This driver adds generic removable drive bay support.
Signed-off-by: Kristen Carlson Accardi <[email protected]>
Index: 2.6-git/drivers/acpi/bay.c
===================================================================
--- /dev/null
+++ 2.6-git/drivers/acpi/bay.c
@@ -0,0 +1,592 @@
+/*
+ * bay.c - ACPI removable drive bay driver
+ *
+ * Copyright (C) 2006 Kristen Carlson Accardi <[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.,
+ * 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA.
+ *
+ * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+ */
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/types.h>
+#include <linux/notifier.h>
+#include <acpi/acpi_bus.h>
+#include <acpi/acpi_drivers.h>
+#include <linux/proc_fs.h>
+#include <linux/seq_file.h>
+#include <asm/uaccess.h>
+
+#define ACPI_BAY_DRIVER_NAME "ACPI Removable Drive Bay Driver"
+
+ACPI_MODULE_NAME("bay")
+MODULE_AUTHOR("Kristen Carlson Accardi");
+MODULE_DESCRIPTION(ACPI_BAY_DRIVER_NAME);
+MODULE_LICENSE("GPL");
+#define ACPI_BAY_CLASS "bay"
+#define ACPI_BAY_COMPONENT 0x10000000
+#define _COMPONENT ACPI_BAY_COMPONENT
+#define bay_dprintk(h,s) {\
+ char prefix[80] = {'\0'};\
+ struct acpi_buffer buffer = {sizeof(prefix), prefix};\
+ acpi_get_name(h, ACPI_FULL_PATHNAME, &buffer);\
+ printk(KERN_DEBUG PREFIX "%s: %s\n", prefix, s); }
+
+static void bay_notify(acpi_handle handle, u32 event, void *data);
+static int acpi_bay_add(struct acpi_device *device);
+static int acpi_bay_remove(struct acpi_device *device, int type);
+static int acpi_bay_match(struct acpi_device *device,
+ struct acpi_driver *driver);
+
+static struct acpi_driver acpi_bay_driver = {
+ .name = ACPI_BAY_DRIVER_NAME,
+ .class = ACPI_BAY_CLASS,
+ .ops = {
+ .add = acpi_bay_add,
+ .remove = acpi_bay_remove,
+ .match = acpi_bay_match,
+ },
+};
+
+struct bay {
+ acpi_handle handle;
+ char *name;
+ struct list_head list;
+ struct proc_dir_entry *proc;
+};
+
+LIST_HEAD(drive_bays);
+
+static struct proc_dir_entry *acpi_bay_dir;
+
+/*****************************************************************************
+ * Drive Bay functions *
+ *****************************************************************************/
+/**
+ * is_ejectable - see if a device is ejectable
+ * @handle: acpi handle of the device
+ *
+ * If an acpi object has a _EJ0 method, then it is ejectable
+ */
+static int is_ejectable(acpi_handle handle)
+{
+ acpi_status status;
+ acpi_handle tmp;
+
+ status = acpi_get_handle(handle, "_EJ0", &tmp);
+ if (ACPI_FAILURE(status))
+ return 0;
+ return 1;
+}
+
+/**
+ * bay_present - see if the bay device is present
+ * @bay: the drive bay
+ *
+ * execute the _STA method.
+ */
+static int bay_present(struct bay *bay)
+{
+ unsigned long sta;
+ acpi_status status;
+
+ if (bay) {
+ status = acpi_evaluate_integer(bay->handle, "_STA", NULL, &sta);
+ if (ACPI_SUCCESS(status) && sta)
+ return 1;
+ }
+ return 0;
+}
+
+/**
+ * eject_device - respond to an eject request
+ * @handle - the device to eject
+ *
+ * Call this devices _EJ0 method.
