[patch 2.6.14-git] SPI core, refresh
From: David Brownell
Date: Fri Nov 11 2005 - 02:55:47 EST
I thought I'd send out a refresh of this simple SPI framework,
updated to build on recent kernels. The patch description
inludes a summary of what changed ... not much, though there
is now a Documentation/spi directory with a FAQ-ish writeup.
- Dave
This is the core of a small SPI framework, implementing the model of a
queue of messages which complete asynchronously (with thin synchronous
wrappers on top).
- It's still less than 2KB of ".text" (ARM). If there's got to be a
mid-layer for something so simple, that's the right size budget. :)
- The guts use board-specific SPI device tables to build the driver
model tree. (Hardware probing is rarely an option.)
- This version of Kconfig includes no drivers. At this writing there
are two known master controller drivers (PXA/SSP, OMAP MicroWire)
and three protocol drivers (CS8415a, ADS7846, DataFlash) with LKML
mentions of other drivers in development.
- No userspace API. There are several implementations to compare.
Implement them like any other driver, and bind them with sysfs.
The changes from last version are minor. One issue turned out to be a
driver model bug, fixed in kernels 2.6.14-rc1 and newer.
- Matching Russell's 2.6.14-git5+ changes for device_driver suspend
and resume calls (removing the pointless "level" parameter);
- Adding a Documentation/spi directory with a FAQ-like summary
(with a "how to write these drivers" overview);
- Adding a new spi_busnum_to_master() for folk doing hotplug style
things (mostly for developers, like with parport bitbanging).
(Started from a patch by Mark Underwood.)
- Renaming the controller-specific fields in spi_device, based on
discussions with Stephen Street.
- Preallocating a static cacheline-sized buffer for the rpc-style
write-then-read calls, removing a lurking dma/portability issue.
- Renaming spi_message.dev to spi_message.spi (matching the naming
convention elsewhere, avoiding ugliness like msg->dev.dev);
This is enough to write (or adapt) real drivers. This API isn't
expected to change much, except adding protocol tweaking hooks as
needed to handle new hardware.
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ lubbock/include/linux/spi/spi.h 2005-11-10 23:03:50.000000000 -0800
@@ -0,0 +1,516 @@
+/*
+ * Copyright (C) 2005 David Brownell
+ *
+ * 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., 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+#ifndef __LINUX_SPI_H
+#define __LINUX_SPI_H
+
+/*
+ * INTERFACES between SPI master drivers and infrastructure
+ * (There's no SPI slave support for Linux yet...)
+ *
+ * A "struct device_driver" for an spi_device uses "spi_bus_type" and
+ * needs no special API wrappers (much like platform_bus). These drivers
+ * are bound to devices based on their names (much like platform_bus),
+ * and are available in dev->driver.
+ */
+extern struct bus_type spi_bus_type;
+
+/**
+ * struct spi_device - Master side proxy for an SPI slave device
+ * @dev: Driver model representation of the device.
+ * @master: SPI controller used with the device.
+ * @max_speed_hz: Maximum clock rate to be used with this chip
+ * (on this board); may be changed by the device's driver.
+ * @chip-select: Chipselect, distinguishing chips handled by "master".
+ * @mode: The spi mode defines how data is clocked out and in.
+ * This may be changed by the device's driver.
+ * @bits_per_word: Data transfers involve one or more words; word sizes
+ * like eight or 12 bits are common. In-memory wordsizes are
+ * powers of two bytes (e.g. 20 bit samples use 32 bits).
+ * This may be changed by the device's driver.
+ * @irq: Negative, or the number passed to request_irq() to receive
+ * interrupts from this device.
+ * @controller_state: Controller's runtime state
+ * @controller_data: Static board-specific definitions for controller, such
+ * as FIFO initialization parameters; from board_info.controller_data
+ *
+ * An spi_device is used to interchange data between an SPI slave
+ * (usually a discrete chip) and CPU memory.
+ *
+ * In "dev", the platform_data is used to hold information about this
+ * device that's meaningful to the device's protocol driver, but not
+ * to its controller. One example might be an identifier for a chip
+ * variant with slightly different functionality.
+ */
+struct spi_device {
+ struct device dev;
+ struct spi_master *master;
+ u32 max_speed_hz;
+ u8 chip_select;
+ u8 mode;
+#define SPI_CPHA 0x01 /* clock phase */
+#define SPI_CPOL 0x02 /* clock polarity */
+#define SPI_MODE_0 (0|0)
+#define SPI_MODE_1 (0|SPI_CPHA)
+#define SPI_MODE_2 (SPI_CPOL|0)
+#define SPI_MODE_3 (SPI_CPOL|SPI_CPHA)
+#define SPI_CS_HIGH 0x04 /* chipselect active high? */
+ u8 bits_per_word;
+ int irq;
+ void *controller_state;
+ const void *controller_data;
+
+ // likely need more hooks for more protocol options affecting how
+ // the controller talks to its chips, like:
+ // - bit order (default is wordwise msb-first)
+ // - memory packing (12 bit samples into low bits, others zeroed)
+ // - priority
+ // - chipselect delays
+ // - ...
+};
+
+static inline struct spi_device *to_spi_device(struct device *dev)
+{
+ return container_of(dev, struct spi_device, dev);
+}
+
+/* most drivers won't need to care about device refcounting */
+static inline struct spi_device *spi_dev_get(struct spi_device *spi)
+{
+ return (spi && get_device(&spi->dev)) ? spi : NULL;
+}
+
+static inline void spi_dev_put(struct spi_device *spi)
+{
+ if (spi)
+ put_device(&spi->dev);
+}
+
+/* ctldata is for the bus_master driver's runtime state */
+static inline void *spi_get_ctldata(struct spi_device *spi)
+{
+ return spi->controller_state;
+}
+
+static inline void spi_set_ctldata(struct spi_device *spi, void *state)
+{
+ spi->controller_state = state;
+}
+
+
+struct spi_message;
+
+
+/**
+ * struct spi_master - interface to SPI master controller
+ * @cdev: class interface to this driver
+ * @bus_num: board-specific (and often SOC-specific) identifier for a
+ * given SPI controller.
+ * @num_chipselects: chipselects are used to distinguish individual
+ * SPI slaves, and are numbered from zero to num_chipselects.
+ * each slave has a chipselect signal, but it's common that not
+ * every chipselect is connected to a slave.
+ * @setup: updates the device mode and clocking records used by a
+ * device's SPI controller; protocol code may call this.
+ * @transfer: adds a message to the controller's transfer queue.
+ * @cleanup: frees controller-specific state
+ *
+ * Each SPI master controller can communicate with one or more spi_device
+ * children. These make a small bus, sharing MOSI, MISO and SCK signals
+ * but not chip select signals. Each device may be configured to use a
+ * different clock rate, since those shared signals are ignored unless
+ * the chip is selected.
+ *
+ * The driver for an SPI controller manages access to those devices through
+ * a queue of spi_message transactions, copyin data between CPU memory and
+ * an SPI slave device). For each such message it queues, it calls the
+ * message's completion function when the transaction completes.
+ */
+struct spi_master {
+ struct class_device cdev;
+
+ /* other than zero (== assign one dynamically), bus_num is fully
+ * board-specific. usually that simplifies to being SOC-specific.
