This is an updated version of SPI framework from me, Dmitry Pervushin.
It seems that now it is good time to consolidate our SPI frameworks to
push it to kernel :)
We've tested our SPI core as well with bus drivers with wireless LAN
driver and achieved good performance with relatively small overhead.
This proves the viability of this framework in real life even in
real-time environment. The size of .text is
still about 2,500 bytes, that is comparable with David Brownell's
framework size.
I think now is the time to start the final convergence process for these
two cores and get the final core
into the mainline kernel. And in order to understand where we need to
converge, I created the main differences
list (see below).
The list of main differences between David Brownell's SPI framework (A)
and my one (B):
- (A) uses more complicated structure of SPI message, that contains one
or more atomic transfers, and (B)
offers the only spi_msg that represents the atomic transfer on SPI bus.
The similar approach can be imple-
mented in (B), and actually is implemented. But my imp[ression is that
such enhancement may be added later..
- (A) uses workqueues to queue and handle SPI messages, and (B)
allocates the kernel thread to the same purpose.
Using workqueues is not very good solution in real-time environment; I
think that allocating and starting the
separate thread will give us more predictable and stable results;
- (A) has some assumptions on buffers that are passed down to spi
functions. If some controller driver (or bus driver
in terms of (B)) tries to perform DMA transfers, it must copy the
passed buffers to some memory allocated
using GFP_DMA flag and map it using dma_map_single. From the other
hand, (B) relies on callbacks provided
by SPI device driver to allocate memory for DMA transfers, but keeps
ability to pass user-allocated buffers down
to SPI functions by specifying flags in SPI message. SPI message being
a fundamental essense looks better to me when
it's as simple as possible. Especially when we don't lose any
flexibility which is exacly our case (buffers that are
allocated as well as message itself/provided by user, DMA-capable
buffers..)
- (A) retrieves SPI message from the queue in sequential order (FIFO),
but (B) provides more flexible way by providing
special callback to retrieve next message from queue. This callback may
implement its own discipline of scheduling
SPI messages. In any way, the default is FIFO.
- (A) uses standartized way to provide CS information, and (B) relies on
functional drivers callbacks, which looks more
flexible to me.
SIgned-off-by: dmitry pervushin <[email protected]>
linux/include/linux/spi.h | 285 +++++++++++++++
linux/Documentation/spi.txt | 382 ++++++++++++++++++++
linux/arch/arm/Kconfig | 2
linux/drivers/Kconfig | 2
linux/drivers/Makefile | 1
linux/drivers/spi/Kconfig | 33 +
linux/drivers/spi/Makefile | 14
linux/drivers/spi/spi-core.c | 645 +++++++++++++++++++++++++++++++++++
linux/drivers/spi/spi-dev.c | 219 +++++++++++
9 files changed, 1583 insertions(+)
Index: linux/arch/arm/Kconfig
===================================================================
--- linux.orig/arch/arm/Kconfig
+++ linux/arch/arm/Kconfig
@@ -834,6 +834,8 @@ source "drivers/ssi/Kconfig"
source "drivers/mmc/Kconfig"
+source "drivers/spi/Kconfig"
+
endmenu
source "ktools/Kconfig"
Index: linux/drivers/Kconfig
===================================================================
--- linux.orig/drivers/Kconfig
+++ linux/drivers/Kconfig
@@ -42,6 +42,8 @@ source "drivers/char/Kconfig"
source "drivers/i2c/Kconfig"
+source "drivers/spi/Kconfig"
+
source "drivers/w1/Kconfig"
source "drivers/misc/Kconfig"
Index: linux/drivers/Makefile
===================================================================
--- linux.orig/drivers/Makefile
+++ linux/drivers/Makefile
@@ -67,3 +67,4 @@ obj-$(CONFIG_DPM) += dpm/
obj-$(CONFIG_MMC) += mmc/
obj-y += firmware/
obj-$(CONFIG_EVENT_BROKER) += evb/
+obj-$(CONFIG_SPI) += spi/
Index: linux/drivers/spi/Kconfig
===================================================================
--- /dev/null
+++ linux/drivers/spi/Kconfig
@@ -0,0 +1,33 @@
+#
+# SPI device configuration
+#
+menu "SPI support"
+
+config SPI
+ default Y
+ tristate "SPI (Serial Peripheral Interface) bus support"
+ default false
+ help
+ Say Y if you need to enable SPI support on your kernel.
+ Say M if you want to create the spi-core loadable module.
+
+config SPI_DEBUG
+ bool "SPI debug output"
+ depends on SPI
+ default false
+ help
+ Say Y there if you'd like to see debug output from SPI drivers
+ If unsure, say N
+
+config SPI_CHARDEV
+ default Y
+ tristate "SPI device interface"
+ depends on SPI
+ help
+ Say Y here to use /dev/spiNN device files. They make it possible to have user-space
+ programs use the SPI bus.
+ This support is also available as a module. If so, the module
+ will be called spi-dev.
+
+endmenu
+
Index: linux/drivers/spi/Makefile
===================================================================
--- /dev/null
+++ linux/drivers/spi/Makefile
@@ -0,0 +1,14 @@
+#
+# Makefile for the kernel spi bus driver.
+#
+
+obj-$(CONFIG_SPI) += spi-core.o
+# bus drivers
+# ...functional drivers
+# ...and the common spi-dev driver
+obj-$(CONFIG_SPI_CHARDEV) += spi-dev.o
+
+ifeq ($(CONFIG_SPI_DEBUG),y)
+EXTRA_CFLAGS += -DDEBUG
+endif
+
Index: linux/drivers/spi/spi-core.c
===================================================================
--- /dev/null
+++ linux/drivers/spi/spi-core.c
@@ -0,0 +1,645 @@
+/*
+ * drivers/spi/spi-core.c
+ *
+ * Copyright (C) 2005 MontaVista Software, Inc <[email protected]>
+ *
+ * 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.
+ *
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/config.h>
+#include <linux/errno.h>
+#include <linux/slab.h>
+#include <linux/device.h>
+#include <linux/proc_fs.h>
+#include <linux/kmod.h>
+#include <linux/init.h>
+#include <linux/wait.h>
+#include <linux/kthread.h>
+#include <linux/spi.h>
+#include <asm/atomic.h>
+
+static int spi_thread(void *context);
+
+/**
+ * spi_bus_match_name - verify that driver matches device on spi bus
+ *
+ * @dev: device that hasn't yet being assigned to any driver
+ * @drv: driver for spi device
+ *
+ * Drivers and devices on SPI bus are matched by name, just like the
+ * platform devices, with exception of SPI_DEV_CHAR. Driver with this name
+ * will be matched against any device
+**/
+static int spi_bus_match_name(struct device *dev, struct device_driver *drv)
+{
+ return !strcmp(drv->name, SPI_DEV_CHAR) ||
+ !strcmp(TO_SPI_DEV(dev)->name, drv->name);
+}
+
+/**
+ * spi_bus_suspend - suspend all devices on the spi bus
+ *
+ * @dev: spi device to be suspended
+ * @state: state of device to be set
+ *
+ * This function set device on SPI bus to state `state', just like platform_bus does
+**/
+static int spi_bus_suspend(struct device * dev, u32 state)
+{
+ int ret = 0;
+
+ if (dev->driver && dev->driver->suspend) {
+ ret = dev->driver->suspend(dev, state, SUSPEND_DISABLE);
+ if (ret == 0)
+ ret = dev->driver->suspend(dev, state, SUSPEND_SAVE_STATE);
+ if (ret == 0)
+ ret = dev->driver->suspend(dev, state, SUSPEND_POWER_DOWN);
+ }
+ return ret;
+}
+
+/**
+ * spi_bus_resume - resume functioning of all devices on spi bus
+ *
+ * @dev: device to resume
+ *
+ * This function resumes device on SPI bus, just like platform_bus does
+**/
+static int spi_bus_resume(struct device * dev)
+{
+ int ret = 0;
+
+ if (dev->driver && dev->driver->resume) {
+ ret = dev->driver->resume(dev, RESUME_POWER_ON);
+ if (ret == 0)
+ ret = dev->driver->resume(dev, RESUME_RESTORE_STATE);
+ if (ret == 0)
+ ret = dev->driver->resume(dev, RESUME_ENABLE);
+ }
+ return ret;
+}
+
+/**
+ * spi_bus - the &bus_type structure for SPI devices and drivers
+ *
+ * @name: the name of subsystem, "spi" here
+ * @match: function that matches devices to their drivers
+ * @suspend: PM callback to suspend device
+ * @resume: PM callback to resume device
+**/
+struct bus_type spi_bus = {
+ .name = "spi",
+ .match = spi_bus_match_name,
+ .suspend = spi_bus_suspend,
+ .resume = spi_bus_resume,
+};
+
+/**
+ * spi_bus_driver_init - init internal bus driver structures
+ *
+ * @bus: registered spi_bus_driver structure
+ * @dev: device that represents spi controller
+ *
+ * Once registered by spi_bus_register, the bus driver needs initialization, that
+ * includes starting thread, initializing internal structures.. The best place where
+ * the spi_bus_driver_init is in the `probe' function, when we already sure that passed
+ * device object is SPI master controller
+**/
+int spi_bus_driver_init(struct spi_bus_driver *bus, struct device *dev)
+{
+ struct spi_bus_data *pd =
+ kmalloc(sizeof(struct spi_bus_data), GFP_KERNEL);
+ int err = 0;
+
+ if (!pd) {
+ err = -ENOMEM;
+ goto init_failed_1;
+ }
+ atomic_set(&pd->exiting, 0);
+ pd->bus = bus;
+ init_MUTEX(&pd->lock);
+ INIT_LIST_HEAD(&pd->msgs);
+ init_waitqueue_head(&pd->queue);
+ pd->id = dev->bus_id;
+ pd->thread = kthread_run(spi_thread, pd, "%s-work", pd->id);
+ if (IS_ERR(pd->thread)) {
+ err = PTR_ERR(pd->thread);
+ goto init_failed_2;
+ }
+ dev->platform_data = pd;
+ return 0;
+
+init_failed_2:
+ kfree(pd);
+init_failed_1:
+ return err;
+}
+
+/**
+ * __spi_bus_free -- unregister all children of the spi bus
+ *
+ * @dev: the spi bus `device' object
+ * @context: not used, will be NULL
+ *
+ * This is internal function that is called when unregistering bus driver. Responsibility
+ * of this function is freeing the resources that were requested by spi_bus_driver_init
+ **/
+static int __spi_bus_free(struct device *dev, void *context)
+{
+ struct spi_bus_data *pd = dev->platform_data;
+
+ if (pd) {
+ atomic_inc(&pd->exiting);
+ kthread_stop(pd->thread);
+ kfree(pd);
+ }
+
+ dev_dbg(dev, "unregistering children\n");
+ /*
+ * NOTE: the loop below might needs redesign. Currently
+ * we delete devices from the head of children list
+ * until the list is empty; that's because the function
+ * device_for_each_child will hold the semaphore needed
+ * for deletion of device
+ */
+ while (!list_empty(&dev->children)) {
+ struct device *child =
+ list_entry(dev->children.next, struct device, node);
+ spi_device_del(TO_SPI_DEV(child));
+ }
+ return 0;
+}
+
+/**
+ * spi_bus_driver_unregister - unregister SPI bus controller from the system
+ *
+ * @bus_driver: driver registered by call to spi_bus_driver_register
+ *
+ * unregisters the SPI bus from the system. Before unregistering, it deletes
+ * each SPI device on the bus using call to __spi_device_free
+**/
+void spi_bus_driver_unregister(struct spi_bus_driver *bus_driver)
+{
+ if (bus_driver) {
+ driver_for_each_dev(&bus_driver->driver, NULL, __spi_bus_free);
+ driver_unregister(&bus_driver->driver);
+ }
+}
+
+/**
+ * spi_device_release - release the spi device structure
+ *
+ * @dev: spi_device to be released
+ *
+ * Pointer to this function will be put to dev->release place
+ * This function gets called as a part of device removing
+**/
+void spi_device_release(struct device *dev)
+{
+ struct spi_device* sdev = TO_SPI_DEV(dev);
+
+ kfree( sdev );
+}
+
+/**
+ * spi_device_add - add the new (discovered) SPI device to the bus. Mostly used by bus drivers
+ *
+ * @parent: the bus device object
+ * @name: name of device (non-null!)