+ */
+static void eject_device(acpi_handle handle)
+{
+ struct acpi_object_list arg_list;
+ union acpi_object arg;
+
+ bay_dprintk(handle, "Ejecting device");
+
+ arg_list.count = 1;
+ arg_list.pointer = &arg;
+ arg.type = ACPI_TYPE_INTEGER;
+ arg.integer.value = 1;
+
+ if (ACPI_FAILURE(acpi_evaluate_object(handle, "_EJ0",
+ &arg_list, NULL)))
+ pr_debug("Failed to evaluate _EJ0!\n");
+}
+
+
+/**
+ * is_ata - see if a device is an ata device
+ * @handle: acpi handle of the device
+ *
+ * If an acpi object has one of 4 ATA ACPI methods defined,
+ * then it is an ATA device
+ */
+static int is_ata(acpi_handle handle)
+{
+ acpi_handle tmp;
+
+ if ((ACPI_SUCCESS(acpi_get_handle(handle, "_GTF", &tmp))) ||
+ (ACPI_SUCCESS(acpi_get_handle(handle, "_GTM", &tmp))) ||
+ (ACPI_SUCCESS(acpi_get_handle(handle, "_STM", &tmp))) ||
+ (ACPI_SUCCESS(acpi_get_handle(handle, "_SDD", &tmp))))
+ return 1;
+
+ return 0;
+}
+
+/**
+ * parent_is_ata(acpi_handle handle)
+ *
+ */
+static int parent_is_ata(acpi_handle handle)
+{
+ acpi_handle phandle;
+
+ if (acpi_get_parent(handle, &phandle))
+ return 0;
+
+ return is_ata(phandle);
+}
+
+/**
+ * is_ejectable_bay - see if a device is an ejectable drive bay
+ * @handle: acpi handle of the device
+ *
+ * If an acpi object is ejectable and has one of the ACPI ATA
+ * methods defined, then we can safely call it an ejectable
+ * drive bay
+ */
+static int is_ejectable_bay(acpi_handle handle)
+{
+ if ((is_ata(handle) || parent_is_ata(handle)) && is_ejectable(handle))
+ return 1;
+ return 0;
+}
+
+/**
+ * eject_removable_drive - try to eject this drive
+ * @dev : the device structure of the drive
+ *
+ * If a device is a removable drive that requires an _EJ0 method
+ * to be executed in order to safely remove from the system, do
+ * it. ATM - always returns success
+ */
+int eject_removable_drive(struct device *dev)
+{
+ acpi_handle handle = DEVICE_ACPI_HANDLE(dev);
+
+ if (handle) {
+ bay_dprintk(handle, "Got device handle");
+ if (is_ejectable_bay(handle))
+ eject_device(handle);
+ } else {
+ printk("No acpi handle for device\n");
+ }
+
+ /* should I return an error code? */
+ return 0;
+}
+EXPORT_SYMBOL_GPL(eject_removable_drive);
+
+static int acpi_bay_add(struct acpi_device *device)
+{
+ bay_dprintk(device->handle, "adding bay device");
+ strcpy(acpi_device_name(device), "Dockable Bay");
+ strcpy(acpi_device_class(device), "bay");
+ return 0;
+}
+
+static int acpi_bay_status_seq_show(struct seq_file *seq, void *offset)
+{
+ struct bay *bay = (struct bay *)seq->private;
+
+ if (!bay)
+ return 0;
+
+ if (bay_present(bay))
+ seq_printf(seq, "present\n");
+ else
+ seq_printf(seq, "removed\n");
+
+ return 0;
+}
+
+static ssize_t
+acpi_bay_write_eject(struct file *file,
+ const char __user * buffer,
+ size_t count, loff_t * data)
+{
+ struct seq_file *m = (struct seq_file *)file->private_data;
+ struct bay *bay = (struct bay *)m->private;
+ char str[12] = { 0 };
+ u32 state = 0;
+
+ /* FIXME - our only valid value here is 1 */
+ if (!bay || count + 1 > sizeof str)
+ return -EINVAL;
+
+ if (copy_from_user(str, buffer, count))
+ return -EFAULT;
+
+ str[count] = 0;
+ state = simple_strtoul(str, NULL, 0);
+ if (state)
+ eject_device(bay->handle);
+
+ return count;
+}
+
+static int
+acpi_bay_status_open_fs(struct inode *inode, struct file *file)
+{
+ return single_open(file, acpi_bay_status_seq_show,
+ PDE(inode)->data);
+}
+
