1struct file_operations { +2 struct module *owner; +3 loff_t (*llseek) (struct file *, loff_t, int); +4 ssize_t (*read) (struct file *, char __user *, size_t, loff_t *); +5 ssize_t (*write) (struct file *, const char __user *, size_t, loff_t *); +6 ssize_t (*read_iter) (struct kiocb *, struct iov_iter *); +7 ssize_t (*write_iter) (struct kiocb *, struct iov_iter *); +8 int (*iopoll)(struct kiocb *kiocb, bool spin); +9 int (*iterate) (struct file *, struct dir_context *); +10 int (*iterate_shared) (struct file *, struct dir_context *); +11 __poll_t (*poll) (struct file *, struct poll_table_struct *); +12 long (*unlocked_ioctl) (struct file *, unsigned int, unsigned long); +13 long (*compat_ioctl) (struct file *, unsigned int, unsigned long); +14 int (*mmap) (struct file *, struct vm_area_struct *); +15 unsigned long mmap_supported_flags; +16 int (*open) (struct inode *, struct file *); +17 int (*flush) (struct file *, fl_owner_t id); +18 int (*release) (struct inode *, struct file *); +19 int (*fsync) (struct file *, loff_t, loff_t, int datasync); +20 int (*fasync) (int, struct file *, int); +21 int (*lock) (struct file *, int, struct file_lock *); +22 ssize_t (*sendpage) (struct file *, struct page *, int, size_t, loff_t *, int); +23 unsigned long (*get_unmapped_area)(struct file *, unsigned long, unsigned long, unsigned long, unsigned long); +24 int (*check_flags)(int); +25 int (*flock) (struct file *, int, struct file_lock *); +26 ssize_t (*splice_write)(struct pipe_inode_info *, struct file *, loff_t *, size_t, unsigned int); +27 ssize_t (*splice_read)(struct file *, loff_t *, struct pipe_inode_info *, size_t, unsigned int); +28 int (*setlease)(struct file *, long, struct file_lock **, void **); +29 long (*fallocate)(struct file *file, int mode, loff_t offset, 30 loff_t len); -31 void (*show_fdinfo)(struct seq_file *m, struct file *f); -32 ssize_t (*copy_file_range)(struct file *, loff_t, struct file *, -33 loff_t, size_t, unsigned int); -34 loff_t (*remap_file_range)(struct file *file_in, loff_t pos_in, -35 struct file *file_out, loff_t pos_out, -36 loff_t len, unsigned int remap_flags); -37 int (*fadvise)(struct file *, loff_t, loff_t, int); +31 void (*show_fdinfo)(struct seq_file *m, struct file *f); +32 ssize_t (*copy_file_range)(struct file *, loff_t, struct file *, +33 loff_t, size_t, unsigned int); +34 loff_t (*remap_file_range)(struct file *file_in, loff_t pos_in, +35 struct file *file_out, loff_t pos_out, +36 loff_t len, unsigned int remap_flags); +37 int (*fadvise)(struct file *, loff_t, loff_t, int); 38} __randomize_layout;
Some operations are not implemented by a driver. For example, a driver that handles a video card will not need to read from a directory structure. The corresponding entries @@ -1225,7 +1222,7 @@ new way of assigning to the structure looks like:
-1struct file_operations fops = { +1struct file_operations fops = { 2 read: device_read, 3 write: device_write, 4 open: device_open, @@ -1237,7 +1234,7 @@ You should use this syntax in case someone wants to port your driver. It will he with compatibility:-
1struct file_operations fops = { +1struct file_operations fops = { 2 .read = device_read, 3 .write = device_write, 4 .open = device_open, @@ -1246,7 +1243,7 @@ with compatibility:-The meaning is clear, and you should be aware that any member of the structure which you do not explicitly assign will be initialized to NULL by gcc. -
An instance of
struct file_operations +An instance of
struct file_operationscontaining pointers to functions that are used to implement read, write, open, … syscalls is commonly named fops.Sin Linux v5.6, the
proc_ops @@ -1259,7 +1256,7 @@ syscalls is commonly named fops.Each device is represented in the kernel by a file structure, which is defined in linux/fs.h. Be aware that a file is a kernel level structure and never appears in a user space program. It is not the same thing as a -
FILE +FILE, which is defined by glibc and would never appear in a kernel space function. Also, its name is a bit misleading; it represents an abstract open ‘file’, not a file on a disk, which is represented by a structure named inode. @@ -1288,24 +1285,24 @@ using theregister_chrdevfunction, defined by linux/fs.h.-
1int register_chrdev(unsigned int major, const char *name, struct file_operations *fops);+1int register_chrdev(unsigned int major, const char *name, struct file_operations *fops);where unsigned int major is the major number you want to request, -
const char *name +const char *nameis the name of the device as it will appear in /proc/devices and -struct file_operations *fops +struct file_operations *fopsis a pointer to thefile_operationstable for your driver. A negative return value means the registration failed. Note that we didn’t pass the minor number toregister_chrdev. That is because the kernel doesn’t care about the minor number; only our driver uses it. -Now the question is, how do you get a major number without hijacking one that’s -already in use? The easiest way would be to look through Documentation -/devices.txt and pick an unused one. That is a bad way of doing things -because you will never be sure if the number you picked will be assigned later. -The answer is that you can ask the kernel to assign you a dynamic major -number. -