+ * @bus_data: bus data to be assigned to device
+ *
+ * SPI devices usually cannot be discovered by SPI bus driver, so it needs to take the configuration
+ * somewhere from hardcoded structures, and prepare bus_data for its devices
+**/
+struct spi_device* spi_device_add(struct device *parent, char *name, void *bus_data)
+{
+ struct spi_device* dev;
+
+ if (!name)
+ goto dev_add_out;
+
+ dev = kmalloc(sizeof(struct spi_device), GFP_KERNEL);
+ if( !dev )
+ goto dev_add_out;
+
+ memset(&dev->dev, 0, sizeof(dev->dev));
+ dev->dev.parent = parent;
+ dev->dev.bus = &spi_bus;
+ strncpy(dev->name, name, sizeof(dev->name));
+ strncpy(dev->dev.bus_id, name, sizeof(dev->dev.bus_id));
+ dev->dev.release = spi_device_release;
+ dev->dev.platform_data = bus_data;
+
+ if (device_register(&dev->dev)<0) {
+ dev_dbg(parent, "device '%s' cannot be added\n", name);
+ goto dev_add_out_2;
+ }
+ return dev;
+
+dev_add_out_2:
+ kfree(dev);
+dev_add_out:
+ return NULL;
+}
+
+/**
+ * spi_queue - queue the message to be processed asynchronously
+ *
+ * @msg: message to be sent
+ *
+ * This function queues the message to spi bus driver's queue. The bus driver
+ * retrieves the message from queue according to its own rules (see retrieve method)
+ * and sends the message to target device. If message has no callback method, originator
+ * of message would get no chance to know where the message is processed. The better
+ * solution is using spi_transfer function, which will return error code if no callback
+ * is provided, or transfer the message synchronously.
+**/
+int spi_queue(struct spi_msg *msg)
+{
+ struct device *dev = &msg->device->dev;
+ struct spi_bus_data *pd = dev->parent->platform_data;
+
+ down(&pd->lock);
+ list_add_tail(&msg->link, &pd->msgs);
+ dev_dbg(dev->parent, "message has been queued\n");
+ up(&pd->lock);
+ wake_up_interruptible(&pd->queue);
+ return 0;
+}
+
+/**
+ * __spi_transfer_callback - callback to process synchronous messages
+ *
+ * @msg: message that is about to complete
+ * @code: message status
+ *
+ * callback for synchronously processed message. If spi_transfer determines
+ * that there is no callback provided neither by msg->status nor callback
+ * parameter, the __spi_transfer_callback will be used, and spi_transfer
+ * does not return until transfer is finished
+ *
+**/
+static void __spi_transfer_callback(struct spi_msg *msg, int code)
+{
+ if (code & (SPIMSG_OK | SPIMSG_FAILED))
+ complete((struct completion *)msg->context);
+}
+
+/*
+ * spi_transfer - transfer the message either in sync or async way
+ *
+ * @msg: message to process
+ * @callback: user-supplied callback
+ *
+ * If both msg->status and callback are set, the error code of -EINVAL
+ * will be returned
+ */
+int spi_transfer(struct spi_msg *msg, void (*callback) (struct spi_msg *, int))
+{
+ struct completion msg_done;
+ int err = -EINVAL;
+
+ if (callback && !msg->status) {
+ msg->status = callback;
+ callback = NULL;
+ }
+
+ if (!callback) {
+ if (!msg->status) {
+ init_completion(&msg_done);
+ msg->context = &msg_done;
+ msg->status = __spi_transfer_callback;
+ spi_queue(msg);
+ wait_for_completion(&msg_done);
+ err = 0;
+ } else {
+ err = spi_queue(msg);
+ }
+ }
+
+ return err;
+}
+
+/**
+ * spi_thread_awake - function that called to determine if thread needs to process any messages
+ *
+ * @bd: pointer to struct spi_bus_data
+ *
+ * Thread wakes up if there is signal to exit (bd->exiting is set) or there are any messages
+ * in bus' queue.
+ */
+static int spi_thread_awake(struct spi_bus_data *bd)
+{
+ int ret;
+
+ if (atomic_read(&bd->exiting)) {
+ return 1;
+ }
+ down(&bd->lock);
+ ret = !list_empty(&bd->msgs);
+ up(&bd->lock);
+ return ret;
+}
+
+static void spi_async_callback (void *_msg )
+{
+ struct spi_msg *msg = _msg;
+
+ msg->status ( msg, SPIMSG_OK );
+}
+
+/**
+ * spi_bus_fifo_retrieve - simple function to retrieve the first message from the queue
+ *
+ * @this: spi_bus_driver that needs to retrieve next message from queue
+ * @data: pointer to spi_bus_data structure associated with spi_bus_driver
+ *
+ * This is pretty simple `retrieve' function. It retrieves the first message from the queue,
+ * and does not care about target of the message. For simple cases, this function is the best
+ * and the fastest solution to provide as retrieve method of bus driver
+ **/
+static struct spi_msg *spi_bus_fifo_retrieve (struct spi_bus_driver *this, struct spi_bus_data *data)
+{
+ return list_entry(data->msgs.next, struct spi_msg, link)
+}
+
+/**
+ * spi_bus_simple_retrieve -- retrieve message from the queue with taking into account previous target
+ *
+ * @this: spi_bus_driver that needs to retrieve next message from queue
+ * @data: pointer to spi_bus_data structure associated with spi_bus_driver
+ *
+ * this function is more complex than spi_bus_fifo_retrieve; it takes into account the already selected
+ * device on SPI bus, and tries to retrieve the message targeted to the same device.
+ *
+ **/
+static struct spi_msg *spi_bus_simple_retrieve( struct spi_bus_driver *this, struct spi_bus_data *data)
+{
+ int found = 0;
+ struct spi_msg *msg;
+
+ list_for_each_entry(msg, &data->msgs, link) {
+ if (!data->selected_device || msg->device == data->selected_device) {
+ found = 1;
+ break;
+ }
+ }
+ if (!found)
+ /*
+ * all messages for current selected_device
+ * are processed.
+ * let's switch to another device
+ */
+ msg = list_entry(data->msgs.next, struct spi_msg, link);
+
+ return msg;
+}
+
+/**
+ * spi_bus_next_msg - the wrapper for retrieve method for bus driver
+ *
+ * @this: spi_bus_driver that needs to retrieve next message from queue
+ * @data: pointer to spi_bus_data structure associated with spi_bus_driver
+ *
+ * If bus driver provides the `retrieve' method, it is called to retrieve the next message
+ * from queue. Otherwise, the spi_bus_fifo_retrieve is called
+ *
+ **/
+static struct spi_msg *spi_bus_next_msg( struct spi_bus_driver *this, struct spi_bus_data *data)
+{
+ if (!this)
+ return NULL;
+ if (this->retrieve)
+ return this->retrieve (this, data);
+ return spi_bus_fifo_retrieve( this, data );
+}
+
+/**
+ * spi_thread - the thread that calls bus functions to perform actual transfers
+ *
+ * @pd: pointer to struct spi_bus_data with bus-specific data
+ *
+ * This function is started as separate thread to perform actual
+ * transfers on SPI bus
+ **/
+static int spi_thread(void *context)
+{
+ struct spi_bus_data *bd = context;
+ struct spi_msg *msg;
+ int xfer_status;
+ struct workqueue_struct *wq;
+
+ wq = create_workqueue ( bd->id );
+ if (!wq)
+ pr_debug( "%s: cannot create workqueue, async callbacks will be unavailable\n", bd->id );
+
+ while (!kthread_should_stop()) {
+
+ wait_event_interruptible(bd->queue, spi_thread_awake(bd));
+
+ if (atomic_read(&bd->exiting))
+ goto thr_exit;
+
+ down(&bd->lock);
+ while (!list_empty(&bd->msgs)) {
+ /*
+ * this part is locked by bus_data->lock,
+ * to protect spi_msg extraction
+ */
+ msg = spi_bus_next_msg( bd->bus, bd );
+
+ /* verify if device needs re-selecting */
+ if (bd->selected_device != msg->device) {
+ if (bd->selected_device && bd->bus->deselect)
+ bd->bus->deselect (bd->selected_device);
+ bd->selected_device = msg->device;
+ if (bd->bus->select)
+ bd->bus->select (bd->selected_device);
+ }
+ list_del(&msg->link);
+ up(&bd->lock);
+
+ /*
+ * and this part is locked by device's lock;
+ * spi_queue will be able to queue new
+ * messages
+ *
+ * note that bd->selected_device is locked, not msg->device
+ * they are the same, but msg can be freed in msg->status function
+ */
+ spi_device_lock(&bd->selected_device);
+ if (bd->bus->set_clock && msg->clock)
+ bd->bus->set_clock(msg->device->dev.parent,
+ msg->clock);
+ xfer_status = bd->bus->xfer(msg);
+ if (msg->status) {
+ if (msg->flags & SPI_M_ASYNC_CB ) {
+ INIT_WORK( &msg->wq_item, spi_async_callback, msg );
+ queue_work (wq, &msg->wq_item );
+ } else {
+ msg->status(msg,
+ xfer_status == 0 ? SPIMSG_OK :
+ SPIMSG_FAILED);
+ }
+ }
+
+ spi_device_unlock(&bd_selected->device);
+
+ /* lock the bus_data again... */
+ down(&bd->lock);
+ }
+ if (bd->bus->deselect)
+ bd->bus->deselect(bd->selected_device);
+ bd->selected_device = NULL;
+ /* device has been just deselected, unlocking the bus */
+ up(&bd->lock);
+ }
+
+thr_exit:
+ if (wq)
+ destroy_workqueue (wq);
+ return 0;
+}
+
+/**
+ * spi_write - send data to a device on an SPI bus
+ *
+ * @dev: the target device
+ * @buf: buffer to be sent
+ * @len: buffer's length
+ *
+ * Returns the number of bytes transferred, or negative error code.
+**/
+int spi_write(struct spi_device *dev, const char *buf, int len)
+{
+ struct spi_msg *msg = spimsg_alloc(dev, SPI_M_WR, len, NULL);
+ int ret;
+
+ memcpy(spimsg_buffer_wr(msg), buf, len);
+ ret = spi_transfer(msg, NULL);
+ return ret == 1 ? len : ret;
+}
+
+/**
+ * spi_read - receive data from a device on an SPI bus
+ *
+ * @dev: the target device
+ * @buf: buffer to be sent
+ * @len: buffer's length
+ *
+ * Returns the number of bytes transferred, or negative error code.
+**/
+int spi_read(struct spi_device *dev, char *buf, int len)
+{
+ int ret;
+ struct spimsg *msg = spimsg_alloc(dev, SPI_M_RD, len, NULL);
+
+ ret = spi_transfer(msg, NULL);
+ memcpy(buf, spimsg_buffer_rd(msg), len);
+ return ret == 1 ? len : ret;
+}
+
+/**
+ * spi_bus_populate/spi_bus_populate2 - populate the bus
+ *
+ * @parent: the SPI bus device object
+ * @devices: string that represents bus population
+ * @devices_s: array of structures that represents bus population
+ * @callback: optional pointer to function that called on each device's add
+ *
+ * These two functions intended to populate the SPI bus corresponding to
+ * device passed as 1st parameter. The difference is in the way to describe
+ * new SPI slave devices: the spi_bus_populate takes the ASCII string delimited
+ * by '\0', where each section matches one SPI device name _and_ its parameters,
+ * and the spi_bus_populate2 takes the array of structures spi_device_desc.