+static int
+acpi_bay_eject_open_fs(struct inode *inode, struct file *file)
+{
+ return single_open(file, acpi_bay_status_seq_show,
+ PDE(inode)->data);
+}
+
+static struct file_operations acpi_bay_status_fops = {
+ .open = acpi_bay_status_open_fs,
+ .read = seq_read,
+ .llseek = seq_lseek,
+ .release = single_release,
+};
+
+static struct file_operations acpi_bay_eject_fops = {
+ .open = acpi_bay_eject_open_fs,
+ .read = seq_read,
+ .write = acpi_bay_write_eject,
+ .llseek = seq_lseek,
+ .release = single_release,
+};
+#if 0
+static struct file_operations acpi_bay_insert_fops = {
+ .open = acpi_bay_insert_open_fs,
+ .read = seq_read,
+ .llseek = seq_lseek,
+ .release = single_release,
+};
+#endif
+static int acpi_bay_add_fs(struct bay *bay)
+{
+ struct proc_dir_entry *entry = NULL;
+
+ if (!bay)
+ return -EINVAL;
+
+ /*
+ * create a proc entry for this device
+ * we need to do this a little bit differently than normal
+ * acpi device drivers because our device may not be present
+ * at the moment, and therefore we have no acpi_device struct
+ */
+
+ bay->proc = proc_mkdir(bay->name, acpi_bay_dir);
+
+ /* 'status' [R] */
+ entry = create_proc_entry("status",
+ S_IRUGO, bay->proc);
+ if (!entry)
+ return -EIO;
+ else {
+ entry->proc_fops = &acpi_bay_status_fops;
+ entry->data = bay;
+ entry->owner = THIS_MODULE;
+ }
+ /* 'eject' [W] */
+ entry = create_proc_entry("eject",
+ S_IWUGO, bay->proc);
+ if (!entry)
+ return -EIO;
+ else {
+ entry->proc_fops = &acpi_bay_eject_fops;
+ entry->data = bay;
+ entry->owner = THIS_MODULE;
+ }
+#if 0
+ /* 'insert' [W] */
+ entry = create_proc_entry("insert",
+ S_IWUGO, bay->proc);
+ if (!entry)
+ return -EIO;
+ else {
+ entry->proc_fops = &acpi_bay_insert_fops;
+ entry->data = bay;
+ entry->owner = THIS_MODULE;
+ }
+#endif
+ return 0;
+}
+
+static void acpi_bay_remove_fs(struct bay *bay)
+{
+ if (!bay)
+ return;
+
+ if (bay->proc) {
+ remove_proc_entry("status", bay->proc);
+ remove_proc_entry("eject", bay->proc);
+#if 0
+ remove_proc_entry("insert", bay->proc);
+#endif
+ remove_proc_entry(bay->name, acpi_bay_dir);
+ bay->proc = NULL;
+ }
+}
+
+static int bay_is_dock_device(acpi_handle handle)
+{
+ acpi_handle parent;
+
+ acpi_get_parent(handle, &parent);
+
+ /* if the device or it's parent is dependent on the
+ * dock, then we are a dock device
+ */
+ return (is_dock_device(handle) || is_dock_device(parent));
+}
+
+static int bay_add(acpi_handle handle)
+{
+ acpi_status status;
+ struct bay *new_bay;
+ struct acpi_buffer nbuffer = {ACPI_ALLOCATE_BUFFER, NULL};
+ acpi_get_name(handle, ACPI_FULL_PATHNAME, &nbuffer);
+
+ bay_dprintk(handle, "Adding notify handler");
+
+ /*
+ * if this is the first bay device found, make the root
+ * proc entry
+ */
+ if (acpi_bay_dir == NULL)
+ acpi_bay_dir = proc_mkdir(ACPI_BAY_CLASS, acpi_root_dir);
+
+ /*
+ * Initialize bay device structure
+ */
+ new_bay = kmalloc(GFP_ATOMIC, sizeof(*new_bay));
+ INIT_LIST_HEAD(&new_bay->list);
+ new_bay->handle = handle;
+ new_bay->name = (char *)nbuffer.pointer;
+ list_add(&new_bay->list, &drive_bays);
+ acpi_bay_add_fs(new_bay);
+
+ /* register for events on this device */
+ status = acpi_install_notify_handler(handle, ACPI_SYSTEM_NOTIFY,
+ bay_notify, new_bay);
+ if (ACPI_FAILURE(status)) {
+ printk(KERN_ERR PREFIX "Error installing bay notify handler\n");
+ }
+
+ /* if we are on a dock station, we should register for dock
+ * notifications.
+ */
+ if (bay_is_dock_device(handle)) {
+ bay_dprintk(handle, "Is dependent on dock\n");
+ register_hotplug_dock_device(handle, bay_notify, new_bay);
+ }
+ printk(KERN_INFO PREFIX "Bay [%s] Added\n", new_bay->name);
+ return 0;
+}
+
+static int acpi_bay_remove(struct acpi_device *device, int type)
+{
+ /*** FIXME: do something here */
+ return 0;
+}
+
+static int acpi_bay_match(struct acpi_device *device,
+ struct acpi_driver *driver)
+{
+ if (!device || !driver)
+ return -EINVAL;
+
+ if (is_ejectable_bay(device->handle)) {
+ bay_dprintk(device->handle, "matching bay device");
+ return 0;
+ }
+
+ return -ENODEV;
+}
+
+/**
+ * bay_create_acpi_device - add new devices to acpi
+ * @handle - handle of the device to add
+ *
+ * This function will create a new acpi_device for the given
+ * handle if one does not exist already. This should cause
+ * acpi to scan for drivers for the given devices, and call
+ * matching driver's add routine.
+ *
+ * Returns a pointer to the acpi_device corresponding to the handle.
+ */
+static struct acpi_device * bay_create_acpi_device(acpi_handle handle)
+{
+ struct acpi_device *device = NULL;
+ struct acpi_device *parent_device;
+ acpi_handle parent;
+ int ret;
+
+ bay_dprintk(handle, "Trying to get device");
+ if (acpi_bus_get_device(handle, &device)) {
+ /*
+ * no device created for this object,
+ * so we should create one.
+ */
+ bay_dprintk(handle, "No device for handle");
+ acpi_get_parent(handle, &parent);
+ if (acpi_bus_get_device(parent, &parent_device))
+ parent_device = NULL;
+
+ ret = acpi_bus_add(&device, parent_device, handle,
+ ACPI_BUS_TYPE_DEVICE);
+ if (ret) {
+ pr_debug("error adding bus, %x\n",
+ -ret);
+ return NULL;
+ }
+ }
+ return device;
+}
+
+/**
+ * bay_notify - act upon an acpi bay notification
+ * @handle: the bay handle
+ * @event: the acpi event
+ * @data: our driver data struct
+ *
+ */
+static void bay_notify(acpi_handle handle, u32 event, void *data)
+{
+ struct acpi_device *dev;
+ struct bay *bay = data;
+
+ bay_dprintk(handle, "Bay event");
+
+ switch(event) {
+ case ACPI_NOTIFY_BUS_CHECK:
+ printk("Bus Check\n");
+ case ACPI_NOTIFY_DEVICE_CHECK:
+ printk("Device Check\n");
+ dev = bay_create_acpi_device(handle);
+ if (dev)
+ acpi_bus_generate_event(dev, event, 0);
+ else
+ printk("No device for generating event\n");
+ /* wouldn't it be a good idea to just rescan SATA
+ * right here?
+ */
+ break;
+ case ACPI_NOTIFY_EJECT_REQUEST:
+ printk("Eject request\n");
+ dev = bay_create_acpi_device(handle);
+ if (dev)
+ acpi_bus_generate_event(dev, event, 0);
+ else
+ printk("No device for generating eventn");
+
+ /* wouldn't it be a good idea to just call the
+ * eject_device here if we were a SATA device?