+ * example: one SOC has three SPI controllers, numbered 1..3,
+ * and one board's schematics might show it using SPI-2. software
+ * would normally use bus_num=2 for that controller.
+ */
+ u16 bus_num;
+
+ /* chipselects will be integral to many controllers; some others
+ * might use board-specific GPIOs.
+ */
+ u16 num_chipselect;
+
+ /* setup mode and clock, etc (spi driver may call many times) */
+ int (*setup)(struct spi_device *spi);
+
+ /* bidirectional bulk transfers
+ *
+ * + The transfer() method may not sleep; its main role is
+ * just to add the message to the queue.
+ * + For now there's no remove-from-queue operation, or
+ * any other request management
+ * + To a given spi_device, message queueing is pure fifo
+ *
+ * + The master's main job is to process its message queue,
+ * selecting a chip then transferring data
+ * + If there are multiple spi_device children, the i/o queue
+ * arbitration algorithm is unspecified (round robin, fifo,
+ * priority, reservations, preemption, etc)
+ *
+ * + Chipselect stays active during the entire message
+ * (unless modified by spi_transfer.cs_change != 0).
+ * + The message transfers use clock and SPI mode parameters
+ * previously established by setup() for this device
+ */
+ int (*transfer)(struct spi_device *spi,
+ struct spi_message *mesg);
+
+ /* called on release() to free memory provided by spi_master */
+ void (*cleanup)(const struct spi_device *spi);
+};
+
+/* the spi driver core manages memory for the spi_master classdev */
+extern struct spi_master *
+spi_alloc_master(struct device *host, unsigned size);
+
+extern int spi_register_master(struct spi_master *master);
+extern void spi_unregister_master(struct spi_master *master);
+
+extern struct spi_master *spi_busnum_to_master(u16 busnum);
+
+/*---------------------------------------------------------------------------*/
+
+/*
+ * I/O INTERFACE between SPI controller and protocol drivers
+ *
+ * Protocol drivers use a queue of spi_messages, each transferring data
+ * between the controller and memory buffers.
+ *
+ * The spi_messages themselves consist of a series of read+write transfer
+ * segments. Those segments always read the same number of bits as they
+ * write; but one or the other is easily ignored by passing a null buffer
+ * pointer. (This is unlike most types of I/O API, because SPI hardware
+ * is full duplex.)
+ *
+ * NOTE: Allocation of spi_transfer and spi_message memory is entirely
+ * up to the protocol driver, which guarantees the integrity of both (as
+ * well as the data buffers) for as long as the message is queued.
+ */
+
+/**
+ * struct spi_transfer - a read/write buffer pair
+ * @tx_buf: data to be written, or NULL
+ * @rx_buf: data to be read, or NULL
+ * @len: size of rx and tx buffers (in bytes)
+ * @cs_change: affects chipselect after this transfer completes
+ * @delay_usecs: microseconds to delay after this transfer before
+ * (optionally) changing the chipselect status, then starting
+ * the next transfer or completing this spi_message.
+ *
+ * SPI transfers always write the same number of bytes as they read.
+ *
+ * All SPI transfers start with the relevant chipselect active. Drivers
+ * can change behavior of the chipselect after the transfer finishes
+ * (including any mandatory delay). The normal behavior is to leave it
+ * selected, except for the last transfer in a message. Setting cs_change
+ * thus allows either of two things:
+ *
+ * (i) If the transfer isn't the last one in the message, chipselect
+ * may briefly go inactive in the middle of the message. Toggling
+ * chipselect in this way may be needed to let a single spi_message
+ * perform all of group of chip transactions together.
+ *
+ * (ii) When the transfer is the last one in the message, the chip will
+ * stay selected until the next transfer in the queue is started. If that
+ * next transfer uses the same chipselect, this removes some overhead.
+ */
+struct spi_transfer {
+ /* it's ok if tx_buf == rx_buf (right?)
+ * for MicroWire, one buffer must be null
+ * buffers must work with dma_*map_single() calls
+ */
+ const void *tx_buf;
+ void *rx_buf;
+ unsigned len;
+
+ /* REVISIT for now, these are only for the controller driver's
+ * use, for recording dma mappings
+ */
+ dma_addr_t tx_dma;
+ dma_addr_t rx_dma;
+
+ unsigned cs_change:1;
+ u16 delay_usecs;
+};
+
+/**
+ * struct spi_message - one multi-segment SPI transaction
+ * @transfers: the segements of the transaction
+ * @n_transfer: how many segments
+ * @spi: SPI device to which the transaction is queued
+ * @complete: called to report transaction completions
+ * @context: the argument to complete() when it's called
+ * @actual_length: how many bytes were transferd
+ * @status: zero for success, else negative errno
+ * @queue: for use by whichever driver currently owns the message
+ * @state: for use by whichever driver currently owns the message
+ */
+struct spi_message {
+ struct spi_transfer *transfers;
+ unsigned n_transfer;
+
+ struct spi_device *spi;
+
+ /* completion is reported through a callback */
+ void FASTCALL((*complete)(void *context));
+ void *context;
+ unsigned actual_length;
+ int status;
+
+ /* for optional use by whatever driver currently owns the
+ * spi_message ... between calls to spi_async and then later
+ * complete(), that's the spi_master controller driver.
+ */
+ struct list_head queue;
+ void *state;
+};
+
+/**
+ * spi_setup -- setup SPI mode and clock rate
+ * @spi: the device whose settings are being modified
+ *
+ * SPI protocol drivers may need to update the transfer mode if the
+ * device doesn't work with the mode 0 default. They may likewise need
+ * to update clock rates or word sizes from initial values. This function
+ * changes those settings, and must be called from a context that can sleep.
+ */
+static inline int
+spi_setup(struct spi_device *spi)
+{
+ return spi->master->setup(spi);
+}
+
+
+/**
+ * spi_async -- asynchronous SPI transfer
+ * @spi: device with which data will be exchanged
+ * @message: describes the data transfers, including completion callback
+ *
+ * This call may be used in_irq and other contexts which can't sleep,
+ * as well as from task contexts which can sleep.
+ *
+ * The completion callback is invoked in a context which can't sleep.
+ * Before that invocation, the value of message->status is undefined.
+ * When the callback is issued, message->status holds either zero (to
+ * indicate complete success) or a negative error code.
+ */
+static inline int
+spi_async(struct spi_device *spi, struct spi_message *message)
+{
+ message->spi = spi;
+ return spi->master->transfer(spi, message);
+}
+
+/*---------------------------------------------------------------------------*/
+
+/* All these synchronous SPI transfer routines are utilities layered
+ * over the core async transfer primitive. Here, "synchronous" means
+ * they will sleep uninterruptibly until the async transfer completes.
+ */
+
+extern int spi_sync(struct spi_device *spi, struct spi_message *message);
+
+/**
+ * spi_write - SPI synchronous write
+ * @spi: device to which data will be written
+ * @buf: data buffer
+ * @len: data buffer size
+ *
+ * This writes the buffer and returns zero or a negative error code.
+ * Callable only from contexts that can sleep.