+ *
+ * If some device cannot be added because of spi_device_add fail, the function will
+ * not try to parse the rest of list
+ */
+int spi_bus_populate(struct device *parent,
+ char *devices,
+ void (*callback) (struct device * bus,
+ struct spi_device * new_dev))
+{
+ struct spi_device *new_device;
+ int count = 0;
+
+ while (devices[0]) {
+ dev_dbg(parent, " discovered new SPI device, name '%s'\n",
+ devices);
+ if ((new_device = spi_device_add(parent, devices, NULL)) == NULL)
+ break;
+ if (callback)
+ callback(parent, new_device);
+ devices += (strlen(devices) + 1);
+ count++;
+ }
+ return count;
+}
+
+int spi_bus_populate2(struct device *parent,
+ struct spi_device_desc* devices_s,
+ void (*callback) (struct device* bus,
+ struct spi_device *new_dev,
+ void* params))
+{
+ struct spi_device *new_device;
+ int count = 0;
+
+ while (devices_s->name) {
+ dev_dbg(parent, " discovered new SPI device, name '%s'\n",
+ devices->name );
+ if ((new_device = spi_device_add(parent, devices_s->name, devices_s->params)) == NULL)
+ break;
+ if (callback)
+ callback(parent, new_device, devices_s->params);
+ devices++;
+ count++;
+ }
+ return count;
+}
+
+/**
+ * spi_bus_reset - reset the spi bus
+ *
+ * @bus: device object that represents the SPI bus
+ * @context: u32 value to be passed to reset method of bus
+ *
+ * This is simple wrapper for bus' `reset' method
+ *
+**/
+void spi_bus_reset (struct device* bus, u32 context)
+{
+ if (bus && bus->driver && TO_SPI_BUS_DRIVER(bus->driver)->reset)
+ TO_SPI_BUS_DRIVER(bus->driver)->reset( bus, context );
+}
+
+int __init spi_core_init(void)
+{
+ return bus_register(&spi_bus);
+}
+
+subsys_initcall(spi_core_init);
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("dmitry pervushin <[email protected]>");
+
+EXPORT_SYMBOL_GPL(spi_bus_reset);
+EXPORT_SYMBOL_GPL(spi_queue);
+EXPORT_SYMBOL_GPL(spi_device_add);
+EXPORT_SYMBOL_GPL(spi_bus_driver_unregister);
+EXPORT_SYMBOL_GPL(spi_bus_populate);
+EXPORT_SYMBOL_GPL(spi_bus_populate2);
+EXPORT_SYMBOL_GPL(spi_transfer);
+EXPORT_SYMBOL_GPL(spi_write);
+EXPORT_SYMBOL_GPL(spi_read);
+EXPORT_SYMBOL_GPL(spi_bus);
+EXPORT_SYMBOL_GPL(spi_bus_driver_init);
+EXPORT_SYMBOL_GPL(spi_bus_fifo_retrieve);
+EXPORT_SYMBOL_GPL(spi_bus_simple_retrieve);
+
Index: linux/drivers/spi/spi-dev.c
===================================================================
--- /dev/null
+++ linux/drivers/spi/spi-dev.c
@@ -0,0 +1,219 @@
+/*
+ spi-dev.c - spi driver, char device interface
+
+ Copyright (C) 2005 MontaVista Software, Inc <[email protected]>
+
+ 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/init.h>
+#include <linux/config.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/device.h>
+#include <linux/fs.h>
+#include <linux/slab.h>
+#include <linux/version.h>
+#include <linux/smp_lock.h>
+
+#include <linux/init.h>
+#include <asm/uaccess.h>
+#include <linux/spi.h>
+
+#define SPI_TRANSFER_MAX 65535L
+
+struct spidev_driver_data {
+ int minor;
+};
+
+static ssize_t spidev_read(struct file *file, char *buf, size_t count,
+ loff_t * offset);
+static ssize_t spidev_write(struct file *file, const char *buf, size_t count,
+ loff_t * offset);
+
+static int spidev_open(struct inode *inode, struct file *file);
+static int spidev_release(struct inode *inode, struct file *file);
+static int __init spidev_init(void);
+
+static void spidev_cleanup(void);
+
+static int spidev_probe(struct device *dev);
+static int spidev_remove(struct device *dev);
+
+static struct file_operations spidev_fops = {
+ .owner = THIS_MODULE,
+ .llseek = no_llseek,
+ .read = spidev_read,
+ .write = spidev_write,
+ .open = spidev_open,
+ .release = spidev_release,
+};
+
+static struct class_simple *spidev_class;
+
+static struct spi_driver spidev_driver = {
+ .driver = {
+ .name = SPI_DEV_CHAR,
+ .probe = spidev_probe,
+ .remove = spidev_remove,
+ },
+};
+
+static int spidev_minor;
+
+static int spidev_probe(struct device *dev)
+{
+ struct spidev_driver_data *drvdata;
+
+ drvdata = kmalloc(sizeof(struct spidev_driver_data), GFP_KERNEL);
+ if (!drvdata) {
+ dev_dbg(dev, "allocating drvdata failed\n");
+ return -ENOMEM;
+ }
+
+ drvdata->minor = spidev_minor++;
+ dev_dbg(dev, "setting device's(%p) minor to %d\n", dev, drvdata->minor);
+ dev_set_drvdata(dev, drvdata);
+
+ class_simple_device_add(spidev_class,
+ MKDEV(SPI_MAJOR, drvdata->minor),
+ NULL, "spi%d", drvdata->minor);
+ dev_dbg(dev, " added\n");
+ return 0;
+}
+
+static int spidev_remove(struct device *dev)
+{
+ struct spidev_driver_data *drvdata;
+
+ drvdata = (struct spidev_driver_data *)dev_get_drvdata(dev);
+ class_simple_device_remove(MKDEV(SPI_MAJOR, drvdata->minor));
+ kfree(drvdata);
+ dev_dbg(dev, " removed\n");
+ return 0;
+}
+
+static ssize_t spidev_read(struct file *file, char *buf, size_t count,
+ loff_t * offset)
+{
+ struct spi_device *dev = (struct spi_device *)file->private_data;
+ if (count > SPI_TRANSFER_MAX)
+ count = SPI_TRANSFER_MAX;
+ return spi_read(dev, buf, count);
+}
+
+static ssize_t spidev_write(struct file *file, const char *buf, size_t count,
+ loff_t * offset)
+{
+ struct spi_device *dev = (struct spi_device *)file->private_data;
+ if (count > SPI_TRANSFER_MAX)
+ count = SPI_TRANSFER_MAX;
+ return spi_write(dev, buf, count);
+}
+
+struct spidev_openclose {
+ unsigned int minor;
+ struct file *file;
+};
+
+static int spidev_do_open(struct device *the_dev, void *context)
+{
+ struct spidev_openclose *o = (struct spidev_openclose *)context;
+ struct spi_device *dev = TO_SPI_DEV(the_dev);
+ struct spidev_driver_data *drvdata;
+
+ drvdata = (struct spidev_driver_data *)dev_get_drvdata(the_dev);
+ if (!drvdata) {
+ pr_debug("%s: oops, drvdata is NULL !\n", __FUNCTION__);
+ goto do_open_fail;
+ }
+
+ pr_debug("drvdata->minor = %d vs %d\n", drvdata->minor, o->minor);
+ if (drvdata->minor == o->minor) {
+ get_device(&dev->dev);
+ o->file->private_data = dev;
+ return 1;
+ }
+
+do_open_fail:
+ return 0;
+}
+
+int spidev_open(struct inode *inode, struct file *file)
+{
+ struct spidev_openclose o;
+ int status;
+
+ o.minor = iminor(inode);
+ o.file = file;
+ status = driver_for_each_dev(&spidev_driver.driver, &o, spidev_do_open);
+ if (status == 0) {
+ status = -ENODEV;
+ }
+ return status < 0 ? status : 0;
+}
+
+static int spidev_release(struct inode *inode, struct file *file)
+{
+ struct spi_device *dev = file->private_data;
+
+ if (dev)
+ put_device(&dev->dev);
+ file->private_data = NULL;
+
+ return 0;
+}
+
+static int __init spidev_init(void)
+{
+ int res;
+
+ if ((res = register_chrdev(SPI_MAJOR, "spi", &spidev_fops)) != 0) {
+ goto out;
+ }
+
+ spidev_class = class_simple_create(THIS_MODULE, "spi");
+ if (IS_ERR(spidev_class)) {
+ printk(KERN_ERR "%s: error creating class\n", __FUNCTION__);
+ res = -EINVAL;
+ goto out_unreg;
+ }
+
+ if ((res = spi_driver_add(&spidev_driver)) != 0)
+ goto out_unreg;
+
+ printk("SPI /dev entries driver.\n");
+ return 0;
+
+out_unreg:
+ unregister_chrdev(SPI_MAJOR, "spi");
+out:
+ printk(KERN_ERR "%s: Driver initialization failed\n", __FILE__);
+ return res;
+}
+
+static void spidev_cleanup(void)
+{
+ spi_driver_del(&spidev_driver);
+ class_simple_destroy(spidev_class);
+ unregister_chrdev(SPI_MAJOR, "spi");
+}
+
+MODULE_AUTHOR("dmitry pervushin <[email protected]>");
+MODULE_DESCRIPTION("SPI /dev entries driver");
+MODULE_LICENSE("GPL");
+
+module_init(spidev_init);
+module_exit(spidev_cleanup);
Index: linux/Documentation/spi.txt
===================================================================
--- /dev/null
+++ linux/Documentation/spi.txt
@@ -0,0 +1,382 @@
+Documentation/spi.txt
+========================================================
+Table of contents
+1. Introduction -- what is SPI ?
+2. Purposes of this code
+3. SPI devices stack
+3.1 SPI outline
+3.2 How the SPI devices gets discovered and probed ?
+3.3 DMA and SPI messages
+4. SPI functions and structures reference
+5. How to contact authors
+========================================================
+
+1. What is SPI ?
+----------------
+SPI stands for "Serial Peripheral Interface", a full-duplex synchronous
+serial interface for connecting low-/medium-bandwidth external devices
+using four wires. SPI devices communicate using a master/slave relation-
+ship over two data lines and two control lines:
+- Master Out Slave In (MOSI): supplies the output data from the master
+ to the inputs of the slaves;
+- Master In Slave Out (MISO): supplies the output data from a slave to
+ the input of the master. It is important to note that there can be no
+ more than one slave that is transmitting data during any particular
+ transfer;
+- Serial Clock (SCLK): a control line driven by the master, regulating
+ the flow of data bits;
+- Slave Select (SS): a control line that allows slaves to be turned on
+ and off with hardware control.
+More information is also available at http://en.wikipedia.org/wiki/Serial_Peripheral_Interface
+
+2. Purposes of this code
+------------------------
+The supplied patch is starting point for implementing drivers for various
+SPI busses as well as devices connected to these busses. Currently, the
+SPI core supports only for MASTER mode for system running Linux.
+
+3. SPI devices stack
+--------------------
+
+3.1 The SPI outline
+
+The SPI infrastructure deals with several levels of abstraction. They are
+"SPI bus", "SPI bus driver", "SPI slave device" and "SPI device driver". The
+"SPI bus" is hardware device, which usually called "SPI adapter", and has
+"SPI slave devices" connected. From the Linux' point of view, the "SPI bus" is
+structure of type platform_device, and "SPI slave device" is structure of type
+spi_device. The "SPI bus driver" is the driver which controls the whole
+SPI bus (and, particularly, creates and destroys "SPI slave devices" on the bus),
+and "SPI device driver" is driver that controls the only device on the SPI
+bus, controlled by "SPI bus driver". "SPI device driver" can indirectly
+call "SPI bus driver" to send/receive messages using API provided by SPI
+core, and provide its own interface to the kernel and/or userland.
+So, the device stack looks as follows:
+
+ +--------------+ +---------+
+ | some_bus | | spi_bus |
+ +--------------+ +---------+
+ |..| |
+ |..|--------+ +---------------+
+ +------------+| is parent to | SPI devices |
+ | SPI busses |+-------------> | |
+ +------------+ +---------------+
+ | |
+ +----------------+ +----------------------+
+ | SPI bus driver | | SPI device driver |
+ +----------------+ +----------------------+
+
+3.2 How do the SPI devices gets discovered and probed ?
