) {
>        err = i;
>        goto ERROR4;
>      }
> @@ -406,14 +405,14 @@ For now, you can ignore the `flags' parameter. It is there for future use.
>  
>      /* This function can write default values to the client registers, if
>         needed. */
> -    foo_init_client(new_client);
> +    foo_init_client(client);
>      return 0;
>  
>      /* OK, this is not exactly good programming practice, usually. But it is
>         very code-efficient in this case. */
>  
>      ERROR4:
> -      i2c_detach_client(new_client);
> +      i2c_detach_client(client);
>      ERROR3:
>      ERROR2:
>      /* SENSORS ONLY START */


-- 
Jean Delvare

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [PATCH 1/2] Eliminate references to new_client in i2c documentation
  2007-05-11  8:01 ` [PATCH 1/2] Eliminate references to new_client in i2c documentation Jean Delvare
@ 2007-05-11 13:32   ` Grant Likely
  0 siblings, 0 replies; 4+ messages in thread
From: Grant Likely @ 2007-05-11 13:32 UTC (permalink / raw)
  To: Jean Delvare; +Cc: i2c, linux-kernel

On 5/11/07, Jean Delvare <khali@linux-fr.org> wrote:
> Hi Grant,
>
> On Thu, 10 May 2007 15:04:16 -0600, Grant Likely wrote:
> > The use of 'new_client' in i2c device allocation has been refered to as
> > a 'disease'.  Replace all occurances of 'struct i2c_client *new_client'
> > in documentation examples with 'struct i2c_client *client'.
> >
> > Also remove unnecessary zero initializtion of .flags since kzalloc is used.
> >
> > Signed-off-by: Grant Likely <grant.likely@secretlab.ca>
>
> Thanks for your contribution, however I made a similar documentation
> cleanup 3 weeks ago:
>   http://lists.lm-sensors.org/pipermail/i2c/2007-April/001075.html
> It's upstream already.

Heh, apparently I don't keep my edge bleeding enough

g.

-- 
Grant Likely, B.Sc. P.Eng.
Secret Lab Technologies Ltd.
grant.likely@secretlab.ca
(403) 399-0195

^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2007-05-11 13:33 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2007-05-10 21:04 [PATCH 1/2] Eliminate references to new_client in i2c documentation Grant Likely
2007-05-10 21:06 ` [PATCH 2/2] Add driver for Dallas DS1682 elapsed time recorder Grant Likely
2007-05-11  8:01 ` [PATCH 1/2] Eliminate references to new_client in i2c documentation Jean Delvare
2007-05-11 13:32   ` Grant Likely

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox