diff mbox

[v4,1/6] NFC: add nfc subsystem core

Message ID 1309285246-8495-2-git-send-email-aloisio.almeida@openbossa.org (mailing list archive)
State Not Applicable, archived
Headers show

Commit Message

Aloisio Almeida June 28, 2011, 6:20 p.m. UTC
From: Lauro Ramos Venancio <lauro.venancio@openbossa.org>

The NFC subsystem core is responsible for providing the device driver
interface. It is also responsible for providing an interface to the control
operations and data exchange.

Signed-off-by: Lauro Ramos Venancio <lauro.venancio@openbossa.org>
Signed-off-by: Aloisio Almeida Jr <aloisio.almeida@openbossa.org>
Signed-off-by: Samuel Ortiz <sameo@linux.intel.com>
---
 drivers/Kconfig      |    2 -
 drivers/Makefile     |    1 +
 drivers/nfc/Kconfig  |   16 +--
 drivers/nfc/Makefile |    2 +
 include/net/nfc.h    |  131 +++++++++++++++++++
 net/Kconfig          |    1 +
 net/Makefile         |    1 +
 net/nfc/Kconfig      |   16 +++
 net/nfc/Makefile     |    7 +
 net/nfc/core.c       |  344 ++++++++++++++++++++++++++++++++++++++++++++++++++
 net/nfc/nfc.h        |   75 +++++++++++
 11 files changed, 581 insertions(+), 15 deletions(-)
 create mode 100644 include/net/nfc.h
 create mode 100644 net/nfc/Kconfig
 create mode 100644 net/nfc/Makefile
 create mode 100644 net/nfc/core.c
 create mode 100644 net/nfc/nfc.h

Comments

Joe Perches June 28, 2011, 8:18 p.m. UTC | #1
On Tue, 2011-06-28 at 15:20 -0300, Aloisio Almeida Jr wrote:
> From: Lauro Ramos Venancio <lauro.venancio@openbossa.org>
> The NFC subsystem core is responsible for providing the device driver
> interface. It is also responsible for providing an interface to the control
> operations and data exchange.
[]
> diff --git a/net/nfc/nfc.h b/net/nfc/nfc.h
[]
> +#define NFC_INFO(fmt, arg...) printk(KERN_INFO "NFC: " fmt "\n", ## arg)
> +#define NFC_ERR(fmt, arg...) printk(KERN_ERR "%s: " fmt "\n", __func__, ## arg)
> +#define NFC_DBG(fmt, arg...) pr_debug("%s: " fmt "\n", __func__, ## arg)

I think these #defines and their uses would be
better lower case.

#define nfc_info(etc...)
#define nfc_err(etc...)
#define nfc_dbg(etc...)

And because these don't really take any nfc specific
struct as an argument, may be better removed altogether
and replaced with the pr_<level> equivalents.

I think emitting __func__ rarely adds useful information.

--
To unsubscribe from this list: send the line "unsubscribe linux-wireless" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Aloisio Almeida June 28, 2011, 11:31 p.m. UTC | #2
Hi Joe,

On Tue, Jun 28, 2011 at 5:18 PM, Joe Perches <joe@perches.com> wrote:
> On Tue, 2011-06-28 at 15:20 -0300, Aloisio Almeida Jr wrote:
>> From: Lauro Ramos Venancio <lauro.venancio@openbossa.org>
>> The NFC subsystem core is responsible for providing the device driver
>> interface. It is also responsible for providing an interface to the control
>> operations and data exchange.
> []
>> diff --git a/net/nfc/nfc.h b/net/nfc/nfc.h
> []
>> +#define NFC_INFO(fmt, arg...) printk(KERN_INFO "NFC: " fmt "\n", ## arg)
>> +#define NFC_ERR(fmt, arg...) printk(KERN_ERR "%s: " fmt "\n", __func__, ## arg)
>> +#define NFC_DBG(fmt, arg...) pr_debug("%s: " fmt "\n", __func__, ## arg)
>
> I think these #defines and their uses would be
> better lower case.
>
> #define nfc_info(etc...)
> #define nfc_err(etc...)
> #define nfc_dbg(etc...)
>
> And because these don't really take any nfc specific
> struct as an argument, may be better removed altogether
> and replaced with the pr_<level> equivalents.
>
I was using pr_*() on previous versions. One of the proposed changes
for v4 was to create these macros to avoid calling pr_*() functions
with same parameters every time. The implementation chosen is based on
BT_* macros from bluetooth subsystem.

> I think emitting __func__ rarely adds useful information.
>
I can say that during the implementation phase I considered __func__
useful, mainly for tracing the execution. That's the reason I left it
in the code.

Aloisio
--
To unsubscribe from this list: send the line "unsubscribe linux-wireless" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Joe Perches June 28, 2011, 11:52 p.m. UTC | #3
On Tue, 2011-06-28 at 20:31 -0300, Aloisio Almeida wrote:
> Hi Joe,

Oi Aloisio.

> On Tue, Jun 28, 2011 at 5:18 PM, Joe Perches <joe@perches.com> wrote:
> > On Tue, 2011-06-28 at 15:20 -0300, Aloisio Almeida Jr wrote:
> >> From: Lauro Ramos Venancio <lauro.venancio@openbossa.org>
> >> The NFC subsystem core is responsible for providing the device driver
> >> interface. It is also responsible for providing an interface to the control
> >> operations and data exchange.
> > []
> >> diff --git a/net/nfc/nfc.h b/net/nfc/nfc.h
> > []
> >> +#define NFC_INFO(fmt, arg...) printk(KERN_INFO "NFC: " fmt "\n", ## arg)
> >> +#define NFC_ERR(fmt, arg...) printk(KERN_ERR "%s: " fmt "\n", __func__, ## arg)
> >> +#define NFC_DBG(fmt, arg...) pr_debug("%s: " fmt "\n", __func__, ## arg)
> >
> > I think these #defines and their uses would be
> > better lower case.
> >
> > #define nfc_info(etc...)
> > #define nfc_err(etc...)
> > #define nfc_dbg(etc...)
> >
> > And because these don't really take any nfc specific
> > struct as an argument, may be better removed altogether
> > and replaced with the pr_<level> equivalents.
> I was using pr_*() on previous versions. One of the proposed changes
> for v4 was to create these macros to avoid calling pr_*() functions
> with same parameters every time. The implementation chosen is based on
> BT_* macros from bluetooth subsystem.
> > I think emitting __func__ rarely adds useful information.
> I can say that during the implementation phase I considered __func__
> useful, mainly for tracing the execution. That's the reason I left it
> in the code.

Up to you, but regardless, I think the bluetooth macros
aren't good ones to follow.  Most all of the other
<foo>_<level> output logging forms are lowercase.

cheers, Joe

--
To unsubscribe from this list: send the line "unsubscribe linux-wireless" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Marcel Holtmann June 29, 2011, 1:31 a.m. UTC | #4
Hi Joe,

> > The NFC subsystem core is responsible for providing the device driver
> > interface. It is also responsible for providing an interface to the control
> > operations and data exchange.
> []
> > diff --git a/net/nfc/nfc.h b/net/nfc/nfc.h
> []
> > +#define NFC_INFO(fmt, arg...) printk(KERN_INFO "NFC: " fmt "\n", ## arg)
> > +#define NFC_ERR(fmt, arg...) printk(KERN_ERR "%s: " fmt "\n", __func__, ## arg)
> > +#define NFC_DBG(fmt, arg...) pr_debug("%s: " fmt "\n", __func__, ## arg)
> 
> I think these #defines and their uses would be
> better lower case.
> 
> #define nfc_info(etc...)
> #define nfc_err(etc...)
> #define nfc_dbg(etc...)
> 
> And because these don't really take any nfc specific
> struct as an argument, may be better removed altogether
> and replaced with the pr_<level> equivalents.

it is similar to what we do in the Bluetooth subsystem, but in the end
upper-case vs lower-case is a bit more personal taste.

The pr_<level> ones are nice and I wished we had them all back in the
days, but the NFC ones actually could take the controller as argument
and then us the dev_* versions of these commands.

At this stage of the project it is a bit too early to tell I guess.

> I think emitting __func__ rarely adds useful information.

Depends on how you are using your debug statements. I find it really
helpful since then you can keep the text detail to a minimum.

Regards

Marcel


--
To unsubscribe from this list: send the line "unsubscribe linux-wireless" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Joe Perches June 29, 2011, 1:49 a.m. UTC | #5
On Tue, 2011-06-28 at 18:31 -0700, Marcel Holtmann wrote:
> Hi Joe,

Hello Marcel.

> > > The NFC subsystem core is responsible for providing the device driver
> > > interface. It is also responsible for providing an interface to the control
> > > operations and data exchange.
> > []
> > > diff --git a/net/nfc/nfc.h b/net/nfc/nfc.h
> > []
> > > +#define NFC_INFO(fmt, arg...) printk(KERN_INFO "NFC: " fmt "\n", ## arg)
> > > +#define NFC_ERR(fmt, arg...) printk(KERN_ERR "%s: " fmt "\n", __func__, ## arg)
> > > +#define NFC_DBG(fmt, arg...) pr_debug("%s: " fmt "\n", __func__, ## arg)
> > I think these #defines and their uses would be
> > better lower case.
> > #define nfc_info(etc...)
> > #define nfc_err(etc...)
> > #define nfc_dbg(etc...)
> > And because these don't really take any nfc specific
> > struct as an argument, may be better removed altogether
> > and replaced with the pr_<level> equivalents.
> 
> it is similar to what we do in the Bluetooth subsystem, but in the end
> upper-case vs lower-case is a bit more personal taste.

And a bit of of coding style agreement too.

> The pr_<level> ones are nice and I wished we had them all back in the
> days, but the NFC ones actually could take the controller as argument
> and then us the dev_* versions of these commands.

I do think that if there's a controller struct that's always
or mostly used with nfc_<level>, then it should be added and
passed to the functions arguments, maybe with NULL used where
necessary.

> At this stage of the project it is a bit too early to tell I guess.
> > I think emitting __func__ rarely adds useful information.
> Depends on how you are using your debug statements. I find it really
> helpful since then you can keep the text detail to a minimum.

I don't disagree that while debugging function names
and tracing function entries/exits are useful.

Today, dynamic_debug can add __func__ to the output as
desired so I think that it's not really necessary
to add to any <foo>_dbg callsite.

I don't think they're ever useful on <foo>_err.
If they are, then the message probably isn't
descriptive enough and adding %s/__func__ as
necessary is OK.

cheers, Joe


--
To unsubscribe from this list: send the line "unsubscribe linux-wireless" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Marcel Holtmann June 29, 2011, 6 p.m. UTC | #6
Hi Joe,

> > The pr_<level> ones are nice and I wished we had them all back in the
> > days, but the NFC ones actually could take the controller as argument
> > and then us the dev_* versions of these commands.
> 
> I do think that if there's a controller struct that's always
> or mostly used with nfc_<level>, then it should be added and
> passed to the functions arguments, maybe with NULL used where
> necessary.
> 
> > At this stage of the project it is a bit too early to tell I guess.
> > > I think emitting __func__ rarely adds useful information.
> > Depends on how you are using your debug statements. I find it really
> > helpful since then you can keep the text detail to a minimum.
> 
> I don't disagree that while debugging function names
> and tracing function entries/exits are useful.
> 
> Today, dynamic_debug can add __func__ to the output as
> desired so I think that it's not really necessary
> to add to any <foo>_dbg callsite.

I did not know that. Then we might should go ahead and also cleanup the
Bluetooth subsystem.

We do use an implied "Bluetooth: " prefix when calling BT_INFO, but that
can be easily changed to bt_info() as well since I do not care about
that part. 

The Bluetooth subsystem has a Linux 2.4 legacy history and a lot of
things can be done a lot nicer these days.

Regards

Marcel


--
To unsubscribe from this list: send the line "unsubscribe linux-wireless" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Joe Perches June 29, 2011, 11:46 p.m. UTC | #7
On Wed, 2011-06-29 at 20:23 -0300, Aloisio Almeida wrote:
> >> Today, dynamic_debug can add __func__ to the output as
> >> desired so I think that it's not really necessary
> >> to add to any <foo>_dbg callsite.
> That's true only for pr_debug() function. You cannot add __func__ info
> on dev_dbg() calls dynamically.

I believe that's false.  It's definitely stored.

#define dynamic_dev_dbg(dev, fmt, ...) do {				\
	static struct _ddebug descriptor				\
	__used								\
	__attribute__((section("__verbose"), aligned(8))) =		\
	{ KBUILD_MODNAME, __func__, __FILE__, fmt, __LINE__,		\
		_DPRINTK_FLAGS_DEFAULT };				\
	if (unlikely(descriptor.enabled))				\
		dev_printk(KERN_DEBUG, dev, fmt, ##__VA_ARGS__);	\
	} while (0)

Jason?  True or false?

> So, for net/nfc/* I propose to use directly pr_*() functions.
> For device drivers the following macros would be provided:
> #define nfc_dev_info(dev, fmt, arg...) dev_info((dev), "NFC: " fmt, ## arg)
> #define nfc_dev_err(dev, fmt, arg...) dev_err((dev), "%s: " fmt,
> __func__, ## arg)
> #define nfc_dev_dbg(dev, fmt, arg...) dev_dbg((dev), "%s: " fmt,
> __func__, ## arg)
> What do you think?

I still think __func__ isn't useful ;)
I think you should add NFC to nfc_dev_err too.

cheers, Joe

--
To unsubscribe from this list: send the line "unsubscribe linux-wireless" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Aloisio Almeida June 30, 2011, 3:26 a.m. UTC | #8
Hi Joe,

On Wed, Jun 29, 2011 at 8:46 PM, Joe Perches <joe@perches.com> wrote:
> On Wed, 2011-06-29 at 20:23 -0300, Aloisio Almeida wrote:
>> >> Today, dynamic_debug can add __func__ to the output as
>> >> desired so I think that it's not really necessary
>> >> to add to any <foo>_dbg callsite.
>> That's true only for pr_debug() function. You cannot add __func__ info
>> on dev_dbg() calls dynamically.
>
> I believe that's false.  It's definitely stored.
>
> #define dynamic_dev_dbg(dev, fmt, ...) do {                             \
>        static struct _ddebug descriptor                                \
>        __used                                                          \
>        __attribute__((section("__verbose"), aligned(8))) =             \
>        { KBUILD_MODNAME, __func__, __FILE__, fmt, __LINE__,            \
>                _DPRINTK_FLAGS_DEFAULT };                               \
>        if (unlikely(descriptor.enabled))                               \
>                dev_printk(KERN_DEBUG, dev, fmt, ##__VA_ARGS__);        \
>        } while (0)
>
> Jason?  True or false?

It's stored but not retrieved. If you check lib/dynamic_debug.c you
see that only __dynamic_pr_debug() (called by dynamic_pr_debug() )
adds such information on prints. The dev_printk() does not check
_DPRINTK_FLAGS_INCL_* flags.

> I think you should add NFC to nfc_dev_err too.

That's make sense to me also.

Aloisio
--
To unsubscribe from this list: send the line "unsubscribe linux-wireless" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
diff mbox

Patch

diff --git a/drivers/Kconfig b/drivers/Kconfig
index 61631ed..a56b0b8 100644
--- a/drivers/Kconfig
+++ b/drivers/Kconfig
@@ -92,8 +92,6 @@  source "drivers/memstick/Kconfig"
 
 source "drivers/leds/Kconfig"
 
-source "drivers/nfc/Kconfig"
-
 source "drivers/accessibility/Kconfig"
 
 source "drivers/infiniband/Kconfig"
diff --git a/drivers/Makefile b/drivers/Makefile
index a29527f..843cd31 100644
--- a/drivers/Makefile
+++ b/drivers/Makefile
@@ -120,3 +120,4 @@  obj-y				+= ieee802154/
 obj-y				+= clk/
 
 obj-$(CONFIG_HWSPINLOCK)	+= hwspinlock/
+obj-$(CONFIG_NFC)		+= nfc/
diff --git a/drivers/nfc/Kconfig b/drivers/nfc/Kconfig
index ea15800..7809289 100644
--- a/drivers/nfc/Kconfig
+++ b/drivers/nfc/Kconfig
@@ -2,17 +2,8 @@ 
 # Near Field Communication (NFC) devices
 #
 
-menuconfig NFC_DEVICES
-	bool "Near Field Communication (NFC) devices"
-	default n
-	---help---
-	  You'll have to say Y if your computer contains an NFC device that
-	  you want to use under Linux.
-
-	  You can say N here if you don't have any Near Field Communication
-	  devices connected to your computer.
-
-if NFC_DEVICES
+menu "Near Field Communication (NFC) devices"
+	depends on NFC
 
 config PN544_NFC
 	tristate "PN544 NFC driver"
@@ -26,5 +17,4 @@  config PN544_NFC
 	  To compile this driver as a module, choose m here. The module will
 	  be called pn544.
 
-
-endif # NFC_DEVICES
+endmenu
diff --git a/drivers/nfc/Makefile b/drivers/nfc/Makefile
index a4efb16..25296f0 100644
--- a/drivers/nfc/Makefile
+++ b/drivers/nfc/Makefile
@@ -3,3 +3,5 @@ 
 #
 
 obj-$(CONFIG_PN544_NFC)		+= pn544.o
+
+ccflags-$(CONFIG_NFC_DEBUG) := -DDEBUG
diff --git a/include/net/nfc.h b/include/net/nfc.h
new file mode 100644
index 0000000..11d63dc
--- /dev/null
+++ b/include/net/nfc.h
@@ -0,0 +1,131 @@ 
+/*
+ * Copyright (C) 2011 Instituto Nokia de Tecnologia
+ *
+ * Authors:
+ *    Lauro Ramos Venancio <lauro.venancio@openbossa.org>
+ *    Aloisio Almeida Jr <aloisio.almeida@openbossa.org>
+ *
+ * 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.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the
+ * Free Software Foundation, Inc.,
+ * 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ */
+
+#ifndef __NET_NFC_H
+#define __NET_NFC_H
+
+#include <linux/device.h>
+#include <linux/skbuff.h>
+
+struct nfc_dev;
+
+/**
+ * data_exchange_cb_t - Definition of nfc_data_exchange callback
+ *
+ * @context: nfc_data_exchange cb_context parameter
+ * @skb: response data
+ * @err: If an error has occurred during data exchange, it is the
+ *	error number. Zero means no error.
+ *
+ * When a rx or tx package is lost or corrupted or the target gets out
+ * of the operating field, err is -EIO.
+ */
+typedef void (*data_exchange_cb_t)(void *context, struct sk_buff *skb,
+								int err);
+
+struct nfc_ops {
+	int (*start_poll)(struct nfc_dev *dev, u32 protocols);
+	void (*stop_poll)(struct nfc_dev *dev);
+	int (*activate_target)(struct nfc_dev *dev, u32 target_idx,
+							u32 protocol);
+	void (*deactivate_target)(struct nfc_dev *dev, u32 target_idx);
+	int (*data_exchange)(struct nfc_dev *dev, u32 target_idx,
+				struct sk_buff *skb, data_exchange_cb_t cb,
+							void *cb_context);
+};
+
+struct nfc_dev {
+	unsigned idx;
+	struct device dev;
+	bool polling;
+	u32 supported_protocols;
+
+	struct nfc_ops *ops;
+};
+#define to_nfc_dev(_dev) container_of(_dev, struct nfc_dev, dev)
+
+extern struct class nfc_class;
+
+struct nfc_dev *nfc_allocate_device(struct nfc_ops *ops,
+					u32 supported_protocols);
+
+/**
+ * nfc_free_device - free nfc device
+ *
+ * @dev: The nfc device to free
+ */
+static inline void nfc_free_device(struct nfc_dev *dev)
+{
+	put_device(&dev->dev);
+}
+
+int nfc_register_device(struct nfc_dev *dev);
+
+void nfc_unregister_device(struct nfc_dev *dev);
+
+/**
+ * nfc_set_parent_dev - set the parent device
+ *
+ * @nfc_dev: The nfc device whose parent is being set
+ * @dev: The parent device
+ */
+static inline void nfc_set_parent_dev(struct nfc_dev *nfc_dev,
+					struct device *dev)
+{
+	nfc_dev->dev.parent = dev;
+}
+
+/**
+ * nfc_set_drvdata - set driver specifc data
+ *
+ * @dev: The nfc device
+ * @data: Pointer to driver specifc data
+ */
+static inline void nfc_set_drvdata(struct nfc_dev *dev, void *data)
+{
+	dev_set_drvdata(&dev->dev, data);
+}
+
+/**
+ * nfc_get_drvdata - get driver specifc data
+ *
+ * @dev: The nfc device
+ */
+static inline void *nfc_get_drvdata(struct nfc_dev *dev)
+{
+	return dev_get_drvdata(&dev->dev);
+}
+
+/**
+ * nfc_device_name - get the nfc device name
+ *
+ * @dev: The nfc device whose name to return
+ */
+static inline const char *nfc_device_name(struct nfc_dev *dev)
+{
+	return dev_name(&dev->dev);
+}
+
+struct sk_buff *nfc_alloc_skb(unsigned int size, gfp_t gfp);
+
+#endif /* __NET_NFC_H */
diff --git a/net/Kconfig b/net/Kconfig
index 878151c..a073148 100644
--- a/net/Kconfig
+++ b/net/Kconfig
@@ -322,6 +322,7 @@  source "net/rfkill/Kconfig"
 source "net/9p/Kconfig"
 source "net/caif/Kconfig"
 source "net/ceph/Kconfig"
+source "net/nfc/Kconfig"
 
 
 endif   # if NET
diff --git a/net/Makefile b/net/Makefile
index a51d946..acdde49 100644
--- a/net/Makefile
+++ b/net/Makefile
@@ -68,3 +68,4 @@  obj-$(CONFIG_WIMAX)		+= wimax/
 obj-$(CONFIG_DNS_RESOLVER)	+= dns_resolver/
 obj-$(CONFIG_CEPH_LIB)		+= ceph/
 obj-$(CONFIG_BATMAN_ADV)	+= batman-adv/
+obj-$(CONFIG_NFC)		+= nfc/
diff --git a/net/nfc/Kconfig b/net/nfc/Kconfig
new file mode 100644
index 0000000..33e095b
--- /dev/null
+++ b/net/nfc/Kconfig
@@ -0,0 +1,16 @@ 
+#
+# NFC sybsystem configuration
+#
+
+menuconfig NFC
+	depends on NET && EXPERIMENTAL
+	tristate "NFC subsystem support (EXPERIMENTAL)"
+	default n
+	help
+	  Say Y here if you want to build support for NFC (Near field
+	  communication) devices.
+
+	  To compile this support as a module, choose M here: the module will
+	  be called nfc.
+
+source "drivers/nfc/Kconfig"
diff --git a/net/nfc/Makefile b/net/nfc/Makefile
new file mode 100644
index 0000000..28bee59
--- /dev/null
+++ b/net/nfc/Makefile
@@ -0,0 +1,7 @@ 
+#
+# Makefile for the Linux NFC subsystem.
+#
+
+obj-$(CONFIG_NFC) += nfc.o
+
+nfc-objs := core.o
diff --git a/net/nfc/core.c b/net/nfc/core.c
new file mode 100644
index 0000000..992bbc5
--- /dev/null
+++ b/net/nfc/core.c
@@ -0,0 +1,344 @@ 
+/*
+ * Copyright (C) 2011 Instituto Nokia de Tecnologia
+ *
+ * Authors:
+ *    Lauro Ramos Venancio <lauro.venancio@openbossa.org>
+ *    Aloisio Almeida Jr <aloisio.almeida@openbossa.org>
+ *
+ * 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.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the
+ * Free Software Foundation, Inc.,
+ * 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ */
+
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/slab.h>
+
+#include "nfc.h"
+
+#define VERSION "0.1"
+
+int nfc_devlist_generation;
+DEFINE_MUTEX(nfc_devlist_mutex);
+
+/**
+ * nfc_start_poll - start polling for nfc targets
+ *
+ * @dev: The nfc device that must start polling
+ * @protocols: bitset of nfc protocols that must be used for polling
+ *
+ * The device remains polling for targets until a target is found or
+ * the nfc_stop_poll function is called.
+ */
+int nfc_start_poll(struct nfc_dev *dev, u32 protocols)
+{
+	int rc;
+
+	NFC_DBG("dev_name:%s protocols=0x%x", dev_name(&dev->dev), protocols);
+
+	if (!protocols)
+		return -EINVAL;
+
+	device_lock(&dev->dev);
+
+	if (!device_is_registered(&dev->dev)) {
+		rc = -ENODEV;
+		goto error;
+	}
+
+	if (dev->polling) {
+		rc = -EBUSY;
+		goto error;
+	}
+
+	rc = dev->ops->start_poll(dev, protocols);
+	if (!rc)
+		dev->polling = true;
+
+error:
+	device_unlock(&dev->dev);
+	return rc;
+}
+
+/**
+ * nfc_stop_poll - stop polling for nfc targets
+ *
+ * @dev: The nfc device that must stop polling
+ */
+int nfc_stop_poll(struct nfc_dev *dev)
+{
+	int rc = 0;
+
+	NFC_DBG("dev_name:%s", dev_name(&dev->dev));
+
+	device_lock(&dev->dev);
+
+	if (!device_is_registered(&dev->dev)) {
+		rc = -ENODEV;
+		goto error;
+	}
+
+	if (!dev->polling) {
+		rc = -EINVAL;
+		goto error;
+	}
+
+	dev->ops->stop_poll(dev);
+	dev->polling = false;
+
+error:
+	device_unlock(&dev->dev);
+	return rc;
+}
+
+/**
+ * nfc_activate_target - prepare the target for data exchange
+ *
+ * @dev: The nfc device that found the target
+ * @target_idx: index of the target that must be activated
+ * @protocol: nfc protocol that will be used for data exchange
+ */
+int nfc_activate_target(struct nfc_dev *dev, u32 target_idx, u32 protocol)
+{
+	int rc;
+
+	NFC_DBG("dev_name:%s", dev_name(&dev->dev));
+
+	device_lock(&dev->dev);
+
+	if (!device_is_registered(&dev->dev)) {
+		rc = -ENODEV;
+		goto error;
+	}
+
+	rc = dev->ops->activate_target(dev, target_idx, protocol);
+
+error:
+	device_unlock(&dev->dev);
+	return rc;
+}
+
+/**
+ * nfc_deactivate_target - deactivate a nfc target
+ *
+ * @dev: The nfc device that found the target
+ * @target_idx: index of the target that must be deactivated
+ */
+int nfc_deactivate_target(struct nfc_dev *dev, u32 target_idx)
+{
+	int rc = 0;
+
+	NFC_DBG("dev_name:%s", dev_name(&dev->dev));
+
+	device_lock(&dev->dev);
+
+	if (!device_is_registered(&dev->dev)) {
+		rc = -ENODEV;
+		goto error;
+	}
+
+	dev->ops->deactivate_target(dev, target_idx);
+
+error:
+	device_unlock(&dev->dev);
+	return rc;
+}
+
+/**
+ * nfc_data_exchange - transceive data
+ *
+ * @dev: The nfc device that found the target
+ * @target_idx: index of the target
+ * @skb: data to be sent
+ * @cb: callback called when the response is received
+ * @cb_context: parameter for the callback function
+ *
+ * The user must wait for the callback before calling this function again.
+ */
+int nfc_data_exchange(struct nfc_dev *dev, u32 target_idx,
+					struct sk_buff *skb,
+					data_exchange_cb_t cb,
+					void *cb_context)
+{
+	int rc;
+
+	NFC_DBG("dev_name:%s", dev_name(&dev->dev));
+
+	device_lock(&dev->dev);
+
+	if (!device_is_registered(&dev->dev)) {
+		rc = -ENODEV;
+		kfree_skb(skb);
+		goto error;
+	}
+
+	rc = dev->ops->data_exchange(dev, target_idx, skb, cb, cb_context);
+
+error:
+	device_unlock(&dev->dev);
+	return rc;
+}
+
+/**
+ * nfc_alloc_skb - allocate a skb for data exchange responses
+ *
+ * @size: size to allocate
+ * @gfp: gfp flags
+ */
+struct sk_buff *nfc_alloc_skb(unsigned int size, gfp_t gfp)
+{
+	struct sk_buff *skb;
+	unsigned int total_size;
+
+	total_size = size + 1;
+	skb = alloc_skb(total_size, gfp);
+
+	if (skb)
+		skb_reserve(skb, 1);
+
+	return skb;
+}
+EXPORT_SYMBOL(nfc_alloc_skb);
+
+static void nfc_release(struct device *d)
+{
+	struct nfc_dev *dev = to_nfc_dev(d);
+
+	NFC_DBG("dev_name:%s", dev_name(&dev->dev));
+
+	kfree(dev);
+}
+
+struct class nfc_class = {
+	.name = "nfc",
+	.dev_release = nfc_release,
+};
+EXPORT_SYMBOL(nfc_class);
+
+static int match_idx(struct device *d, void *data)
+{
+	struct nfc_dev *dev = to_nfc_dev(d);
+	unsigned *idx = data;
+
+	return dev->idx == *idx;
+}
+
+struct nfc_dev *nfc_get_device(unsigned idx)
+{
+	struct device *d;
+
+	d = class_find_device(&nfc_class, NULL, &idx, match_idx);
+	if (!d)
+		return NULL;
+
+	return to_nfc_dev(d);
+}
+
+/**
+ * nfc_allocate_device - allocate a new nfc device
+ *
+ * @ops: device operations
+ * @supported_protocols: NFC protocols supported by the device
+ */
+struct nfc_dev *nfc_allocate_device(struct nfc_ops *ops,
+					u32 supported_protocols)
+{
+	static atomic_t dev_no = ATOMIC_INIT(0);
+	struct nfc_dev *dev;
+
+	if (!ops->start_poll || !ops->stop_poll || !ops->activate_target ||
+		!ops->deactivate_target || !ops->data_exchange)
+		return NULL;
+
+	if (!supported_protocols)
+		return NULL;
+
+	dev = kzalloc(sizeof(struct nfc_dev), GFP_KERNEL);
+	if (!dev)
+		return NULL;
+
+	dev->dev.class = &nfc_class;
+	dev->idx = atomic_inc_return(&dev_no) - 1;
+	dev_set_name(&dev->dev, "nfc%d", dev->idx);
+	device_initialize(&dev->dev);
+
+	dev->ops = ops;