+
+In general, the SPI bus driver cannot effective discover devices
+on its bus. Fortunately, the devices on SPI bus usually implemented
+onboard, so the following method has been chosen: the SPI bus driver
+calls the function named spi_bus_populate and passed the `topology
+string' to it. The function will parse the string and call the callback
+for each device, just before registering it. This allows bus driver
+to determine parameters like CS# for each device, retrieve them from
+string and store somewhere like spi_device->platform_data. An example:
+ err = spi_bus_populate( the_spi_bus,
+ "Dev1 0 1 2\0" "Dev2 2 1 0\0",
+ extract_name )
+In this example, function like extract_name would put the '\0' on the
+1st space of device's name, so names will become just "Dev1", "Dev2",
+and the rest of string will become parameters of device.
+
+The another way is to create array of structures spi_device_desc and pass
+this array to function spi_bus_populate2, like this:
+ struct spi_device_desc spi_slaves[] = {
+ [0] = {
+ .name = "device1",
+ .param = device1_params,
+ },
+ [1] = {
+ .name = "device2",
+ .param = NULL,
+ }
+ [2] = {
+ NULL, NULL
+ };
+ spi_bus_populate2( the_spi_bus, spi_slaves, callback );
+
+3.3. DMA and SPI messages
+-------------------------
+
+To handle DMA transfers on SPI bus, any device driver might provide special
+callbacks to allocate/free/get access to buffer. These callbacks are defined
+in subsection iii of section 4.
+To send data using DMA, the buffers should be allocated using
+dma_alloc_coherent function. Usually buffers are allocated statically or
+using kmalloc function.
+To allow drivers to allocate buffers in non-standard
+When one allocates the structure for spi message, it needs to provide target
+device. If its driver wants to allocate buffer in driver-specific way, it may
+provide its own allocation/free methods: alloc and free. If driver does not
+provide these methods, kmalloc and kfree will be used.
+After allocation, the buffer must be accessed to copy the buffer to be send
+or retrieve buffer that has been just received from device. If buffer was
+allocated using driver's alloc method, it(buffer) will be accessed using
+get_buffer. Driver should provide accessible buffer that corresponds buffer
+allocated by driver's alloc method. If there is no get_buffer method,
+the result of alloc will be used.
+After reading/writing from/to buffer, it will be released by call to driver's
+release_buffer method.
+
+
+4. SPI functions are structures reference
+-----------------------------------------
+This section describes structures and functions that listed
+in include/linux/spi.h
+
+i. struct spi_msg
+~~~~~~~~~~~~~~~~~
+
+struct spi_msg {
+ unsigned char flags;
+ unsigned short len;
+ unsigned long clock;
+ struct spi_device* device;
+ void *context;
+ struct list_head link;
+ void (*status)( struct spi_msg* msg, int code );
+ void *devbuf_rd, *devbuf_wr;
+ u8 *databuf_rd, *databuf_wr;
+};
+This structure represents the message that SPI device driver sends to the
+SPI bus driver to handle.
+Fields:
+ flags combination of message flags
+ SPI_M_RD "read" operation (from device to host)
+ SPI_M_WR "write" operation (from host to device)
+ SPI_M_CS assert the CS signal before sending the message
+ SPI_M_CSREL clear the CS signal after sending the message
+ SPI_M_CSPOL set clock polarity to high
+ SPI_M_CPHA set clock phase to high
+ len length, in bytes, of allocated buffer
+ clock reserved, set to zero
+ device the target device of the message
+ context user-defined field; to associate any user data with the message
+ link used by bus driver to queue messages
+ status user-provided callback function to inform about message flow
+ devbuf_rd, devbuf_wr
+ so-called "device buffers". These buffers allocated by the
+ device driver, if device driver provides approproate callback.
+ Otherwise, the kmalloc API will be used.
+ databuf_rd, databuf_wr
+ pointers to access content of device buffers. They are acquired
+ using get_buffer callback, if device driver provides one.
+ Otherwise, they are just pointers to corresponding
+ device buffers
+
+struct spi_msg* spimsg_alloc( struct spi_device* device,
+ unsigned flags,
+ unsigned short len,
+ void (*status)( struct spi_msg*, int code ) )
+This functions is called to allocate the spi_msg structure and set the
+corresponding fields in structure. If device->platform_data provides callbacks
+to handle buffers, alloc/get_buffer are to be used. Returns NULL on errors.
+
+struct void spimsg_free( struct spi_msg* msg )
+Deallocate spi_msg as well as internal buffers. If msg->device->platform_data
+provides callbacks to handle buffers, release_buffer and free are to be used.
+
+u8* spimsg_buffer_rd( struct spi_msg* msg )
+u8* spimsg_buffer_wr( struct spi_msg* msg )
+u8* spimsg_buffer( struct spi_msg* )
+Return the corresponding data buffer, which can be directly modified by driver.
+spimsg_buffer checks flags and return either databuf_rd or databuf_wr basing on
+value of `flags' in spi_msg structure.
+
+ii. struct spi_device
+~~~~~~~~~~~~~~~~~~~~~
+
+struct spi_device
+{
+ char name[ BUS_ID_SIZE ];
+ struct device dev;
+};
+This structure represents the physical device on SPI bus. The SPI bus driver
+will create and register this structure for you.
+ name the name of the device. It should match to the SPI device
+ driver name
+ dev field used to be registered with core
+
+struct spi_device* spi_device_add( struct device* parent,
+ char* name )
+This function registers the device on the spi bus, and set its parent
+to `parent', which represents the SPI bus. The device name will be set to name,
+that should be non-empty, non-NULL string. Returns pointer to allocated
+spi_device structure, NULL on error.
+
+void spi_device_del( struct spi_device* dev )
+Unregister the SPI device. Return value is ignored
+
+iii. struct spi_driver
+~~~~~~~~~~~~~~~~~~~~~~
+
+struct spi_driver {
+ void* (*alloc)( size_t, int );
+ void (*free)( const void* );
+ unsigned char* (*get_buffer)( struct spi_device*, void* );
+ void (*release_buffer)( struct spi_device*, unsigned char*);
+ void (*control)( struct spi_device*, int mode, u32 ctl );
+ struct device_driver driver;
+};
+This structure represents the SPI device driver object. Before registering,
+all fields of driver sub-structure should be properly filled, e.g., the
+`bus_type' should be set to spi_bus. Otherwise, the driver will be incorrectly
+registered and its callbacks might never been called. An example of will-
+formed spi_driver structure:
+ extern struct bus_type spi_bus;
+ static struct spi_driver pnx4008_eeprom_driver = {
+ .driver = {
+ .bus = &spi_bus,
+ .name = "pnx4008-eeprom",
+ .probe = pnx4008_spiee_probe,
+ .remove = pnx4008_spiee_remove,
+ .suspend = pnx4008_spiee_suspend,
+ .resume = pnx4008_spiee_resume,
+ },
+};
+The method control gets called during the processing of SPI message.
+For detailed description of malloc/free/get_buffer/release_buffer, please
+look to section 3.3, "DMA and SPI messages"
+
+
+int spi_driver_add( struct spi_driver* driver )
+Register the SPI device driver with core; returns 0 on no errors, error code
+otherwise.
+
+void spi_driver_del( struct spi_driver* driver )
+Unregisters the SPI device driver; return value ignored.
+
+iv. struct spi_bus_driver
+~~~~~~~~~~~~~~~~~~~~~~~~~
+To handle transactions over the SPI bus, the spi_bus_driver structure must
+be prepared and registered with core. Any transactions, either synchronous
+or asynchronous, go through spi_bus_driver->xfer function.
+
+struct spi_bus_driver
+{
+ int (*xfer)( struct spi_msg* msgs );
+ void (*select) ( struct spi_device* arg );
+ void (*deselect)( struct spi_device* arg );
+
+ struct spi_msg *(*retrieve)( struct spi_bus_driver *this, struct spi_bus_data *bd);
+ void (*reset)( struct spi_bus_driver *this, u32 context);
+
+ struct device_driver driver;
+};
+
+Fields:
+ xfer pointer to function to execute actual transaction on SPI bus
+ msg message to handle
+ select pointer to function that gets called when bus needs to
+ select another device to be target of transfers
+ deselect
+ pointer to function that gets called before another device
+ is selected to be the target of transfers
+ reset
+ pointer to function that performs reset of SPI bus
+ retrieve
+ this function is used to retrieve next message from queue. If NULL,
+ spi_bus_fifo_retrieve is used
+
+
+spi_bus_driver_register( struct spi_bus_driver* )
+
+Register the SPI bus driver with the system. The driver sub-structure should
+be properly filled before using this function, otherwise you may get unpredi-
+ctable results when trying to exchange data. An example of correctly prepared
+spi_bus_driver structure:
+ static struct spi_bus_driver spipnx_driver = {
+ .driver = {
+ .bus = &platform_bus_type,
+ .name = "spipnx",
+ .probe = spipnx_probe,
+ .remove = spipnx_remove,
+ .suspend = spipnx_suspend,
+ .resume = spipnx_resume,
+ },
+ .xfer = spipnx_xfer,
+};
+The driver and corresponding platform device are matched by name, so, in
+order the example abive to work, the platform_device named "spipnx" should
+be registered somewhere.
+
+void spi_bus_driver_unregister( struct spi_bus_driver* )
+
+Unregister the SPI bus driver registered by call to spi_buys_driver_register
+function; returns void.
+
+int spi_bus_populate( struct device* parent,
+ char* devices,
+ void (*callback)( struct device* parent, struct spi_device* new_one ) )
+This function usually called by SPI bus drivers in order to populate the SPI
+bus (see also section 3.2, "How the SPI devices gets discovered and probed ?").
+After creating the spi_device, the spi_bus_populate calls the `callback'
+function to allow to modify spi_device's fields before registering it with core.
+ parent pointer to SPI bus
+ devices string representing the current topology of SPI bus. It should
+ be formed like
+ "dev-1_and_its_info\0dev-2_and_its_info\0another_device\0\0"
+ the spi_bus_populate delimits this string by '\0' characters,
+ creates spi_device and after calling the callback registers the
+ spi_device
+ callback
+ pointer to function which could modify spi_device fields just
+ before registering them with core
+
+int spi_bus_populate2 (struct device *parent, struct spi_device_desc *devices,
+ void (*callback) (struct device* parent, struct spi_device* new_dev,
+ void *params ))
+This is another way to populate the bus; but instead of string with device names and
+parameters, the array of structures spi_device_desc is passed. Each item in array describes
+the SPI slave device to create.
+
+
+v. spi_transfer and spi_queue
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+The driver that uses SPI core can initiate transfers either by calling
+spi_transfer function (that will wait till the transfer is funished) or
+queueing the message using spi_queue function (you need to provide function
+that will be called during message is processed). In any way, you need to
+prepare the spimsg structure and know the target device to your message to
+be sent.
+
+int spi_transfer( struct spi_msg msgs,
+ void (*callback)( struct spi_msg* msg, int ) )
+If callback is zero, start synchronous transfer. Otherwise, queue
+the message.
+ msg message to be handled
+ callback the callback function to be called during
+ message processing. If NULL, the function
+ will wait until end of processing.
+
+int spi_queue( struct spi_device* device,
+ struct spi_msg* msg )
+Queue the only message to the device. Returns status of queueing. To obtain
+status of message processing, you have to provide `status' callback in message
+and examine its parameters
+ msg message to be queued
+
+vi. the spi_bus variable
+~~~~~~~~~~~~~~~~~~~~~~~~
+This variable is created during initialization of spi core, and has to be
+specified as `bus' on any SPI device driver (look to section iii, "struct
+spi_driver" ). If you do not specify spi_bus, your driver will be never
+matched to spi_device and never be probed with hardware. Note that
+spi_bus.match points to function that matches drivers and devices by name,
+so SPI devices and their drivers should have the same name.
+
+5. How to contact authors
+-------------------------
+Do you have any comments ? Enhancements ? Device driver ? Feel free
+to contact me:
+ [email protected]
+ [email protected]
+Visit our project page:
+ http://spi-devel.sourceforge.net
+Subscribe to mailing list:
+ [email protected]
+
Index: linux/include/linux/spi.h
===================================================================
--- /dev/null
+++ linux/include/linux/spi.h
@@ -0,0 +1,285 @@
+/*
+ * linux/include/linux/spi/spi.h
+ *
+ * Copyright (C) 2005 MontaVista Software, Inc <[email protected]>
+ *
+ * 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.
+ *
+ * Derived from l3.h by Jamey Hicks
+ */
+#ifndef SPI_H
+#define SPI_H
+
+#include <linux/types.h>
+#include <linux/device.h>
+
+#define kzalloc(size,type) kcalloc(1,size,type)
+
+struct spi_device;
+struct spi_driver;
+struct spi_msg;
+struct spi_bus_driver;
+
+extern struct bus_type spi_bus;
+
+struct spi_bus_data {
+ struct semaphore lock;
+ struct list_head msgs;
+ atomic_t exiting;
+ struct task_struct *thread;
+ wait_queue_head_t queue;
+ struct spi_device *selected_device;
+ struct spi_bus_driver *bus;
+ char *id;
+};
+
+#define TO_SPI_BUS_DRIVER(drv) container_of( drv, struct spi_bus_driver, driver )
+struct spi_bus_driver {
+ int (*xfer) (struct spi_msg * msg);
+ void (*select) (struct spi_device * dev);
+ void (*deselect) (struct spi_device * dev);
+ void (*set_clock) (struct device * bus_device, u32 clock_hz);
+ void (*reset) (struct device *bus_device, u32 context);
+ struct spi_msg *
+ (*retrieve) (struct spi_bus_driver *bus, struct spi_bus_data *data);
+ struct device_driver driver;
+};
+
+#define TO_SPI_DEV(device) container_of( device, struct spi_device, dev )
+struct spi_device {
+ char name[BUS_ID_SIZE];
+ void *private;
+ struct device dev;
+};
+
+#define TO_SPI_DRIVER(drv) container_of( drv, struct spi_driver, driver )
+struct spi_driver {
+ void *(*alloc) (size_t, int);
+ void (*free) (const void *);
+ unsigned char *(*get_buffer) (struct spi_device *, void *);
+ void (*release_buffer) (struct spi_device *, unsigned char *);
+ void (*control) (struct spi_device *, int mode, u32 ctl);
+ struct device_driver driver;
+};
+
+#define SPI_DEV_DRV( device ) TO_SPI_DRIVER( (device)->dev.driver )
+
+#define spi_device_lock( dev ) /* down( dev->dev.sem ) */
+#define spi_device_unlock( dev ) /* up( dev->dev.sem ) */
+
+/*
+ * struct spi_msg
+ *
+ * This structure represent the SPI message internally. You should never use fields of this structure directly
+ * Please use corresponding functions to create/destroy/access fields
+ *
+ */
+struct spi_msg {
+ u32 flags;
+#define SPI_M_RD 0x00000001
+#define SPI_M_WR 0x00000002 /**< Write mode flag */
+#define SPI_M_CSREL 0x00000004 /**< CS release level at end of the frame */
+#define SPI_M_CS 0x00000008 /**< CS active level at begining of frame ( default low ) */
+#define SPI_M_CPOL 0x00000010 /**< Clock polarity */
+#define SPI_M_CPHA 0x00000020 /**< Clock Phase */
+#define SPI_M_EXTBUF 0x80000000 /** externally allocated buffers */
+#define SPI_M_ASYNC_CB 0x40000000 /** use workqueue to deliver callbacks */
+#define SPI_M_DNA 0x20000000 /** do not allocate buffers */
+
+ unsigned short len; /* msg length */
+ unsigned long clock;
+ struct spi_device *device;
+ void *context;
+ void *arg;
+ void *parent; /* in case of complex messages */
+ struct list_head link;
+
+ void (*status) (struct spi_msg * msg, int code);
+
+ void *devbuf_rd, *devbuf_wr;
+ u8 *databuf_rd, *databuf_wr;
+
+ struct work_struct wq_item;
+};
+
+static inline struct spi_msg *spimsg_alloc(struct spi_device *device,
+ unsigned flags,
+ unsigned short len,
+ void (*status) (struct spi_msg *,
+ int code))
+{
+ struct spi_msg *msg;
+ struct spi_driver *drv = SPI_DEV_DRV(device);
+ int msgsize = sizeof (struct spi_msg);
+
+ if (drv->alloc || (flags & (SPI_M_RD|SPI_M_WR)) == (SPI_M_RD | SPI_M_WR ) ) {
+ pr_debug ( "%s: external buffers\n", __FUNCTION__ );
+ flags |= SPI_M_EXTBUF;
+ } else {
+ pr_debug ("%s: no ext buffers, msgsize increased from %d by %d to %d\n", __FUNCTION__,
+ msgsize, len, msgsize + len );
+ msgsize += len;
+ }
+
+ msg = kmalloc( msgsize, GFP_KERNEL);
+ if (!msg)
+ return NULL;
+ memset(msg, 0, sizeof(struct spi_msg));
+ msg->len = len;
+ msg->status = status;
+ msg->device = device;
+ msg->flags = flags;
+ INIT_LIST_HEAD(&msg->link);
+
+ if (flags & SPI_M_DNA )
+ return msg;
+
+ /* otherwise, we need to set up pointers */
+ if (!(flags & SPI_M_EXTBUF)) {
+ msg->databuf_rd = msg->databuf_wr =
+ (u8*)msg + sizeof ( struct spi_msg);
+ } else {
+ if (flags & SPI_M_RD) {
+ msg->devbuf_rd = drv->alloc ?
+ drv->alloc(len, GFP_KERNEL) : kmalloc(len, GFP_KERNEL);
+ msg->databuf_rd = drv->get_buffer ?
+ drv->get_buffer(device, msg->devbuf_rd) : msg->devbuf_rd;
+ }
+ if (flags & SPI_M_WR) {
+ msg->devbuf_wr = drv->alloc ?
+ drv->alloc(len, GFP_KERNEL) : kmalloc(len, GFP_KERNEL);
+ msg->databuf_wr = drv->get_buffer ?
+ drv->get_buffer(device, msg->devbuf_wr) : msg->devbuf_wr;
+ }
+ }
+ pr_debug("%s: msg = %p, RD=(%p,%p) WR=(%p,%p). Actual flags = %s+%s\n",
+ __FUNCTION__,
+ msg,
+ msg->devbuf_rd, msg->databuf_rd,
+ msg->devbuf_wr, msg->databuf_wr,
+ msg->flags & SPI_M_RD ? "RD" : "~rd",
+ msg->flags & SPI_M_WR ? "WR" : "~wr");
+ return msg;
+}
+
+static inline void spimsg_free(struct spi_msg *msg)
+{
+ void (*do_free) (const void *) = kfree;
+ struct spi_driver *drv = SPI_DEV_DRV(msg->device);
+
+ if (msg) {
+
+ if ( !(msg->flags & SPI_M_DNA) || (msg->flags & SPI_M_EXTBUF) ) {
+ if (drv->free)
+ do_free = drv->free;
+ if (drv->release_buffer) {
+ if (msg->databuf_rd)
+ drv->release_buffer(msg->device,
+ msg->databuf_rd);
+ if (msg->databuf_wr)
+ drv->release_buffer(msg->device,
+ msg->databuf_wr);
+ }
+ if (msg->devbuf_rd)
+ do_free(msg->devbuf_rd);
+ if (msg->devbuf_wr)
+ do_free(msg->devbuf_wr);
+ }
+ kfree(msg);
+ }
+}
+
+static inline u8 *spimsg_buffer_rd(struct spi_msg *msg)
+{
+ return msg ? msg->databuf_rd : NULL;
+}
+
+static inline u8 *spimsg_buffer_wr(struct spi_msg *msg)
+{
+ return msg ? msg->databuf_wr : NULL;
+}
+
+static inline u8 *spimsg_buffer(struct spi_msg *msg)
+{
+ if (!msg)
+ return NULL;
+ if ((msg->flags & (SPI_M_RD | SPI_M_WR)) == (SPI_M_RD | SPI_M_WR)) {
+ printk(KERN_ERR "%s: what buffer do you really want ?\n",
+ __FUNCTION__);
+ return NULL;
+ }
+ if (msg->flags & SPI_M_RD)
+ return msg->databuf_rd;
+ if (msg->flags & SPI_M_WR)
+ return msg->databuf_wr;
+}
+
+static inline void spimsg_set_rd( struct spi_msg* msg, void* buf )
+{
+ msg->databuf_rd = buf;
+}
+
+static inline void spimsg_set_wr (struct spi_msg *msg, void *buf )
+{
+ msg->databuf_wr = buf;
+}
+
+
+#define SPIMSG_OK 0x01
+#define SPIMSG_FAILED 0x80
+#define SPIMSG_STARTED 0x02
+#define SPIMSG_DONE 0x04
+
+#define SPI_MAJOR 153
+
+struct spi_driver;
+struct spi_device;
+
+static inline int spi_bus_driver_register (struct spi_bus_driver *bus_driver )
+{
+ return driver_register (&bus_driver->driver);
+}
+
+void spi_bus_driver_unregister(struct spi_bus_driver *);
+int spi_bus_driver_init(struct spi_bus_driver *driver, struct device *dev);
+struct spi_device* spi_device_add(struct device *parent, char *name, void *private);
+
+static inline void spi_device_del(struct spi_device *dev)
+{
+ device_unregister(&dev->dev);
+}
+static inline int spi_driver_add(struct spi_driver *drv)
+{
+ drv->driver.bus = &spi_bus;
+ return driver_register(&drv->driver);
+}
+static inline void spi_driver_del(struct spi_driver *drv)
+{
+ driver_unregister(&drv->driver);
+}
+
+#define SPI_DEV_CHAR "spi-char"
+
+extern void spi_bus_reset(struct device* bus, u32 context);
+extern int spi_write(struct spi_device *dev, const char *buf, int len);
+extern int spi_read(struct spi_device *dev, char *buf, int len);
+
+extern int spi_queue(struct spi_msg *message);
+extern int spi_transfer(struct spi_msg *message,
+ void (*status) (struct spi_msg *, int));
+extern int spi_bus_populate(struct device *parent, char *device,
+ void (*assign) (struct device *parent,
+ struct spi_device *));
+struct spi_device_desc {
+ char* name;
+ void* params;
+};
+extern int spi_bus_populate2(struct device *parent,
+ struct spi_device_desc *devices,
+ void (*assign) (struct device *parent,
+ struct spi_device *,
+ void *));
+
+#endif /* SPI_H */
--- dmitry pervushin <[email protected]> wrote:
> This is an updated version of SPI framework from me, Dmitry Pervushin.
> It seems that now it is good time to consolidate our SPI frameworks to
> push it to kernel :)
>
> We've tested our SPI core as well with bus drivers with wireless LAN
> driver and achieved good performance with relatively small overhead.
> This proves the viability of this framework in real life even in
> real-time environment. The size of .text is
> still about 2,500 bytes, that is comparable with David Brownell's
> framework size.
>
> I think now is the time to start the final convergence process for these
> two cores and get the final core
> into the mainline kernel. And in order to understand where we need to
> converge, I created the main differences
> list (see below).
>
> The list of main differences between David Brownell's SPI framework (A)
> and my one (B):
> - (A) uses more complicated structure of SPI message, that contains one
> or more atomic transfers, and (B)
> offers the only spi_msg that represents the atomic transfer on SPI bus.
> The similar approach can be imple-
> mented in (B), and actually is implemented. But my imp[ression is that
> such enhancement may be added later..
I wouldn't have said that the message structure in (A) is more complex then (B). For example, in
(B) you have many flags which controls things like SPI mode which are not needed in every message.
Once the SPI controller has been setup for a particular slave device you don't need to constantly
send this information.
In (B) how to do you handle SPI devices which require to send several messages with out releasing
their cs? There are/will be some devices which require this.
> - (A) uses workqueues to queue and handle SPI messages, and (B)
> allocates the kernel thread to the same purpose.
> Using workqueues is not very good solution in real-time environment; I
> think that allocating and starting the
> separate thread will give us more predictable and stable results;
Where does (A) use a workqueue? (A) doesn't use a workqueue or thread and instead leaves it up to
the adapter driver how to handle the messages that it gets sent (which in the case of some drivers
will mean no thread or workqueue). (B) is _forcing_ a thread on the adapter which the adapter may
not need.
> - (A) has some assumptions on buffers that are passed down to spi
> functions. If some controller driver (or bus driver
> in terms of (B)) tries to perform DMA transfers, it must copy the
> passed buffers to some memory allocated
> using GFP_DMA flag and map it using dma_map_single. From the other
> hand, (B) relies on callbacks provided
> by SPI device driver to allocate memory for DMA transfers, but keeps
> ability to pass user-allocated buffers down
> to SPI functions by specifying flags in SPI message. SPI message being
> a fundamental essense looks better to me when
> it's as simple as possible. Especially when we don't lose any
> flexibility which is exacly our case (buffers that are
> allocated as well as message itself/provided by user, DMA-capable
> buffers..)
But allocating and freeing buffer is a core kernel thing not a SPI thing. To me you are adding
more complexity then is needed and your saying this is keeping things simple?
> - (A) retrieves SPI message from the queue in sequential order (FIFO),
> but (B) provides more flexible way by providing
> special callback to retrieve next message from queue. This callback may
> implement its own discipline of scheduling
> SPI messages. In any way, the default is FIFO.
I think (A) is missing a method of adding extra spi_message(s) in callback to extend the current
transfer on that SPI device. I can imagine a case where you will be required to read status
information from a device and in this status information is the length of the data it has just
received (for example if it was a network adapter). Straight after reading this information the
device would start sending the data it has received but when the read status message was issued
the length of the data wasn't known. Currently with (A) we would have to stop the transfer and
restart the whole thing again, this time using the length of the data we found form the last
message.
A better solution would to be able to add an extra message during the callback from the first
message as now we know then length we can setup a transfer that would be the correct size.
However, this message must be the next message that the adapter sends as if another message for
another SPI device was sent before then the cs line of the device we are talking to would be
deselected and we would have to start again.
My proposal is that in the callback from a spi_message being sent it returns a pointer to the next
spi_message which the adapter will send before it continues sending any other messages (this is
like the adapter being locked by the SPI device), if no other messages need to be sent atomically
in the callback of current message then the SPI device driver would just return NULL.
Example:
========
/* callback/complete routine of a SPI device/protocol driver */
int my_spi_callback (void *data)
{
struct my_status_struct = data;
struct spi_message read_message;
/* Check to see if we have received any data */
if (my_status_struct->read_length)
{
/* Create a new spi_message to read the data which will be
* the very next thing the device will send */
read_message = kzalloc(...)
...
return read_message;
}
else
/* No data to be read so don't append another message */
return NULL;
}
> - (A) uses standartized way to provide CS information, and (B) relies on
> functional drivers callbacks, which looks more
> flexible to me.
I'm not sure what you mean here. You need to provide the cs numbers with SPI device in order for
the core to create the unique addres and entry in sysfs.
However, (A) is not checking to see if the cs that a registering device wants to use is already in
use, this needs to be added, and the same is true for registering spi masters.
Mark
___________________________________________________________
To help you stay safe and secure online, we've developed the all new Yahoo! Security Centre. http://uk.security.yahoo.com
Mark Underwood wrote:
>>The list of main differences between David Brownell's SPI framework (A)
>>and my one (B):
>>- (A) uses more complicated structure of SPI message, that contains one
>>or more atomic transfers, and (B)
>> offers the only spi_msg that represents the atomic transfer on SPI bus.
>>The similar approach can be imple-
>> mented in (B), and actually is implemented. But my imp[ression is that
>>such enhancement may be added later..
>>
>>
>
>I wouldn't have said that the message structure in (A) is more complex then (B). For example, in
>(B) you have many flags which controls things like SPI mode which are not needed in every message.
>Once the SPI controller has been setup for a particular slave device you don't need to constantly
>send this information.
>In (B) how to do you handle SPI devices which require to send several messages with out releasing
>their cs? There are/will be some devices which require this.
>
>
Please see the explanation for the 'flags' in Documentation/spi.txt
within the patch.
>
>
>>- (A) uses workqueues to queue and handle SPI messages, and (B)
>>allocates the kernel thread to the same purpose.
>> Using workqueues is not very good solution in real-time environment; I
>>think that allocating and starting the
>> separate thread will give us more predictable and stable results;
>>
>>
>
>Where does (A) use a workqueue? (A) doesn't use a workqueue or thread and instead leaves it up to
>the adapter driver how to handle the messages that it gets sent (which in the case of some drivers
>will mean no thread or workqueue). (B) is _forcing_ a thread on the adapter which the adapter may
>not need.
>
>
I bet the drivers that don't need neither threads not workqueue there's
no need in async transfers as well. :)
On the other hand, threads is a flexible mechanism for handling async
stuff, and there won't be a lot of threads so the overhead won't be high.
You might also want to ask why you can't change the steering wheel
placement in your car from right-side to rleft-side although you travel
by car to continental Europe once per decade. ;-)
>
>
>>- (A) has some assumptions on buffers that are passed down to spi
>>functions. If some controller driver (or bus driver
>> in terms of (B)) tries to perform DMA transfers, it must copy the
>>passed buffers to some memory allocated
>> using GFP_DMA flag and map it using dma_map_single. From the other
>>hand, (B) relies on callbacks provided
>> by SPI device driver to allocate memory for DMA transfers, but keeps
>>ability to pass user-allocated buffers down
>> to SPI functions by specifying flags in SPI message. SPI message being
>>a fundamental essense looks better to me when
>> it's as simple as possible. Especially when we don't lose any
>>flexibility which is exacly our case (buffers that are
>> allocated as well as message itself/provided by user, DMA-capable
>>buffers..)
>>
>>
>
>But allocating and freeing buffer is a core kernel thing not a SPI thing. To me you are adding
>more complexity then is needed and your saying this is keeping things simple?
>
>
I'm afraid that you're not quite getting the whole concept. The concept
is to provide thorough and stable solution.
Given that the buffer passed is declared as, say, static, the whole
kernel might crash if we try to pass it to DMA. David's core itself is
not capable of filtering that and letting the driver decide adds more
complexity to the driver.
If we're choosing between adding complexity to the core and adding it to
the particular drivers, it's definitely better to add it to the core cuz
it's done _once_.
>>- (A) uses standartized way to provide CS information, and (B) relies on
>>functional drivers callbacks, which looks more
>> flexible to me.
>>
>>
>
>I'm not sure what you mean here. You need to provide the cs numbers with SPI device in order for
>the core to create the unique addres and entry in sysfs.
>However, (A) is not checking to see if the cs that a registering device wants to use is already in
>use, this needs to be added, and the same is true for registering spi masters.
>
>
Can you please elaborate on that?
Vitaly
--- Vitaly Wool <[email protected]> wrote:
> Mark Underwood wrote:
>
> >>The list of main differences between David Brownell's SPI framework (A)
> >>and my one (B):
> >>- (A) uses more complicated structure of SPI message, that contains one
> >>or more atomic transfers, and (B)
> >> offers the only spi_msg that represents the atomic transfer on SPI bus.
> >>The similar approach can be imple-
> >> mented in (B), and actually is implemented. But my imp[ression is that
> >>such enhancement may be added later..
> >>
> >>
> >
> >I wouldn't have said that the message structure in (A) is more complex then (B). For example,
> in
> >(B) you have many flags which controls things like SPI mode which are not needed in every
> message.
> >Once the SPI controller has been setup for a particular slave device you don't need to
> constantly
> >send this information.
> >In (B) how to do you handle SPI devices which require to send several messages with out
> releasing
> >their cs? There are/will be some devices which require this.
> >
> >
> Please see the explanation for the 'flags' in Documentation/spi.txt
> within the patch.
I can see that you can leave cs active at the end of a transfer but that's not my point. How do
you make sure that message for other SPI devices don't get send while the cs of the current device
is high?
>
> >
> >
> >>- (A) uses workqueues to queue and handle SPI messages, and (B)
> >>allocates the kernel thread to the same purpose.
> >> Using workqueues is not very good solution in real-time environment; I
> >>think that allocating and starting the
> >> separate thread will give us more predictable and stable results;
> >>
> >>
> >
> >Where does (A) use a workqueue? (A) doesn't use a workqueue or thread and instead leaves it up
> to
> >the adapter driver how to handle the messages that it gets sent (which in the case of some
> drivers
> >will mean no thread or workqueue). (B) is _forcing_ a thread on the adapter which the adapter
> may
> >not need.
> >
> >
> I bet the drivers that don't need neither threads not workqueue there's
> no need in async transfers as well. :)
> On the other hand, threads is a flexible mechanism for handling async
> stuff, and there won't be a lot of threads so the overhead won't be high.
> You might also want to ask why you can't change the steering wheel
> placement in your car from right-side to rleft-side although you travel
> by car to continental Europe once per decade. ;-)
Sorry I'm not following you here.
Example:
An interrupt driven PIO doesn't need a thread or a workqueue. When it is idle the call to its
transfer function can start off the first transfer and after that the interrupt routine will check
for a new transfer when it has finished the current one. David provided other examples so if you
are still not sure search through the archives.
>
> >
> >
> >>- (A) has some assumptions on buffers that are passed down to spi
> >>functions. If some controller driver (or bus driver
> >> in terms of (B)) tries to perform DMA transfers, it must copy the
> >>passed buffers to some memory allocated
> >> using GFP_DMA flag and map it using dma_map_single. From the other
> >>hand, (B) relies on callbacks provided
> >> by SPI device driver to allocate memory for DMA transfers, but keeps
> >>ability to pass user-allocated buffers down
> >> to SPI functions by specifying flags in SPI message. SPI message being
> >>a fundamental essense looks better to me when
> >> it's as simple as possible. Especially when we don't lose any
> >>flexibility which is exacly our case (buffers that are
> >> allocated as well as message itself/provided by user, DMA-capable
> >>buffers..)
> >>
> >>
> >
> >But allocating and freeing buffer is a core kernel thing not a SPI thing. To me you are adding
> >more complexity then is needed and your saying this is keeping things simple?
> >
> >
> I'm afraid that you're not quite getting the whole concept. The concept
> is to provide thorough and stable solution.
> Given that the buffer passed is declared as, say, static, the whole
> kernel might crash if we try to pass it to DMA. David's core itself is
> not capable of filtering that and letting the driver decide adds more
> complexity to the driver.
> If we're choosing between adding complexity to the core and adding it to
> the particular drivers, it's definitely better to add it to the core cuz
> it's done _once_.
I'm not 100% sure how David is handling this, but one option would be to have a not_dmaable flag
which states that the buffers used in this message are not DMAable and in this case the adapter
driver will either do a PIO transfer or bounce the data to/from a DMAable buffer it allocated
itself. I don't see why a SPI adapter driver needs to supply alloc/free callbacks when a simple
flag would do the job.
>
> >>- (A) uses standartized way to provide CS information, and (B) relies on
> >>functional drivers callbacks, which looks more
> >> flexible to me.
> >>
> >>
> >
> >I'm not sure what you mean here. You need to provide the cs numbers with SPI device in order
> for
> >the core to create the unique addres and entry in sysfs.
> >However, (A) is not checking to see if the cs that a registering device wants to use is already
> in
> >use, this needs to be added, and the same is true for registering spi masters.
> >
> >
> Can you please elaborate on that?
If I register a SPI device on cs1 of spi-1 and later try to register another device on cs1 of
spi-1 I would expect the spi core layer to fail the registration of the second device.
Mark
>
> Vitaly
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to [email protected]
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/
>
___________________________________________________________
To help you stay safe and secure online, we've developed the all new Yahoo! Security Centre. http://uk.security.yahoo.com
On Monday 21 November 2005 12:15 pm, Mark Underwood wrote:
>
> --- dmitry pervushin <[email protected]> wrote:
>
> > The list of main differences between David Brownell's SPI framework (A)
> > and my one (B):
A == http://marc.theaimsgroup.com/?l=linux-kernel&m=113169588230519&w=2
plus a handful of rather minor tweaks (ssize_t, comments, etc).
B == http://sourceforge.net/mailarchive/message.php?msg_id=13824397
but it needs updates to match current 2.6.15-rc code.
> > - (A) uses more complicated structure of SPI message, that contains one
> > or more atomic transfers, and (B)
> > offers the only spi_msg that represents the atomic transfer on SPI bus.