+ */
+static inline int
+spi_write(struct spi_device *spi, const u8 *buf, size_t len)
+{
+ struct spi_transfer t = {
+ .tx_buf = buf,
+ .rx_buf = NULL,
+ .len = len,
+ .cs_change = 0,
+ };
+ struct spi_message m = {
+ .transfers = &t,
+ .n_transfer = 1,
+ };
+
+ return spi_sync(spi, &m);
+}
+
+/**
+ * spi_read - SPI synchronous read
+ * @spi: device from which data will be read
+ * @buf: data buffer
+ * @len: data buffer size
+ *
+ * This writes the buffer and returns zero or a negative error code.
+ * Callable only from contexts that can sleep.
+ */
+static inline int
+spi_read(struct spi_device *spi, u8 *buf, size_t len)
+{
+ struct spi_transfer t = {
+ .tx_buf = NULL,
+ .rx_buf = buf,
+ .len = len,
+ .cs_change = 0,
+ };
+ struct spi_message m = {
+ .transfers = &t,
+ .n_transfer = 1,
+ };
+
+ return spi_sync(spi, &m);
+}
+
+extern int spi_write_then_read(struct spi_device *spi,
+ const u8 *txbuf, unsigned n_tx,
+ u8 *rxbuf, unsigned n_rx);
+
+/**
+ * spi_w8r8 - SPI synchronous 8 bit write followed by 8 bit read
+ * @spi: device with which data will be exchanged
+ * @cmd: command to be written before data is read back
+ *
+ * This returns the (unsigned) eight bit number returned by the
+ * device, or else a negative error code. Callable only from
+ * contexts that can sleep.
+ */
+static inline int spi_w8r8(struct spi_device *spi, u8 cmd)
+{
+ int status;
+ u8 result;
+
+ status = spi_write_then_read(spi, &cmd, 1, &result, 1);
+
+ /* return negative errno or unsigned value */
+ return (status < 0) ? status : result;
+}
+
+/**
+ * spi_w8r16 - SPI synchronous 8 bit write followed by 16 bit read
+ * @spi: device with which data will be exchanged
+ * @cmd: command to be written before data is read back
+ *
+ * This returns the (unsigned) sixteen bit number returned by the
+ * device, or else a negative error code. Callable only from
+ * contexts that can sleep.
+ *
+ * The number is returned in wire-order, which is at least sometimes
+ * big-endian.
+ */
+static inline int spi_w8r16(struct spi_device *spi, u8 cmd)
+{
+ int status;
+ u16 result;
+
+ status = spi_write_then_read(spi, &cmd, 1, (u8 *) &result, 2);
+
+ /* return negative errno or unsigned value */
+ return (status < 0) ? status : result;
+}
+
+/*---------------------------------------------------------------------------*/
+
+/*
+ * INTERFACE between board init code and SPI infrastructure.
+ *
+ * No SPI driver ever sees these SPI device table segments, but
+ * it's how the SPI core (or adapters that get hotplugged) grows
+ * the driver model tree.
+ *
+ * As a rule, SPI devices can't be probed. Instead, board init code
+ * provides a table listing the devices which are present, with enough
+ * information to bind and set up the device's driver. There's basic
+ * support for nonstatic configurations too; enough to handle adding
+ * parport adapters, or microcontrollers acting as USB-to-SPI bridges.
+ */
+
+/* board-specific information about each SPI device */
+struct spi_board_info {
+ /* the device name and module name are coupled, like platform_bus;
+ * "modalias" is normally the driver name.
+ *
+ * platform_data goes to spi_device.dev.platform_data,
+ * controller_data goes to spi_device.platform_data,
+ * irq is copied too
+ */
+ char modalias[KOBJ_NAME_LEN];
+ const void *platform_data;
+ const void *controller_data;
+ int irq;
+
+ /* slower signaling on noisy or low voltage boards */
+ u32 max_speed_hz;
+
+
+ /* bus_num is board specific and matches the bus_num of some
+ * spi_master that will probably be registered later.
+ *
+ * chip_select reflects how this chip is wired to that master;
+ * it's less than num_chipselect.
+ */
+ u16 bus_num;
+ u16 chip_select;
+
+ /* ... may need additional spi_device chip config data here.
+ * avoid stuff protocol drivers can set; but include stuff
+ * needed to behave without being bound to a driver:
+ * - chipselect polarity
+ * - quirks like clock rate mattering when not selected
+ */
+};
+
+#ifdef CONFIG_SPI
+extern int
+spi_register_board_info(struct spi_board_info const *info, unsigned n);
+#else
+/* board init code may ignore whether SPI is configured or not */
+static inline int
+spi_register_board_info(struct spi_board_info const *info, unsigned n)
+ { return 0; }
+#endif
+
+
+/* If you're hotplugging an adapter with devices (parport, usb, etc)
+ * use spi_new_device() to describe each device. You can also call
+ * spi_unregister_device() to get start making that device vanish,
+ * but normally that would be handled by spi_unregister_master().
+ */
+extern struct spi_device *
+spi_new_device(struct spi_master *, struct spi_board_info *);
+
+static inline void
+spi_unregister_device(struct spi_device *spi)
+{
+ if (spi)
+ device_unregister(&spi->dev);
+}
+
+#endif /* __LINUX_SPI_H */
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ lubbock/drivers/spi/spi.c 2005-11-10 23:03:50.000000000 -0800
@@ -0,0 +1,560 @@
+/*
+ * spi.c - SPI init/core code
+ *
+ * Copyright (C) 2005 David Brownell
+ *
+ * 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., 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+#include <linux/autoconf.h>
+#include <linux/kernel.h>
+#include <linux/device.h>
+#include <linux/init.h>
+#include <linux/cache.h>
+#include <linux/spi/spi.h>
+
+
+/* SPI bustype and spi_master class are registered during early boot,
+ * usually before board init code provides the SPI device tables, and
+ * are available later when driver init code needs them.
+ *
+ * Drivers for SPI devices are like those for platform bus devices:
+ * (a) few bus-specific API wrappers (== needless bloat here)
+ * (b) matched to devices using device names
+ * (c) should provide "native" suspend and resume methods
+ */
+static void spidev_release(struct device *dev)
+{
+ const struct spi_device *spi = to_spi_device(dev);
+
+ /* spi masters may cleanup for released devices */
+ if (spi->master->cleanup)
+ spi->master->cleanup(spi);
+
+ class_device_put(&spi->master->cdev);
+ kfree(dev);
+}
+
+static ssize_t
+modalias_show(struct device *dev, struct device_attribute *a, char *buf)
+{
+ const char *modalias = strchr(dev->bus_id, '-') + 1;
+
+ return snprintf(buf, BUS_ID_SIZE + 1, "%s\n", modalias);
+}
+
+static struct device_attribute spi_dev_attrs[] = {
+ __ATTR_RO(modalias),
+ __ATTR_NULL,
+};
+
+/* modalias support makes "modprobe $MODALIAS" new-style hotplug work,
+ * and the sysfs version makes coldplug work too.
+ */
+
+static int spi_match_device(struct device *dev, struct device_driver *drv)
+{
+ const char *modalias = strchr(dev->bus_id, '-') + 1;
+
+ return strncmp(modalias, drv->name, BUS_ID_SIZE) == 0;
+}
+
+static int spi_hotplug(struct device *dev, char **envp, int num_envp,
+ char *buffer, int buffer_size)
+{
+ const char *modalias = strchr(dev->bus_id, '-') + 1;
+
+ envp[0] = buffer;
+ snprintf(buffer, buffer_size, "MODALIAS=%s", modalias);
+ envp[1] = NULL;
+ return 0;
+}
+
+#ifdef CONFIG_PM
+
+/* Suspend/resume in "struct device_driver" don't really need that
+ * strange third parameter, so we just make it a constant and expect
+ * SPI drivers to ignore it just like most platform drivers do.
+ *
+ * NOTE: the suspend() method for an spi_master controller driver
+ * should verify that all its child devices are marked as suspended;
+ * suspend requests delivered through sysfs power/state files don't
+ * enforce such constraints.
+ */
+static int spi_suspend(struct device *dev, pm_message_t message)
+{
+ int value;
+
+ if (!dev->driver || !dev->driver->suspend)
+ return 0;
+
+ /* suspend will stop irqs and dma; no more i/o */
+ value = dev->driver->suspend(dev, message);
+ if (value == 0)
+ dev->power.power_state = message;
+ return value;
+}
+
+static int spi_resume(struct device *dev)
+{
+ int value;
+
+ if (!dev->driver || !dev->driver->resume)
+ return 0;
+
+ /* resume may restart the i/o queue */
+ value = dev->driver->resume(dev);
+ if (value == 0)
+ dev->power.power_state = PMSG_ON;
+ return value;
+}
+
+#else
+#define spi_suspend NULL
+#define spi_resume NULL
+#endif
+
+struct bus_type spi_bus_type = {
+ .name = "spi",
+ .dev_attrs = spi_dev_attrs,
+ .match = spi_match_device,
+ .hotplug = spi_hotplug,
+ .suspend = spi_suspend,
+ .resume = spi_resume,
+};
+EXPORT_SYMBOL_GPL(spi_bus_type);
+
+/*-------------------------------------------------------------------------*/
+
+/* SPI devices should normally not be created by SPI device drivers; that
+ * would make them board-specific. Similarly with SPI master drivers.
+ * Device registration normally goes into like arch/.../mach.../board-YYY.c
+ * with other readonly (flashable) information about mainboard devices.
+ */
+
+struct boardinfo {
+ struct list_head list;
+ unsigned n_board_info;
+ struct spi_board_info board_info[0];
+};
+
+static LIST_HEAD(board_list);
+static DECLARE_MUTEX(board_lock);
+
+#define kzalloc(n, flags) kcalloc(1,(n),(flags))
+
+static int __init_or_module
+check_child(struct device *dev, void *data)
+{
+ const struct spi_device *spi = to_spi_device(dev);
+ const struct spi_board_info *chip = data;
+
+ return (spi->chip_select == chip->chip_select);
+}
+
+
+/* On typical mainboards, this is purely internal; and it's not needed
+ * after board init creates the hard-wired devices. Some development
+ * platforms may not be able to use spi_register_board_info though, and
+ * this is exported so that for example a USB or parport based adapter
+ * driver could add devices.
+ */
+struct spi_device *__init_or_module
+spi_new_device(struct spi_master *master, struct spi_board_info *chip)
+{
+ struct spi_device *proxy;
+ struct device *dev = master->cdev.dev;
+ int status;
+
+ /* NOTE: caller did any chip->bus_num checks necessary */
+
+ /* only one child per chipselect, ever */
+ if (device_for_each_child(dev, chip, check_child))
+ return NULL;
+
+ if (!class_device_get(&master->cdev))
+ return NULL;
+
+ proxy = kzalloc(sizeof *proxy, GFP_KERNEL);
+ if (!proxy) {
+ dev_err(dev, "can't alloc dev for cs%d\n",
+ chip->chip_select);
+ goto fail;
+ }
+ proxy->master = master;
+ proxy->chip_select = chip->chip_select;
+ proxy->max_speed_hz = chip->max_speed_hz;
+ proxy->irq = chip->irq;
+
+ snprintf(proxy->dev.bus_id, sizeof proxy->dev.bus_id,
+ "%s.%u-%s", master->cdev.class_id,
+ chip->chip_select, chip->modalias);
+ proxy->dev.parent = dev;
+ proxy->dev.bus = &spi_bus_type;
+ proxy->dev.platform_data = (void *) chip->platform_data;
+ proxy->controller_data = chip->controller_data;
+ proxy->controller_state = NULL;
+ proxy->dev.release = spidev_release;
+
+ /* drivers may modify this default i/o setup */
+ status = master->setup(proxy);
+ if (status < 0) {
+ dev_dbg(dev, "can't %s %s, status %d\n",
+ "setup", proxy->dev.bus_id, status);
+ goto fail;
+ }
+
+ /* FIXME Paranoia argues that we detect callers that misbehave by
+ * defining a second device with the same bus and chipselect numbers,
+ * but different driver name; and fail cleanly. Here's where we'd
+ * want to catch that ... atomically with respect to device
+ * registration.
+ */
+
+ status = device_register(&proxy->dev);
+ if (status < 0) {
+ dev_dbg(dev, "can't %s %s, status %d\n",
+ "add", proxy->dev.bus_id, status);
+fail:
+ class_device_put(&master->cdev);
+ kfree(proxy);
+ return NULL;
+ }
+ dev_dbg(dev, "registered child %s\n", proxy->dev.bus_id);
+ return proxy;
+}
+EXPORT_SYMBOL_GPL(spi_new_device);
+
+/*
+ * Board-specific early init code calls this (probably during arch_initcall)
+ * with segments of the SPI device table. Any device nodes are created later,
+ * after the relevant parent SPI controller (bus_num) is defined. We keep
+ * this table of devices forever, so that reloading a controller driver will
+ * not make Linux forget about these hard-wired devices.
+ *
+ * Other code can also call this, e.g. a particular add-on board might provide
+ * SPI devices through its expansion connector, so code initializing that board
+ * would naturally declare its SPI devices.
+ *
+ * The board info passed can safely be __initdata ... but be careful of
+ * any embedded pointers (platform_data, etc), they're copied as-is.
+ */
+int __init
+spi_register_board_info(struct spi_board_info const *info, unsigned n)
+{
+ struct boardinfo *bi;
+
+ bi = kmalloc (sizeof (*bi) + n * sizeof (*info), GFP_KERNEL);
+ if (!bi)
+ return -ENOMEM;
+ bi->n_board_info = n;
+ memcpy(bi->board_info, info, n * sizeof (*info));
+
+ down(&board_lock);
+ list_add_tail(&bi->list, &board_list);
+ up(&board_lock);
+ return 0;
+}
+EXPORT_SYMBOL_GPL(spi_register_board_info);
+
+/* FIXME someone should add support for a __setup("spi", ...) that
+ * creates board info from kernel command lines
+ */
+
+static void __init_or_module
+scan_boardinfo(struct spi_master *master)
+{
+ struct boardinfo *bi;
+ struct device *dev = master->cdev.dev;
+
+ down(&board_lock);
+ list_for_each_entry(bi, &board_list, list) {
+ struct spi_board_info *chip = bi->board_info;
+ unsigned n;
+
+ for (n = bi->n_board_info; n > 0; n--, chip++) {
+ if (chip->bus_num != master->bus_num)
+ continue;
+ /* some controllers only have one chip, so they
+ * might not use chipselects. otherwise, the
+ * chipselects are numbered 0..max.
+ */
+ if (chip->chip_select >= master->num_chipselect
+ && master->num_chipselect) {
+ dev_dbg(dev, "cs%d > max %d\n",
+ chip->chip_select,
+ master->num_chipselect);
+ continue;
+ }
+ (void) spi_new_device(master, chip);
+ }
+ }
+ up(&board_lock);
+}
+
+/*-------------------------------------------------------------------------*/
+
+static void spi_master_release(struct class_device *cdev)
+{
+ struct spi_master *master;
+
+ master = container_of(cdev, struct spi_master, cdev);
+ put_device(master->cdev.dev);
+ master->cdev.dev = NULL;
+ kfree(master);
+}
+
+static struct class spi_master_class = {
+ .name = "spi_master",
+ .owner = THIS_MODULE,
+ .release = spi_master_release,
+};
+
+
+/**
+ * spi_alloc_master - allocate SPI master controller
+ * @dev: the controller, possibly using the platform_bus
+ * @size: how much driver-private data to preallocate; a pointer to this
+ * memory in the class_data field of the returned class_device
+ *
+ * This call is used only by SPI master controller drivers, which are the
+ * only ones directly touching chip registers. It's how they allocate
+ * an spi_master structure, prior to calling spi_add_master().
+ *
+ * This must be called from context that can sleep. It returns the SPI
+ * master structure on success, else NULL.
+ *
+ * The caller is responsible for assigning the bus number and initializing
+ * the master's methods before calling spi_add_master(), or else (on error)
+ * calling class_device_put() to prevent a memory leak.
+ */
+struct spi_master * __init_or_module
+spi_alloc_master(struct device *dev, unsigned size)
+{
+ struct spi_master *master;
+
+ master = kzalloc(size + sizeof *master, SLAB_KERNEL);
+ if (!master)
+ return NULL;
+
+ master->cdev.class = &spi_master_class;
+ master->cdev.dev = get_device(dev);
+ class_set_devdata(&master->cdev, &master[1]);
+
+ return master;
+}
+EXPORT_SYMBOL_GPL(spi_alloc_master);
+
+/**
+ * spi_register_master - register SPI master controller
+ * @master: initialized master, originally from spi_alloc_master()
+ *
+ * SPI master controllers connect to their drivers using some non-SPI bus,
+ * such as the platform bus. The final stage of probe() in that code
+ * includes calling spi_register_master() to hook up to this SPI bus glue.
+ *
+ * SPI controllers use board specific (often SOC specific) bus numbers,
+ * and board-specific addressing for SPI devices combines those numbers
+ * with chip select numbers. Since SPI does not directly support dynamic
+ * device identification, boards need configuration tables telling which
+ * chip is at which address.
+ *
+ * This must be called from context that can sleep. It returns zero on
+ * success, else a negative error code (dropping the master's refcount).
+ */
+int __init_or_module
+spi_register_master(struct spi_master *master)
+{
+ static atomic_t dyn_bus_id = ATOMIC_INIT(0);
+ struct device *dev = master->cdev.dev;
+ int status = -ENODEV;
+
+ /* convention: dynamically assigned bus IDs count down from the max */
+ if (master->bus_num == 0) {
+ master->bus_num = atomic_dec_return(&dyn_bus_id);
+ dev_dbg(dev, "spi%d, dynamic bus number\n", master->bus_num);
+ }
+
+ /* register the device, then userspace will see it.
+ * registration fails if the bus ID is in use.
+ */
+ snprintf(master->cdev.class_id, sizeof master->cdev.class_id,
+ "spi%u", master->bus_num);
+ status = class_device_register(&master->cdev);
+ if (status < 0) {
+ class_device_put(&master->cdev);
+ goto done;
+ }
+ dev_dbg(dev, "registered master %s\n", master->cdev.class_id);
+
+ /* populate children from any spi device tables */
+ scan_boardinfo(master);
+ status = 0;
+done:
+ return status;
+}
+EXPORT_SYMBOL_GPL(spi_register_master);
+
+
+static int __unregister(struct device *dev, void *unused)
+{
+ /* note: before about 2.6.14-rc1 this would corrupt memory: */
+ device_unregister(dev);
+ return 0;
+}
+
+/**
+ * spi_unregister_master - unregister SPI master controller
+ * @master: the master being unregistered
+ *
+ * This call is used only by SPI master controller drivers, which are the
+ * only ones directly touching chip registers.
+ *
+ * This must be called from context that can sleep.
+ */
+void spi_unregister_master(struct spi_master *master)
+{
+ class_device_unregister(&master->cdev);
+ (void) device_for_each_child(master->cdev.dev, NULL, __unregister);
+}
+EXPORT_SYMBOL_GPL(spi_unregister_master);
+
+/**
+ * spi_busnum_to_master - look up master associated with bus_num
+ * @bus_num: the master's bus number
+ *
+ * This call may be used with devices that are registered after
+ * arch init time. It returns a refcounted pointer to the relevant
+ * spi_master (which the caller must release), or NULL if there is
+ * no such master registered.
+ */
+struct spi_master *spi_busnum_to_master(u16 bus_num)
+{
+ if (bus_num) {
+ char name[8];
+ struct kobject *bus;
+
+ snprintf(name, sizeof name, "spi%u", bus_num);
+ bus = kset_find_obj(&spi_master_class.subsys.kset, name);
+ if (bus)
+ return container_of(bus, struct spi_master, cdev.kobj);
+ }
+ return NULL;
+}
+EXPORT_SYMBOL_GPL(spi_busnum_to_master);
+
+
+/*-------------------------------------------------------------------------*/
+
+/**
+ * spi_sync - blocking/synchronous SPI data transfers
+ * @spi: device with which data will be exchanged
+ * @message: describes the data transfers
+ *
+ * This call may only be used from a context that may sleep. The sleep
+ * is non-interruptible, and has no timeout. Low-overhead controller
+ * drivers may DMA directly into and out of the message buffers.
+ *
+ * Note that the SPI device's chip select is active during the message,
+ * and then is normally disabled between messages. Drivers for some
+ * frequently-used devices may want to minimize costs of selecting a chip,
+ * by leaving it selected in anticipation that the next message will go
+ * to the same chip. (That may increase power usage.)
+ *
+ * The return value is a negative error code if the message could not be
+ * submitted, else zero. When the value is zero, then message->status is
+ * also defined: it's the completion code for the transfer, either zero
+ * or a negative error code from the controller driver.
+ */
+int spi_sync(struct spi_device *spi, struct spi_message *message)
+{
+ DECLARE_COMPLETION(done);
+ int status;
+
+ message->complete = (void (*)(void *)) complete;
+ message->context = &done;
+ status = spi_async(spi, message);
+ if (status == 0)
+ wait_for_completion(&done);
+ message->context = NULL;
+ return status;
+}
+EXPORT_SYMBOL(spi_sync);
+
+/**
+ * spi_write_then_read - SPI synchronous write followed by read
+ * @spi: device with which data will be exchanged
+ * @txbuf: data to be written (need not be dma-safe)
+ * @n_tx: size of txbuf, in bytes
+ * @rxbuf: buffer into which data will be read
+ * @n_rx: size of rxbuf, in bytes (need not be dma-safe)
+ *
+ * This performs a half duplex MicroWire style transaction with the
+ * device, sending txbuf and then reading rxbuf. The return value
+ * is zero for success, else a negative errno status code.
+ *
+ * For large transfers, use spi_sync() and dma-safe buffers.
+ */
+int spi_write_then_read(struct spi_device *spi,
+ const u8 *txbuf, unsigned n_tx,
+ u8 *rxbuf, unsigned n_rx)
+{
+ static u8 buf[SMP_CACHE_BYTES] ____cacheline_aligned;
+ static DECLARE_MUTEX(lock);
+
+ int status;
+ struct spi_message message;
+ struct spi_transfer x[2];
+
+ /* Rather than kmalloc to ensure safe dma, use a static buffer.
+ * we can't avoid copying here, but we can avoid heap costs.
+ */
+ if ((n_tx + n_rx) > sizeof buf)
+ return -EINVAL;
+
+ down(&lock);
+ memset(x, 0, sizeof x);
+
+ memcpy(buf, txbuf, n_tx);
+ x[0].tx_buf = buf;
+ x[0].len = n_tx;
+
+ x[1].rx_buf = buf + n_tx;
+ x[1].len = n_rx;
+
+ /* do the i/o */
+ message.transfers = x;
+ message.n_transfer = ARRAY_SIZE(x);
+ status = spi_sync(spi, &message);
+ if (status == 0) {
+ memcpy(rxbuf, x[1].rx_buf, n_rx);
+ status = message.status;
+ }
+
+ up(&lock);
+ return status;
+}
+EXPORT_SYMBOL(spi_write_then_read);
+
+/*-------------------------------------------------------------------------*/
+
+static int __init spi_init(void)
+{
+ bus_register(&spi_bus_type);
+ class_register(&spi_master_class);
+ return 0;
+}
+postcore_initcall(spi_init);
+
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ lubbock/drivers/spi/Kconfig 2005-11-10 23:03:50.000000000 -0800
@@ -0,0 +1,91 @@
+#
+# SPI driver configuration
+#
+# NOTE: the reason this doesn't show SPI slave support is mostly that
+# nobody's needed a slave side API yet. The master-role API is not
+# fully appropriate there, so it'd need some thought to do well.
+#
+menu "SPI support"
+
+# someday this stuff should be set using arch/CPU/PLATFORM/Kconfig
+config SPI_ARCH_HAS_MASTER
+ boolean
+ default y if ARCH_AT91
+ default y if ARCH_OMAP
+ default y if ARCH_PXA
+ default y if X86 # devel hack only!! (ICH7 can...)
+
+config SPI_ARCH_HAS_SLAVE
+ boolean
+ default y if ARCH_AT91
+ default y if ARCH_OMAP
+ default y if ARCH_PXA
+
+config SPI
+ bool "SPI support"
+ depends on SPI_ARCH_HAS_MASTER || SPI_ARCH_HAS_SLAVE
+ help
+ The "Serial Peripheral Interface" is a low level synchronous
+ protocol. Chips that support SPI can have data transfer rates
+ up to several tens of Mbit/sec. Chips are addressed with a
+ controller and a chipselect. Most SPI slaves don't support
+ dynamic device discovery; some are even write-only or read-only.
+
+ SPI is widely used by microcontollers to talk with sensors,
+ eeprom and flash memory, codecs and various other controller
+ chips, analog to digital (and d-to-a) converters, and more.
+ MMC and SD cards can be accessed using SPI protocol; and for
+ DataFlash cards used in MMC sockets, SPI must always be used.
+
+ SPI is one of a family of similar protocols using a four wire
+ interface (select, clock, data in, data out) including Microwire
+ (half duplex), SSP, SSI, and PSP. This driver framework should
+ work with most such devices and controllers.
+
+config SPI_DEBUG
+ boolean "Debug support for SPI drivers"
+ depends on SPI && DEBUG_KERNEL
+ help
+ Say "yes" to enable debug messaging (like dev_dbg and pr_debug),
+ sysfs, and debugfs support in SPI controller and protocol drivers.
+
+#
+# MASTER side ... talking to discrete SPI slave chips including microcontrollers
+#
+
+config SPI_MASTER
+# boolean "SPI Master Support"
+ boolean
+ default SPI && SPI_ARCH_HAS_MASTER
+ help
+ If your system has an master-capable SPI controller (which
+ provides the clock and chipselect), you can enable that
+ controller and the protocol drivers for the SPI slave chips
+ that are connected.
+
+comment "SPI Master Controller Drivers"
+ depends on SPI_MASTER
+
+
+#
+# Add new SPI master controllers in alphabetical order above this line
+#
+
+
+#
+# There are lots of SPI device types, with sensors and memory
+# being probably the most widely used ones.
+#
+comment "SPI Protocol Masters"
+ depends on SPI_MASTER
+
+
+#
+# Add new SPI protocol masters in alphabetical order above this line
+#
+
+
+# (slave support would go here)
+
+endmenu # "SPI support"
+
--- lubbock.orig/arch/arm/Kconfig 2005-11-10 23:03:44.000000000 -0800
+++ lubbock/arch/arm/Kconfig 2005-11-10 23:03:50.000000000 -0800
@@ -774,6 +774,8 @@ source "drivers/char/Kconfig"
source "drivers/i2c/Kconfig"
+source "drivers/spi/Kconfig"
+
source "drivers/hwmon/Kconfig"
#source "drivers/l3/Kconfig"
--- lubbock.orig/drivers/Kconfig 2005-11-10 21:20:07.000000000 -0800
+++ lubbock/drivers/Kconfig 2005-11-10 23:03:50.000000000 -0800
@@ -44,6 +44,8 @@ source "drivers/char/Kconfig"
source "drivers/i2c/Kconfig"
+source "drivers/spi/Kconfig"
+
source "drivers/w1/Kconfig"
source "drivers/hwmon/Kconfig"
--- lubbock.orig/drivers/Makefile 2005-11-10 21:22:43.000000000 -0800
+++ lubbock/drivers/Makefile 2005-11-10 23:03:50.000000000 -0800
@@ -40,6 +40,7 @@ obj-$(CONFIG_FUSION) += message/
obj-$(CONFIG_IEEE1394) += ieee1394/
obj-y += cdrom/
obj-$(CONFIG_MTD) += mtd/
+obj-$(CONFIG_SPI) += spi/
obj-$(CONFIG_PCCARD) += pcmcia/
obj-$(CONFIG_DIO) += dio/
obj-$(CONFIG_SBUS) += sbus/
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ lubbock/drivers/spi/Makefile 2005-11-10 23:03:50.000000000 -0800
@@ -0,0 +1,23 @@
+#
+# Makefile for kernel SPI drivers.
+#
+
+ifeq ($(CONFIG_SPI_DEBUG),y)
+EXTRA_CFLAGS += -DDEBUG
+endif
+
+# small core, mostly translating board-specific
+# config declarations into driver model code
+obj-$(CONFIG_SPI_MASTER) += spi.o
+
+# SPI master controller drivers (bus)
+# ... add above this line ...
+
+# SPI protocol drivers (device/link on bus)
+# ... add above this line ...
+
+# SPI slave controller drivers (upstream link)
+# ... add above this line ...
+
+# SPI slave drivers (protocol for that link)
+# ... add above this line ...
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ lubbock/Documentation/spi/spi-summary 2005-11-10 23:03:50.000000000 -0800
@@ -0,0 +1,363 @@
+Overview of Linux kernel SPI support
+====================================
+
+07-Nov-2005
+
+What is SPI?
+------------
+The "Serial Peripheral Interface" (SPI) is a three-wire point-to-point
+serial link used to connect microcontrollers to sensors and memory.
+
+SPI masters may use a fourth "chip select" line to activate a given SPI
+slave device, so that those three signal wires may be connected to several
+chips in parallel. All SPI slaves support chipselects. Some devices
+have other signals, often including an interrupt to the master.
+
+The three signal wires hold a clock (often on the order of 10 MHz), and
+parallel data lines with "Master Out, Slave In" (MOSI) or "Master In,
+Slave Out" (MISO) signals. There are four modes through which data is
+exchanged; mode-0 and mode-3 are most commonly used.
+
+Unlike serial busses like USB or SMBUS, even low level protocols for
+SPI slave functions are usually not interoperable between vendors
+(except for cases like SPI memory chips).
+
+ - SPI may be used for request/response style device protocols, as with
+ touchscreen sensors and memory chips.
+
+ - It may also be used to stream data in either direction, or both of
+ them at the same time.
+
+ - Some devices may use eight bit words. Others may different word
+ lengths, such as streams of 12-bit or 20-bit digital samples.
+
+In the same way, SPI slaves will only rarely support any kind of automatic
+discovery/enumeration protocol. The tree of slave devices accessible from
+a given SPI master will normally be set up manually, with configuration
+tables.
+
+SPI is only one of the names used by such three-wire protocols, and
+most controllers have no problem handling "MicroWire" (think of it as
+half-duplex SPI, for request/response protocols), SSP ("Synchronous
+Serial Protocol"), PSP ("Programmable Serial Protocol"), and other
+related protocols.
+
+Microcontrollers often support both master and slave sides of the SPI
+protocol. This document (and Linux) currently only supports the master
+side of SPI interactions.
+
+
+Who uses it? On what kinds of systems?
+---------------------------------------
+Linux developers using SPI are probably writing device drivers for embedded
+systems boards. SPI is used to control external chips, and it is also a
+protocol supported by every MMC or SD memory card. (The older "DataFlash"
+cards, predating MMC cards but using the same connectors and card shape,
+support only SPI.) Some PC hardware uses SPI flash for BIOS code.
+
+SPI slave chips range from digital/analog converters used for analog
+sensors and codecs, to memory, to peripherals like USB controllers
+or Ethernet adapters; and more.
+
+Most systems using SPI will integrate a few devices on a mainboard.
+Some provide SPI links on expansion connectors; in cases where no
+dedicated SPI controller exists, GPIO pins can be used to create a
+low speed "bitbanging" adapter. Very few systems will "hotplug" an SPI
+controller; the reasons to use SPI focus on low cost and simple operation,
+and if dynamic reconfiguration is important, USB will often be a more
+appropriate low-pincount peripheral bus.
+
+Many microcontrollers that can run Linux integrate one or more I/O
+interfaces with SPI modes. Given SPI support, they could use MMC or SD
+cards without needing a special purpose MMC/SD/SDIO controller.
+
+
+How do these driver programming interfaces work?
+------------------------------------------------
+There are two types of SPI driver, here called:
+
+ Controller drivers ... these are often built in to System-On-Chip
+ processors, and often support both Master and Slave roles.
+ These drivers touch hardware registers and may use DMA.
+
+ Protocol drivers ... these pass messages through the controller
+ driver to communicate with a Slave or Master device on the
+ other side of an SPI link.
+
+So for example one protocol driver might talk to the MTD layer to export
+data to filesystems stored on SPI flash like DataFlash; and others might
+control audio interfaces, present touchscreen sensors as input interfaces,
+or monitor temperature and voltage levels during industrial processing.
+And those might all be sharing the same controller driver.
+
+A "struct spi_device" encapsulates the master-side interface between
+those two types of driver. At this writing, Linux has no slave side
+programming interface.
+
+There is a minimal core of SPI programming interfaces, focussing on
+using driver model to connect controller and protocol drivers using
+device tables provided by board specific initialization code.
+
+The basic I/O primitive submits an asynchronous message to an I/O queue
+maintained by the controller driver. A completion callback is issued
+asynchronously when the data transfer(s) in that message completes.
+There are also some simple synchronous wrappers for those calls.
+
+
+How does board-specific init code declare SPI devices?
+------------------------------------------------------
+Linux needs several kinds of information to properly configure SPI devices.
+That information is normally provided by board-specific code, even for
+chips that do support some of automated discovery/enumeration.
+
+DECLARE CONTROLLERS
+
+The first kind of information is a list of what SPI controllers exist.
+For System-on-Chip (SOC) based boards, these will usually be platform
+devices, and the controller may need some platform_data in order to
+operate properly. The "struct platform_device" will include resources
+like the physical address of the controller's first register and its IRQ.
+
+Platforms will often abstract the "register SPI controller" operation,
+maybe coupling it with code to initialize pin configurations, so that
+the arch/.../mach-*/board-*.c files for several boards can all share the
+same basic controller setup code. This is because most SOCs have several
+SPI-capable controllers, and only the ones actually usable on a given
+board should normally be set up and registered.
+
+So for example arch/.../mach-*/board-*.c files might have code like:
+
+ #include <asm/arch/spi.h> /* for mysoc_spi_data */
+
+ /* if your mach-* infrastructure doesn't support kernels that can
+ * run on multiple boards, pdata wouldn't benefit from "__init".
+ */
+ static struct mysoc_spi_data __init pdata = { ... };
+
+ static __init board_init(void)
+ {
+ ...
+ /* this board only uses SPI controller #2 */
+ mysoc_register_spi(2, &pdata);
+ ...
+ }
+
+And SOC-specific utility code might look something like:
+
+ #include <asm/arch/spi.h>
+
+ static struct platform_device spi2 = { ... };
+
+ void mysoc_register_spi(unsigned n, struct mysoc_spi_data *pdata)
+ {
+ struct mysoc_spi_data *pdata2;
+
+ pdata2 = kmalloc(sizeof *pdata2, GFP_KERNEL);
+ *pdata2 = pdata;
+ ...
+ if (n == 2) {
+ spi2->dev.platform_data = pdata2;
+ register_platform_device(&spi2);
+
+ /* also: set up pin modes so the spi2 signals are
+ * visible on the relevant pins ... bootloaders on
+ * production boards may already have done this, but
+ * developer boards will often need Linux to do it.
+ */
+ }
+ ...
+ }
+
+Notice how the platform_data for boards may be different, even if the
+same SOC controller is used. For example, on one board SPI might use
+an external clock, where another derives the SPI clock from current
+settings of some master clock.
+
+
+DECLARE SLAVE DEVICES
+
+The second kind of information is a list of what SPI slave devices exist
+on the target board, often with some board-specific data needed for the
+driver to work correctly.
+
+Normally your arch/.../mach-*/board-*.c files would provide a small table
+listing the SPI devices on each board. (This would typically be only a
+small handful.) That might look like:
+
+ static struct ads7846_platform_data ads_info = {
+ .vref_delay_usecs = 100,
+ .x_plate_ohms = 580,
+ .y_plate_ohms = 410,
+ };
+
+ static struct spi_board_info spi_board_info[] __initdata = {
+ {
+ .modalias = "ads7846",
+ .platform_data = &ads_info,
+ .mode = SPI_MODE_0,
+ .irq = GPIO_IRQ(31),
+ .max_speed_hz = 120000 /* max sample rate at 3V */ * 16,
+ .bus_num = 1,
+ .chip_select = 0,
+ },
+ };
+
+Again, notice how board-specific information is provided; each chip may need
+several types. This example shows generic constraints like the fastest SPI
+clock to allow (a function of board voltage in this case) or how an IRQ pin
+is wired, plus chip-specific constraints like an important delay that's
+changed by the capacitance at one pin.
+
+(There's also "controller_data", information that may be useful to the
+controller driver. An example would be peripheral-specific DMA tuning
+data or special chipselect logic. This is stored in spi_device later.)
+
+The board_info should provide enough information to let the system work
+without the chip's driver being loaded. The most troublesome aspect of
+that is likely the SPI_CS_HIGH bit in the spi_device.mode field, since
+sharing a bus with a device that interprets chipselect "backwards" is
+not possible.
+
+Then your board initialization code would register that table with the SPI
+infrastructure, so that it's available later when the SPI master controller
+driver is registered:
+
+ spi_register_board_info(spi_board_info, ARRAY_SIZE(spi_board_info));
+
+Like with other static board-specific setup, you won't unregister those.
+
+
+NON-STATIC CONFIGURATIONS
+
+Developer boards often play by different rules than product boards, and one
+example is the potential need to hotplug SPI devices and/or controllers.
+
+For those cases you might need to use use spi_busnum_to_master() to look
+up the spi bus master, and will likely need spi_new_device() to provide the
+board info based on the board that was hotplugged. Of course, you'd later
+call at least spi_unregister_device() when that board is removed.
+
+
+How do I write an "SPI Protocol Driver"?
+----------------------------------------
+All SPI drivers are currently kernel drivers. A userspace driver API
+would just be another kernel driver, probably offering some lowlevel
+access through aio_read(), aio_write(), and ioctl() calls and using the
+standard userspace sysfs mechanisms to bind to a given SPI device.
+
+SPI protocol drivers are normal device drivers, with no more wrapper
+than needed by platform devices:
+
+ static struct device_driver CHIP_driver = {
+ .name = "CHIP",
+ .bus = &spi_bus_type,
+ .probe = CHIP_probe,
+ .remove = __exit_p(CHIP_remove),
+ .suspend = CHIP_suspend,
+ .resume = CHIP_resume,
+ };
+
+The SPI core will autmatically attempt to bind this driver to any SPI
+device whose board_info gave a modalias of "CHIP". Your probe() code
+might look like this unless you're creating a class_device:
+
+ static int __init CHIP_probe(struct device *dev)
+ {
+ struct spi_device *spi = to_spi_device(dev);
+ struct CHIP *chip;
+ struct CHIP_platform_data *pdata = dev->platform_data;
+
+ /* get memory for driver's per-chip state */
+ chip = kzalloc(sizeof *chip, GFP_KERNEL);
+ if (!chip)
+ return -ENOMEM;
+ dev_set_drvdata(dev, chip);
+
+ ... etc
+ return 0;
+ }
+
+As soon as it enters probe(), the driver may issue I/O requests to
+the SPI device using "struct spi_message". A message is a series of
+bidirectional I/O transfers, executed in sequence.
+
+ - The basic I/O primitive is spi_async(). Async requests may be
+ issued in any context (irq handler, task, etc) and completion
+ is reported using a callback provided with the message. The
+ buffers follow normal rules for kernel data buffers, which means
+ that controller drivers using DMA aren't forced to make extra
+ copies unless the hardware requires it (e.g. working around
+ hardware errata that force the use of bounce buffering).
+
+ - There are also synchronous wrappers like spi_sync(), and wrappers
+ like spi_read(), spi_write(), and spi_write_then_read(). These
+ may be issued only in contexts that may sleep, and they're all
+ clean (and small, and "optional") layers over spi_async().
+
+ - The spi_write_then_read() call, and convenience wrappers around
+ it, should only be used with small amounts of data where the
+ cost of an extra copy may be ignored. It's designed to support
+ common RPC-style requests, such as writing an eight bit command
+ and reading a sixteen bit response -- spi_w8r16() being one its
+ wrappers, doing exactly that.
+
+Some drivers may need to modify spi_device characteristics like the
+transfer mode, wordsize, or clock rate. This is done with spi_setup(),
+which would normally be called from probe() before the first I/O is
+done to the device.
+
+While "spi_device" would be the bottom boundary of the driver, the
+upper boundaries might include sysfs (especially for sensor readings),
+the input layer, ALSA, networking, MTD, the character device framework,
+or other Linux subsystems.
+
+
+How do I write an "SPI Master Controller Driver"?
+-------------------------------------------------
+An SPI controller will probably be registered on the platform_bus; write
+a driver to bind to the device, whichever bus is involved.
+
+The main task of this type of driver is to provide an "spi_master".
+Use spi_alloc_master() to allocate the master, and class_get_devdata()
+to get the driver-private data allocated for that device.
+
+ struct spi_master *master;
+ struct CONTROLLER *c;
+
+ master = spi_alloc_master(dev, sizeof *c);
+ if (!master)
+ return -ENODEV;
+
+ c = class_get_devdata(&master->cdev);
+
+The driver will initialize the fields of that spi_master, including the
+bus number (maybe the same as the platform device ID) and three methods
+used to interact with the SPI core and SPI protocol drivers. It will
+also initialize its own internal state.
+
+ master->setup(struct spi_device *spi)
+ This sets up the device clock rate, SPI mode, and word sizes.
+ Drivers may change the defaults provided by board_info, and then
+ call spi_setup(spi) to invoke this routine. It may sleep.
+
+ master->transfer(struct spi_device *spi, struct spi_message *message)
+ This must not sleep. Its responsibility is arrange that the
+ transfer happens and its complete() callback is issued; the two
+ will normally happen later, after other transfers complete.
+
+ master->cleanup(struct spi_device *spi)
+ Your controller driver may use spi_device.controller_state to hold
+ state it dynamically associates with that device. If you do that,
+ be sure to provide the cleanup() method to free that state.
+
+The bulk of the driver will be managing the I/O queue fed by transfer().
+
+That queue could be purely conceptual. For example, a driver used only
+for low-frequency sensor acess might be fine using synchronous PIO.
+
+But the queue will probably be very real, using message->queue, PIO,
+often DMA (especially if the root filesystem is in SPI flash), and
+execution contexts like IRQ handlers, tasklets, or workqueues (such
+as keventd).
+
+