+ */
+ break;
+ default:
+ printk("unknown event %d\n", event);
+ }
+}
+
+static acpi_status
+find_bay(acpi_handle handle, u32 lvl, void *context, void **rv)
+{
+ int *count = (int *)context;
+
+ /*
+ * there could be more than one ejectable bay.
+ * so, just return AE_OK always so that every object
+ * will be checked.
+ */
+ if (is_ejectable_bay(handle)) {
+ bay_dprintk(handle, "found ejectable bay");
+ bay_add(handle);
+ (*count)++;
+ }
+ return AE_OK;
+}
+
+static int __init bay_init(void)
+{
+ int bays = 0;
+
+ acpi_bay_dir = NULL;
+ INIT_LIST_HEAD(&drive_bays);
+
+ /* look for dockable drive bays */
+ acpi_walk_namespace(ACPI_TYPE_DEVICE, ACPI_ROOT_OBJECT,
+ ACPI_UINT32_MAX, find_bay, &bays, NULL);
+
+ if (bays) {
+ if ((acpi_bus_register_driver(&acpi_bay_driver) < 0)) {
+ printk(KERN_ERR "Unable to register bay driver\n");
+ if (acpi_bay_dir)
+ remove_proc_entry(ACPI_BAY_CLASS,
+ acpi_root_dir);
+ }
+ }
+
+ if (!bays)
+ return -ENODEV;
+
+ return 0;
+}
+
+static void __exit bay_exit(void)
+{
+ struct bay *bay, *tmp;
+
+ list_for_each_entry_safe(bay, tmp, &drive_bays, list) {
+ if (is_dock_device(bay->handle))
+ unregister_hotplug_dock_device(bay->handle);
+ acpi_bay_remove_fs(bay);
+ acpi_remove_notify_handler(bay->handle, ACPI_SYSTEM_NOTIFY,
+ bay_notify);
+ kfree(bay->name);
+ kfree(bay);
+ }
+
+ if (acpi_bay_dir)
+ remove_proc_entry(ACPI_BAY_CLASS, acpi_root_dir);
+
+ acpi_bus_unregister_driver(&acpi_bay_driver);
+}
+
+postcore_initcall(bay_init);
+module_exit(bay_exit);
+
Index: 2.6-git/drivers/acpi/Kconfig
===================================================================
--- 2.6-git.orig/drivers/acpi/Kconfig
+++ 2.6-git/drivers/acpi/Kconfig
@@ -138,6 +138,13 @@ config ACPI_DOCK
help
This driver adds support for ACPI controlled docking stations
+config ACPI_BAY
+ tristate "Removable Drive Bay"
+ depends on EXPERIMENTAL
+ help
+ This driver adds support for ACPI controlled removable drive
+ bays such as the IBM ultrabay or the Dell Module Bay.
+
config ACPI_PROCESSOR
tristate "Processor"
default y
Index: 2.6-git/drivers/acpi/Makefile
===================================================================
--- 2.6-git.orig/drivers/acpi/Makefile
+++ 2.6-git/drivers/acpi/Makefile
@@ -43,6 +43,7 @@ obj-$(CONFIG_ACPI_BUTTON) += button.o
obj-$(CONFIG_ACPI_EC) += ec.o
obj-$(CONFIG_ACPI_FAN) += fan.o
obj-$(CONFIG_ACPI_DOCK) += dock.o
+obj-$(CONFIG_ACPI_BAY) += bay.o
obj-$(CONFIG_ACPI_VIDEO) += video.o
obj-$(CONFIG_ACPI_HOTKEY) += hotkey.o
obj-y += pci_root.o pci_link.o pci_irq.o pci_bind.o
-
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/
[Index of Archives]
[Kernel Newbies]
[Netfilter]
[Bugtraq]
[Photo]
[Stuff]
[Gimp]
[Yosemite News]
[MIPS Linux]
[ARM Linux]
[Linux Security]
[Linux RAID]
[Video 4 Linux]
[Linux for the blind]
[Linux Resources]