> > The similar approach can be implemented
> > in (B), and actually is implemented. But my imp[ression is that
> > such enhancement may be added later..
>
> I wouldn't have said that the message structure in (A) is more complex then (B). For example, in
> (B) you have many flags which controls things like SPI mode which are not needed in every message.
> Once the SPI controller has been setup for a particular slave device you don't need to constantly
> send this information.
And in fact, constantly sending it means some drivers will have to waste
time constantly checking it, in case it changed. If that setup is stored
in controller registers, it's a lot better to just have the setup() call
be responsible for changing the communication parameters. (This is the
approach used by both MMC and PCMCIA, for what it's worth...)
> In (B) how to do you handle SPI devices which require to send several messages with out releasing
> their cs? There are/will be some devices which require this.
In fact, that's why the transfer segments are grouped. One builds SPI
protocol requests out of several such segments. A very common idiom is
writing a command, then reading its response. Chipselect must stay
active during the whole sequence.
Adding support for such a basic mechanism "later" doesn't seem like
a good idea to me.
> > - (A) uses workqueues to queue and handle SPI messages, and (B)
> > allocates the kernel thread to the same purpose.
> > Using workqueues is not very good solution in real-time environment; I
> > think that allocating and starting the
> > separate thread will give us more predictable and stable results;
>
> Where does (A) use a workqueue? (A) doesn't use a workqueue or thread and instead leaves it up to
> the adapter driver how to handle the messages that it gets sent (which in the case of some drivers
> will mean no thread or workqueue). (B) is _forcing_ a thread on the adapter which the adapter may
> not need.
Exactly. That's one of the things I meant when I recently listed some of the
top goals of the framework I did:
(a) SPI shouldn't perpetuate the driver model botches of I2C;
(b) ditto I2C's "everything is synchronous" botches;
(c) it should work well with DMA, to support things like DataFlash;
(d) given the variety of SPI chips, protocol controls are needed;
(e) place minimal implementation constraints on controller drivers.
So for example one way you know that (c) is well met is that it's the same
approach used in USB (both host and peripheral/gadget sides); that's been
working well for quite a few years now. (Despite comments from Dmitry
and Vitaly to the contrary.)
> > - (A) has some assumptions on buffers that are passed down to spi
> > functions.
Make that "requirements"; FWIW they're the same ones that apply to all
other kernel driver frameworks I've seen: that buffers be DMA-safe.
It would not be helpful (IMO) to define different rules; that's also
called the "Principle of Least Astonishment". :)
> > If some controller driver (or bus driver
> > in terms of (B)) tries to perform DMA transfers, it must copy the
> > passed buffers to some memory allocated
> > using GFP_DMA flag and map it using dma_map_single.
Based on this and other comments from Dmitry/Vitaly, I suspect they
don't see how the Linux DMA APIs work. The correct statement is that if
a controller driver wants to use DMA, it must dma_{map,unmap}_single().
The upper level drivers don't _need_ to worry about that.
However, some key infrastructure is in place to let SPI protocol drivers
(the ones passing messages through the controller then the bus) provide
pre-mapped buffers if the eventually _want_ to do that. They'd likely be
allocated with dma_alloc_coherent() or through a dma_pool. That would be
useful for cases like an MMC/SD block driver that talks SPI, since the
scatterlists will come down from the block layer ... lower level drivers
should be able to ignore details like how dma_{map,unmap}_sg() works.
> > From the other
> > hand, (B) relies on callbacks provided
> > by SPI device driver to allocate memory for DMA transfers, but keeps
> > ability to pass user-allocated buffers down
> > to SPI functions by specifying flags in SPI message. SPI message being
> > a fundamental essense looks better to me when
> > it's as simple as possible. Especially when we don't lose any
> > flexibility which is exacly our case (buffers that are
> > allocated as well as message itself/provided by user, DMA-capable
> > buffers..)
>
> But allocating and freeing buffer is a core kernel thing not a SPI thing. To me you are adding
> more complexity then is needed and your saying this is keeping things simple?
That's how I read his comments too. Moreover, that particular kind of
complexity is the confusing kind ... it makes it a lot harder to see what's
going on, since it's all hidden behind layers of indirection. Indirection
is of course useful sometimes. But not in this case, where there are much
simpler idioms, with the advantage that most other kernel APIs use them.
> > - (A) retrieves SPI message from the queue in sequential order (FIFO),
Only with respect to a given device. It would make no sense to reorder the
queue so that writing X, then Y, then Z would morph into "X Z Y" or "Z Y X". :)
It's specifically _undefined_ how requests going to different devices are ordered.
Some hardware will be happier if things are synchronized (e.g. to a vertical
retrace IRQ), some systems might need to prioritize certain devices, and so on.
I do think FIFO makes a good general policy, for boards without any of those
special requirements.
> > but (B) provides more flexible way by providing
> > special callback to retrieve next message from queue. This callback may
> > implement its own discipline of scheduling
> > SPI messages. In any way, the default is FIFO.
>
> I think (A) is missing a method of adding extra spi_message(s) in callback to extend the current
> transfer on that SPI device. I can imagine a case where you will be required to read status
> information from a device and in this status information is the length of the data it has just
> received (for example if it was a network adapter). Straight after reading this information the
> device would start sending the data it has received but when the read status message was issued
> the length of the data wasn't known.
Do you actually have hardware which works that way? That would be an example
of a system that needs some specific prioritization of transfers (see below).
> Currently with (A) we would have to stop the transfer and
> restart the whole thing again, this time using the length of the data we found form the last
> message.
Well, each transfer segement would clearly stop, but if that segment had
the flag set which says "leave chipselect active", then the controller
driver would have the flexibility to prioritize transfers to that chip.
> A better solution would to be able to add an extra message during the callback from the first
> message as now we know then length we can setup a transfer that would be the correct size.
> However, this message must be the next message that the adapter sends as if another message for
> another SPI device was sent before then the cs line of the device we are talking to would be
> deselected and we would have to start again.
I suspect that in terms of API, *if* that semantic is really needed (as in,
you have hardware that needs it) then it should be made into a flag in
the spi_message.
Clearly, it'd make things more complicated for the SPI controller driver;
drivers that don't implement that semantic would need to know when it's
required, so they could fail cleanly. And drivers that _do_ know about it
would need to avoid doing things like shuffling completions off to some
tasklet (while starting the next transfer ASAP, getting I/O overlap) ...
they'd need to make stronger guarantees about transaction sequencing, at
a certain cost in terms of potential throughput.
> My proposal is that in the callback from a spi_message being sent it returns a pointer to the next
> spi_message which the adapter will send before it continues sending any other messages (this is
> like the adapter being locked by the SPI device), if no other messages need to be sent atomically
> in the callback of current message then the SPI device driver would just return NULL.
The thing I don't like about that model is that, just like the Linux-USB API
for interrupt and isochronous transfers in the 2.4 kernels, it swallows fault
modes so that drivers can't know when things break.
Better IMO to just keep the same API in all cases, and require that callback
to directly submit a (new?) transfer if that's needed. If the controller
can't accept it, it'll know right away, and then the protocol driver will be
able to do something appropriate.
> > - (A) uses standartized way to provide CS information, and (B) relies on
> > functional drivers callbacks, which looks more
> > flexible to me.
>
> I'm not sure what you mean here. You need to provide the cs numbers with SPI device in order for
> the core to create the unique addres and entry in sysfs.
I'm not sure what he means either. :)
Stephen's PXA2xx SPI driver uses callbacks internally, but that's kind of
specific to that PXA hardware ... there's no chipselect handled by the
controller, one of the dozens of GPIOs must be chosen and that's clearly
a board-specific mechanism (uses controller_data as I recall). He tells
me he plans to post the latest version of that -- many updates including
PXA255 SSP support (not just NSSP) and code shrinkage -- early next week.
But most of the SPI controllers I've seen just have a fixed number of
chipselects, typically four, handled directly by the controller. That's
why the "standardized way" is just to use a 0..N chipselect number.
> However, (A) is not checking to see if the cs that a registering device wants to use is already in
> use, this needs to be added, and the same is true for registering spi masters.
Yes, I even have a "FIXME Paranoia argues that ..." comment in that code. I think
the best way to handle that is probably to get the driver name out of the device
name, thereby punting that check to the driver model. So the devices would have
names like "spi3.2" and the driver name would be in "modalias"; I think that'll
be a simple enough change to the framework, now that I've thought of it. This is
not a case where we _need_ to act much like platform_device.
- Dave
David Brownell wrote:
>>>- (A) uses more complicated structure of SPI message, that contains one
>>> or more atomic transfers, and (B)
>>> offers the only spi_msg that represents the atomic transfer on SPI bus.
>>> The similar approach can be implemented
>>> in (B), and actually is implemented. But my imp[ression is that
>>> such enhancement may be added later..
>>>
>>>
>>I wouldn't have said that the message structure in (A) is more complex then (B). For example, in
>>(B) you have many flags which controls things like SPI mode which are not needed in every message.
>>Once the SPI controller has been setup for a particular slave device you don't need to constantly
>>send this information.
>>
>>
>
>And in fact, constantly sending it means some drivers will have to waste
>time constantly checking it, in case it changed. If that setup is stored
>in controller registers, it's a lot better to just have the setup() call
>be responsible for changing the communication parameters. (This is the
>approach used by both MMC and PCMCIA, for what it's worth...)
>
>
>
I'm not aware if MMC/PCMCIA guys are happy with this approach :), but
anyway what you're talking about here makes sense.
>
>
>>In (B) how to do you handle SPI devices which require to send several messages with out releasing
>>their cs? There are/will be some devices which require this.
>>
>>
>
>In fact, that's why the transfer segments are grouped. One builds SPI
>protocol requests out of several such segments. A very common idiom is
>writing a command, then reading its response. Chipselect must stay
>active during the whole sequence.
>
>Adding support for such a basic mechanism "later" doesn't seem like
>a good idea to me.
>
>
It is supported by means of flags. I'm afraid your note is pointless here.
>
>
>
>>>- (A) uses workqueues to queue and handle SPI messages, and (B)
>>> allocates the kernel thread to the same purpose.
>>> Using workqueues is not very good solution in real-time environment; I
>>> think that allocating and starting the
>>> separate thread will give us more predictable and stable results;
>>>
>>>
>>Where does (A) use a workqueue? (A) doesn't use a workqueue or thread and instead leaves it up to
>>the adapter driver how to handle the messages that it gets sent (which in the case of some drivers
>>will mean no thread or workqueue). (B) is _forcing_ a thread on the adapter which the adapter may
>>not need.
>>
>>
>
>Exactly. That's one of the things I meant when I recently listed some of the
>top goals of the framework I did:
>
>(a) SPI shouldn't perpetuate the driver model botches of I2C;
>(b) ditto I2C's "everything is synchronous" botches;
>(c) it should work well with DMA, to support things like DataFlash;
>(d) given the variety of SPI chips, protocol controls are needed;
>(e) place minimal implementation constraints on controller drivers.
>
>So for example one way you know that (c) is well met is that it's the same
>approach used in USB (both host and peripheral/gadget sides); that's been
>working well for quite a few years now. (Despite comments from Dmitry
>and Vitaly to the contrary.)
>
>
Lemme point you out that if somehting is "working" on a limited number
of platforms within the limited number of use cases, that's not
necessarily a correct implementation.
>
>
>
>>>- (A) has some assumptions on buffers that are passed down to spi
>>> functions.
>>>
>>>
>
>Make that "requirements"; FWIW they're the same ones that apply to all
>other kernel driver frameworks I've seen: that buffers be DMA-safe.
>It would not be helpful (IMO) to define different rules; that's also
>called the "Principle of Least Astonishment". :)
>
>
Yeah within this requirement it's correct. But that requirement may
really make the SPI controller driver a lot more complex if
- it has to send something received from the userland
- it needs to timely send some credentials (what is the case for the
WLAN driver, for instance).
>
>
>
>>> If some controller driver (or bus driver
>>> in terms of (B)) tries to perform DMA transfers, it must copy the
>>> passed buffers to some memory allocated
>>> using GFP_DMA flag and map it using dma_map_single.