+	dev->supported_protocols = supported_protocols;
+
+	return dev;
+}
+EXPORT_SYMBOL(nfc_allocate_device);
+
+/**
+ * nfc_register_device - register a nfc device in the nfc subsystem
+ *
+ * @dev: The nfc device to register
+ */
+int nfc_register_device(struct nfc_dev *dev)
+{
+	int rc;
+
+	NFC_DBG("dev_name:%s", dev_name(&dev->dev));
+
+	mutex_lock(&nfc_devlist_mutex);
+	nfc_devlist_generation++;
+	rc = device_add(&dev->dev);
+	mutex_unlock(&nfc_devlist_mutex);
+
+	return rc;
+}
+EXPORT_SYMBOL(nfc_register_device);
+
+/**
+ * nfc_unregister_device - unregister a nfc device in the nfc subsystem
+ *
+ * @dev: The nfc device to unregister
+ */
+void nfc_unregister_device(struct nfc_dev *dev)
+{
+	NFC_DBG("dev_name:%s", dev_name(&dev->dev));
+
+	mutex_lock(&nfc_devlist_mutex);
+	nfc_devlist_generation++;
+
+	/* lock to avoid unregistering a device while an operation
+	   is in progress */
+	device_lock(&dev->dev);
+	device_del(&dev->dev);
+	device_unlock(&dev->dev);
+
+	mutex_unlock(&nfc_devlist_mutex);
+}
+EXPORT_SYMBOL(nfc_unregister_device);
+
+static int __init nfc_init(void)
+{
+	NFC_INFO("NFC Core ver %s", VERSION);
+
+	return class_register(&nfc_class);
+}
+
+static void __exit nfc_exit(void)
+{
+	class_unregister(&nfc_class);
+}
+
+subsys_initcall(nfc_init);
+module_exit(nfc_exit);
+
+MODULE_AUTHOR("Lauro Ramos Venancio <lauro.venancio@openbossa.org>");
+MODULE_DESCRIPTION("NFC Core ver " VERSION);
+MODULE_VERSION(VERSION);
+MODULE_LICENSE("GPL");
diff --git a/net/nfc/nfc.h b/net/nfc/nfc.h
new file mode 100644
index 0000000..b0ff15c
--- /dev/null
+++ b/net/nfc/nfc.h
@@ -0,0 +1,75 @@ 
+/*
+ * Copyright (C) 2011 Instituto Nokia de Tecnologia
+ *
+ * Authors:
+ *    Lauro Ramos Venancio <lauro.venancio@openbossa.org>
+ *    Aloisio Almeida Jr <aloisio.almeida@openbossa.org>
+ *
+ * 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.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the
+ * Free Software Foundation, Inc.,
+ * 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ */
+
+#ifndef __LOCAL_NFC_H
+#define __LOCAL_NFC_H
+
+#include <net/nfc.h>
+
+#define NFC_INFO(fmt, arg...) printk(KERN_INFO "NFC: " fmt "\n", ## arg)
+#define NFC_ERR(fmt, arg...) printk(KERN_ERR "%s: " fmt "\n", __func__, ## arg)
+#define NFC_DBG(fmt, arg...) pr_debug("%s: " fmt "\n", __func__, ## arg)
+
+extern int nfc_devlist_generation;
+extern struct mutex nfc_devlist_mutex;
+
+struct nfc_dev *nfc_get_device(unsigned idx);
+
+static inline void nfc_put_device(struct nfc_dev *dev)
+{
+	put_device(&dev->dev);
+}
+
+static inline void nfc_device_iter_init(struct class_dev_iter *iter)
+{
+	class_dev_iter_init(iter, &nfc_class, NULL, NULL);
+}
+
+static inline struct nfc_dev *nfc_device_iter_next(struct class_dev_iter *iter)
+{
+	struct device *d = class_dev_iter_next(iter);
+	if (!d)
+		return NULL;
+
+	return to_nfc_dev(d);
+}
+
+static inline void nfc_device_iter_exit(struct class_dev_iter *iter)
+{
+	class_dev_iter_exit(iter);
+}
+
+int nfc_start_poll(struct nfc_dev *dev, u32 protocols);
+
+int nfc_stop_poll(struct nfc_dev *dev);
+
+int nfc_activate_target(struct nfc_dev *dev, u32 target_idx, u32 protocol);
+
+int nfc_deactivate_target(struct nfc_dev *dev, u32 target_idx);
+
+int nfc_data_exchange(struct nfc_dev *dev, u32 target_idx,
+					struct sk_buff *skb,
+					data_exchange_cb_t cb,
+					void *cb_context);
+
+#endif /* __LOCAL_NFC_H */