>>>
>>>
>
>Based on this and other comments from Dmitry/Vitaly, I suspect they
>don't see how the Linux DMA APIs work. The correct statement is that if
>a controller driver wants to use DMA, it must dma_{map,unmap}_single().
>The upper level drivers don't _need_ to worry about that.
>
>
The upper level drivers do need to worry about their buffers being
DMAable then which requirement adds more complexity.
For instance, that means that the upper level drivers can't use
static/const for the data being sent, what might be pretty much annoying
when you have to send the same data multiple times.
Our approach is definitely more robust, although it may be reasonable to
simplify it somehow.
>>>- (A) retrieves SPI message from the queue in sequential order (FIFO),
>>>
>>>
>
>Only with respect to a given device. It would make no sense to reorder the
>queue so that writing X, then Y, then Z would morph into "X Z Y" or "Z Y X". :)
>
>It's specifically _undefined_ how requests going to different devices are ordered.
>Some hardware will be happier if things are synchronized (e.g. to a vertical
>retrace IRQ), some systems might need to prioritize certain devices, and so on.
>
>I do think FIFO makes a good general policy, for boards without any of those
>special requirements.
>
>
>
>
>>> but (B) provides more flexible way by providing
>>> special callback to retrieve next message from queue. This callback may
>>> implement its own discipline of scheduling
>>> SPI messages. In any way, the default is FIFO.
>>>
>>>
>>I think (A) is missing a method of adding extra spi_message(s) in callback to extend the current
>>transfer on that SPI device. I can imagine a case where you will be required to read status
>>information from a device and in this status information is the length of the data it has just
>>received (for example if it was a network adapter). Straight after reading this information the
>>device would start sending the data it has received but when the read status message was issued
>>the length of the data wasn't known.
>>
>>
>
>Do you actually have hardware which works that way? That would be an example
>of a system that needs some specific prioritization of transfers (see below).
>
>
Let's just agree on the _fact_ that here our approach overcame yours.
IMO it's pretty evident.
>
>
>
>>Currently with (A) we would have to stop the transfer and
>>restart the whole thing again, this time using the length of the data we found form the last
>>message.
>>
>>
>
>Well, each transfer segement would clearly stop, but if that segment had
>the flag set which says "leave chipselect active", then the controller
>driver would have the flexibility to prioritize transfers to that chip.
>
>
Hey, you've told us earlier you don't need any flags, right?
If we start talking about flags necessity, then your *complicated*
approach to the spi_message doesn't make sense, sorry.
>>>- (A) uses standartized way to provide CS information, and (B) relies on
>>> functional drivers callbacks, which looks more
>>> flexible to me.
>>>
>>>
>>I'm not sure what you mean here. You need to provide the cs numbers with SPI device in order for
>>the core to create the unique addres and entry in sysfs.
>>
>>
>
>I'm not sure what he means either. :)
>
>Stephen's PXA2xx SPI driver uses callbacks internally, but that's kind of
>specific to that PXA hardware ... there's no chipselect handled by the
>controller, one of the dozens of GPIOs must be chosen and that's clearly
>a board-specific mechanism (uses controller_data as I recall). He tells
>me he plans to post the latest version of that -- many updates including
>PXA255 SSP support (not just NSSP) and code shrinkage -- early next week.
>
>But most of the SPI controllers I've seen just have a fixed number of
>chipselects, typically four, handled directly by the controller. That's
>why the "standardized way" is just to use a 0..N chipselect number.
>
>
Suppose we have a specific driver or system service (API) implemented to
handle chip selects. We'll have to duplicate its functionality within
your approach what is incorrect in architectural terms. Our approach is
free from this drawback.
Vitaly
--- Vitaly Wool <[email protected]> wrote:
> David Brownell wrote:
>
> >>>- (A) uses more complicated structure of SPI message, that contains one
> >>> or more atomic transfers, and (B)
> >>> offers the only spi_msg that represents the atomic transfer on SPI bus.
> >>> The similar approach can be implemented
> >>> in (B), and actually is implemented. But my imp[ression is that
> >>> such enhancement may be added later..
> >>>
> >>>
> >>I wouldn't have said that the message structure in (A) is more complex then (B). For example,
> in
> >>(B) you have many flags which controls things like SPI mode which are not needed in every
> message.
> >>Once the SPI controller has been setup for a particular slave device you don't need to
> constantly
> >>send this information.
> >>
> >>
> >
> >And in fact, constantly sending it means some drivers will have to waste
> >time constantly checking it, in case it changed. If that setup is stored
> >in controller registers, it's a lot better to just have the setup() call
> >be responsible for changing the communication parameters. (This is the
> >approach used by both MMC and PCMCIA, for what it's worth...)
> >
> >
> >
> I'm not aware if MMC/PCMCIA guys are happy with this approach :), but
> anyway what you're talking about here makes sense.
>
> >
> >
> >>In (B) how to do you handle SPI devices which require to send several messages with out
> releasing
> >>their cs? There are/will be some devices which require this.
> >>
> >>
> >
> >In fact, that's why the transfer segments are grouped. One builds SPI
> >protocol requests out of several such segments. A very common idiom is
> >writing a command, then reading its response. Chipselect must stay
> >active during the whole sequence.
> >
> >Adding support for such a basic mechanism "later" doesn't seem like
> >a good idea to me.
> >
> >
> It is supported by means of flags. I'm afraid your note is pointless here.
How does this work? As I said before I can see that you can leave a cs active, but how can you
make sure the next message is for that SPI device and not for another one?
>
> >
> >
> >
> >>>- (A) uses workqueues to queue and handle SPI messages, and (B)
> >>> allocates the kernel thread to the same purpose.
> >>> Using workqueues is not very good solution in real-time environment; I
> >>> think that allocating and starting the
> >>> separate thread will give us more predictable and stable results;
> >>>
> >>>
> >>Where does (A) use a workqueue? (A) doesn't use a workqueue or thread and instead leaves it up
> to
> >>the adapter driver how to handle the messages that it gets sent (which in the case of some
> drivers
> >>will mean no thread or workqueue). (B) is _forcing_ a thread on the adapter which the adapter
> may
> >>not need.
> >>
> >>
> >
> >Exactly. That's one of the things I meant when I recently listed some of the
> >top goals of the framework I did:
> >
> >(a) SPI shouldn't perpetuate the driver model botches of I2C;
> >(b) ditto I2C's "everything is synchronous" botches;
> >(c) it should work well with DMA, to support things like DataFlash;
> >(d) given the variety of SPI chips, protocol controls are needed;
> >(e) place minimal implementation constraints on controller drivers.
> >
> >So for example one way you know that (c) is well met is that it's the same
> >approach used in USB (both host and peripheral/gadget sides); that's been
> >working well for quite a few years now. (Despite comments from Dmitry
> >and Vitaly to the contrary.)
> >
> >
> Lemme point you out that if somehting is "working" on a limited number
> of platforms within the limited number of use cases, that's not
> necessarily a correct implementation.
No, but it's a good indication :).
>
> >
> >
> >
> >>>- (A) has some assumptions on buffers that are passed down to spi
> >>> functions.
> >>>
> >>>
> >
> >Make that "requirements"; FWIW they're the same ones that apply to all
> >other kernel driver frameworks I've seen: that buffers be DMA-safe.
> >It would not be helpful (IMO) to define different rules; that's also
> >called the "Principle of Least Astonishment". :)
> >
> >
> Yeah within this requirement it's correct. But that requirement may
> really make the SPI controller driver a lot more complex if
> - it has to send something received from the userland
> - it needs to timely send some credentials (what is the case for the
> WLAN driver, for instance).
>
> >
> >
> >
> >>> If some controller driver (or bus driver
> >>> in terms of (B)) tries to perform DMA transfers, it must copy the
> >>> passed buffers to some memory allocated
> >>> using GFP_DMA flag and map it using dma_map_single.
> >>>
> >>>
> >
> >Based on this and other comments from Dmitry/Vitaly, I suspect they
> >don't see how the Linux DMA APIs work. The correct statement is that if
> >a controller driver wants to use DMA, it must dma_{map,unmap}_single().
> >The upper level drivers don't _need_ to worry about that.
> >
> >
> The upper level drivers do need to worry about their buffers being
> DMAable then which requirement adds more complexity.
> For instance, that means that the upper level drivers can't use
> static/const for the data being sent, what might be pretty much annoying
> when you have to send the same data multiple times.
> Our approach is definitely more robust, although it may be reasonable to
> simplify it somehow.
>
> >>>- (A) retrieves SPI message from the queue in sequential order (FIFO),
> >>>
> >>>
> >
> >Only with respect to a given device. It would make no sense to reorder the
> >queue so that writing X, then Y, then Z would morph into "X Z Y" or "Z Y X". :)
> >
> >It's specifically _undefined_ how requests going to different devices are ordered.
> >Some hardware will be happier if things are synchronized (e.g. to a vertical
> >retrace IRQ), some systems might need to prioritize certain devices, and so on.
> >
> >I do think FIFO makes a good general policy, for boards without any of those
> >special requirements.
> >
> >
> >
> >
> >>> but (B) provides more flexible way by providing
> >>> special callback to retrieve next message from queue. This callback may
> >>> implement its own discipline of scheduling
> >>> SPI messages. In any way, the default is FIFO.
> >>>
> >>>
> >>I think (A) is missing a method of adding extra spi_message(s) in callback to extend the
> current
> >>transfer on that SPI device. I can imagine a case where you will be required to read status
> >>information from a device and in this status information is the length of the data it has just
> >>received (for example if it was a network adapter). Straight after reading this information
> the
> >>device would start sending the data it has received but when the read status message was
> issued
> >>the length of the data wasn't known.
> >>
> >>
> >
> >Do you actually have hardware which works that way? That would be an example
> >of a system that needs some specific prioritization of transfers (see below).
> >
> >
> Let's just agree on the _fact_ that here our approach overcame yours.
> IMO it's pretty evident.
Neither proposal is complete.
>
> >
> >
> >
> >>Currently with (A) we would have to stop the transfer and
> >>restart the whole thing again, this time using the length of the data we found form the last
> >>message.
> >>
> >>
> >
> >Well, each transfer segement would clearly stop, but if that segment had
> >the flag set which says "leave chipselect active", then the controller
> >driver would have the flexibility to prioritize transfers to that chip.
> >
> >
> Hey, you've told us earlier you don't need any flags, right?
> If we start talking about flags necessity, then your *complicated*
> approach to the spi_message doesn't make sense, sorry.
You don't need _as many_ flags as in (B)
>
> >>>- (A) uses standartized way to provide CS information, and (B) relies on
> >>> functional drivers callbacks, which looks more
> >>> flexible to me.
> >>>
> >>>
> >>I'm not sure what you mean here. You need to provide the cs numbers with SPI device in order
> for
> >>the core to create the unique addres and entry in sysfs.
> >>
> >>
> >
> >I'm not sure what he means either. :)
> >
> >Stephen's PXA2xx SPI driver uses callbacks internally, but that's kind of
> >specific to that PXA hardware ... there's no chipselect handled by the
> >controller, one of the dozens of GPIOs must be chosen and that's clearly
> >a board-specific mechanism (uses controller_data as I recall). He tells
> >me he plans to post the latest version of that -- many updates including
> >PXA255 SSP support (not just NSSP) and code shrinkage -- early next week.
> >
> >But most of the SPI controllers I've seen just have a fixed number of
> >chipselects, typically four, handled directly by the controller. That's
> >why the "standardized way" is just to use a 0..N chipselect number.
> >
> >
> Suppose we have a specific driver or system service (API) implemented to
> handle chip selects. We'll have to duplicate its functionality within
> your approach what is incorrect in architectural terms. Our approach is
> free from this drawback.
But suppose you have a SPI controller which has cs lines in the IP block. Are you saying you will
write another driver which manipulates the registers in the SPI controller out side of the adapter
driver? Some hardware might not support that (i.e. you can't directly controll the cs pin).
Every solution will always work better on some hardware then other hardware. The point is to make
the solution work on all types hardware, here I think David's solution is better.
Mark
>
> Vitaly
>
>
___________________________________________________________
Does your mail provider give you FREE antivirus protection?
Get Yahoo! Mail http://uk.mail.yahoo.com