If you pass a major number of 0 to
register_chrdev +Now the question is, how do you get a major number without hijacking +one that’s already in use? The easiest way would be to look through +Documentation/admin-guide/devices.txt and pick an unused one. That is a bad +way of doing things because you will never be sure if the number you picked will be +assigned later. The answer is that you can ask the kernel to assign you a dynamic +major number. +
If you pass a major number of 0 to
register_chrdev, the return value will be the dynamically allocated major number. The downside is that you can not make a device file in advance, since you do not know what the major number will be. There are a couple of ways to do @@ -1319,13 +1316,13 @@ third method is we can have our driver make the the device file using thedevice_destroyduring the call tocleanup_module. -+
0.6.4 Unregistering A Device
-We can not allow the kernel module to be rmmod’ed whenever root feels like it. If the +
We can not allow the kernel module to be rmmod’ed whenever root feels like it. If the device file is opened by a process and then we remove the kernel module, using the file would cause a call to the memory location where the appropriate function (read/write) used to be. If we are lucky, no other code was loaded there, and we’ll get @@ -1333,7 +1330,7 @@ an ugly error message. If we are unlucky, another kernel module was loaded into same location, which means a jump into the middle of another function within the kernel. The results of this would be impossible to predict, but they can not be very positive. -
Normally, when you do not want to allow something, you return an error code +
Normally, when you do not want to allow something, you return an error code (a negative number) from the function which is supposed to do it. With
cleanup_modulethat’s impossible because it is a void function. However, there is a counter which @@ -1353,212 +1350,212 @@ this counter:- module_put(THIS_MODULE): Decrement the use count.It is important to keep the counter accurate; if you ever do lose track of the +
It is important to keep the counter accurate; if you ever do lose track of the correct usage count, you will never be able to unload the module; it’s now reboot time, boys and girls. This is bound to happen to you sooner or later during a module’s development. -
+
0.6.5 chardev.c
-The next code sample creates a char driver named chardev. You can cat its device +
The next code sample creates a char driver named chardev. You can cat its device file.
1cat /proc/devices-(or open the file with a program) and the driver will put the number of times the +
(or open the file with a program) and the driver will put the number of times the device file has been read from into the file. We do not support writing to the file (like -
echo "hi" > /dev/hello +echo "hi" > /dev/hello), but catch these attempts and tell the user that the operation is not supported. Don’t worry if you don’t see what we do with the data we read into the buffer; we don’t do much with it. We simply read in the data and print a message acknowledging that we received it.-
1/* -2 * chardev.c: Creates a read-only char device that says how many times -3 * you have read from the dev file -4 */ ++1/* +2 * chardev.c: Creates a read-only char device that says how many times +3 * you have read from the dev file +4 */ 5 -6#include <linux/cdev.h> -7#include <linux/delay.h> -8#include <linux/device.h> -9#include <linux/fs.h> -10#include <linux/init.h> -11#include <linux/irq.h> -12#include <linux/kernel.h> -13#include <linux/module.h> -14#include <linux/poll.h> +6#include <linux/cdev.h> +7#include <linux/delay.h> +8#include <linux/device.h> +9#include <linux/fs.h> +10#include <linux/init.h> +11#include <linux/irq.h> +12#include <linux/kernel.h> +13#include <linux/module.h> +14#include <linux/poll.h> 15 -16/* Prototypes - this would normally go in a .h file */ -17static int device_open(struct inode *, struct file *); -18static int device_release(struct inode *, struct file *); -19static ssize_t device_read(struct file *, char *, size_t, loff_t *); -20static ssize_t device_write(struct file *, const char *, size_t, loff_t *); +16/* Prototypes - this would normally go in a .h file */ +17static int device_open(struct inode *, struct file *); +18static int device_release(struct inode *, struct file *); +19static ssize_t device_read(struct file *, char *, size_t, loff_t *); +20static ssize_t device_write(struct file *, const char *, size_t, loff_t *); 21 -22#define SUCCESS 0 -23#define DEVICE_NAME "chardev" /* Dev name as it appears in /proc/devices */ -24#define BUF_LEN 80 /* Max length of the message from the device */ +22#define SUCCESS 0 +23#define DEVICE_NAME "chardev" /* Dev name as it appears in /proc/devices */ +24#define BUF_LEN 80 /* Max length of the message from the device */ 25 -26/* Global variables are declared as static, so are global within the file. */ +26/* Global variables are declared as static, so are global within the file. */ 27 -28static int major; /* major number assigned to our device driver */ -29static int open_device_cnt = 0; /* Is device open? -30 * Used to prevent multiple access to device */ -31static char msg[BUF_LEN]; /* The msg the device will give when asked */ -32static char *msg_ptr; +28static int major; /* major number assigned to our device driver */ +29static int open_device_cnt = 0; /* Is device open? +30 * Used to prevent multiple access to device */ +31static char msg[BUF_LEN]; /* The msg the device will give when asked */ +32static char *msg_ptr; 33 -34static struct class *cls; +34static struct class *cls; 35 -36static struct file_operations chardev_fops = { +36static struct file_operations chardev_fops = { 37 .read = device_read, 38 .write = device_write, 39 .open = device_open, 40 .release = device_release, 41}; 42 -43static int __init chardev_init(void) +43static int __init chardev_init(void) 44{ 45 major = register_chrdev(0, DEVICE_NAME, &chardev_fops); 46 -47 if (major < 0) { -48 pr_alert("Registering char device failed with %d\n", major); -49 return major; +47 if (major < 0) { +48 pr_alert("Registering char device failed with %d\n", major); +49 return major; 50 } 51 -52 pr_info("I was assigned major number %d.\n", major); +52 pr_info("I was assigned major number %d.\n", major); 53 54 cls = class_create(THIS_MODULE, DEVICE_NAME); 55 device_create(cls, NULL, MKDEV(major, 0), NULL, DEVICE_NAME); 56 -57 pr_info("Device created on /dev/%s\n", DEVICE_NAME); +57 pr_info("Device created on /dev/%s\n", DEVICE_NAME); 58 -59 return SUCCESS; +59 return SUCCESS; 60} 61 -62static void __exit chardev_exit(void) +62static void __exit chardev_exit(void) 63{ 64 device_destroy(cls, MKDEV(major, 0)); 65 class_destroy(cls); 66 -67 /* Unregister the device */ +67 /* Unregister the device */ 68 unregister_chrdev(major, DEVICE_NAME); 69} 70 -71/* Methods */ +71/* Methods */ 72 -73/* Called when a process tries to open the device file, like -74 * "sudo cat /dev/chardev" -75 */ -76static int device_open(struct inode *inode, struct file *file) +73/* Called when a process tries to open the device file, like +74 * "sudo cat /dev/chardev" +75 */ +76static int device_open(struct inode *inode, struct file *file) 77{ -78 static int counter = 0; +78 static int counter = 0; 79 -80 if (open_device_cnt) -81 return -EBUSY; +80 if (open_device_cnt) +81 return -EBUSY; 82 83 open_device_cnt++; -84 sprintf(msg, "I already told you %d times Hello world!\n", counter++); +84 sprintf(msg, "I already told you %d times Hello world!\n", counter++); 85 msg_ptr = msg; 86 try_module_get(THIS_MODULE); 87 -88 return SUCCESS; +88 return SUCCESS; 89} 90 -91/* Called when a process closes the device file. */ -92static int device_release(struct inode *inode, struct file *file) +91/* Called when a process closes the device file. */ +92static int device_release(struct inode *inode, struct file *file) 93{ -94 open_device_cnt--; /* We're now ready for our next caller */ +94 open_device_cnt--; /* We're now ready for our next caller */ 95 -96 /* Decrement the usage count, or else once you opened the file, you will -97 * never get get rid of the module. -98 */ +96 /* Decrement the usage count, or else once you opened the file, you will +97 * never get get rid of the module. +98 */ 99 module_put(THIS_MODULE); 100 -101 return SUCCESS; +101 return SUCCESS; 102} 103 -104/* Called when a process, which already opened the dev file, attempts to -105 * read from it. -106 */ -107static ssize_t device_read(struct file *filp, /* see include/linux/fs.h */ -108 char *buffer, /* buffer to fill with data */ -109 size_t length, /* length of the buffer */ +104/* Called when a process, which already opened the dev file, attempts to +105 * read from it. +106 */ +107static ssize_t device_read(struct file *filp, /* see include/linux/fs.h */ +108 char *buffer, /* buffer to fill with data */ +109 size_t length, /* length of the buffer */ 110 loff_t *offset) 111{ -112 /* Number of bytes actually written to the buffer */ -113 int bytes_read = 0; +112 /* Number of bytes actually written to the buffer */ +113 int bytes_read = 0; 114 -115 /* If we are at the end of message, return 0 signifying end of file. */ -116 if (*msg_ptr == 0) -117 return 0; +115 /* If we are at the end of message, return 0 signifying end of file. */ +116 if (*msg_ptr == 0) +117 return 0; 118 -119 /* Actually put the data into the buffer */ -120 while (length && *msg_ptr) { -121 /* The buffer is in the user data segment, not the kernel -122 * segment so "*" assignment won't work. We have to use -123 * put_user which copies data from the kernel data segment to -124 * the user data segment. -125 */ +119 /* Actually put the data into the buffer */ +120 while (length && *msg_ptr) { +121 /* The buffer is in the user data segment, not the kernel +122 * segment so "*" assignment won't work. We have to use +123 * put_user which copies data from the kernel data segment to +124 * the user data segment. +125 */ 126 put_user(*(msg_ptr++), buffer++); 127 128 length--; 129 bytes_read++; 130 } 131 -132 /* Most read functions return the number of bytes put into the buffer. */ -133 return bytes_read; +132 /* Most read functions return the number of bytes put into the buffer. */ +133 return bytes_read; 134} 135 -136/* Called when a process writes to dev file: echo "hi" > /dev/hello */ -137static ssize_t device_write(struct file *filp, -138 const char *buff, -139 size_t len, +136/* Called when a process writes to dev file: echo "hi" > /dev/hello */ +137static ssize_t device_write(struct file *filp, +138 const char *buff, +139 size_t len, 140 loff_t *off) 141{ -142 pr_alert("Sorry, this operation is not supported.\n"); -143 return -EINVAL; +142 pr_alert("Sorry, this operation is not supported.\n"); +143 return -EINVAL; 144} 145 146module_init(chardev_init); 147module_exit(chardev_exit); 148 -149MODULE_LICENSE("GPL");-
0.6.6 Writing Modules for Multiple Kernel Versions
-The system calls, which are the major interface the kernel shows to the processes, +
The system calls, which are the major interface the kernel shows to the processes, generally stay the same across versions. A new system call may be added, but usually the old ones will behave exactly like they used to. This is necessary for backward compatibility – a new kernel version is not supposed to break regular processes. In most cases, the device files will also remain the same. On the other hand, the internal interfaces within the kernel can and do change between versions. -
There are differences between different kernel versions, and if you want +
There are differences between different kernel versions, and if you want to support multiple kernel versions, you will find yourself having to code conditional compilation directives. The way to do this to compare the macro
LINUX_VERSION_CODEto the macroKERNEL_VERSION. In version a.b.c of the kernel, the value of this macro would be. -
While previous versions of this guide showed how you can write backward +
While previous versions of this guide showed how you can write backward compatible code with such constructs in great detail, we decided to break with this tradition for the better. People interested in doing such might now use a LKMPG with a version matching to their kernel. -
+
0.7 The /proc File System
-In Linux, there is an additional mechanism for the kernel and kernel modules to send +
In Linux, there is an additional mechanism for the kernel and kernel modules to send information to processes — the /proc file system. Originally designed to allow easy access to information about processes (hence the name), it is now used by every bit of the kernel which has something interesting to report, such as /proc/modules which provides the list of modules and /proc/meminfo which stats memory usage statistics. -
The method to use the proc file system is very similar to the one used with device +
The method to use the proc file system is very similar to the one used with device drivers — a structure is created with all the information needed for the /proc file, including pointers to any handler functions (in our case there is only one, the one called when somebody attempts to read from the /proc file). Then, @@ -1566,18 +1563,18 @@ one called when somebody attempts to read from the
registers the structure with the kernel and cleanup_moduleunregisters it. -Normal file systems are located on a disk, rather than just in memory (which is +
Normal file systems are located on a disk, rather than just in memory (which is where /proc is), and in that case the inode number is a pointer to a disk location where the file’s index-node (inode for short) is located. The inode contains information about the file, for example the file’s permissions, together with a pointer to the disk location or locations where the file’s data can be found. -
Because we don’t get called when the file is opened or closed, there’s nowhere for +
Because we don’t get called when the file is opened or closed, there’s nowhere for us to put
try_module_getandtry_module_putin this module, and if the file is opened and then the module is removed, there’s no way to avoid the consequences. -Here a simple example showing how to use a /proc file. This is the HelloWorld for +
Here a simple example showing how to use a /proc file. This is the HelloWorld for the /proc filesystem. There are three parts: create the file /proc/helloworld in the function
init_module, return a value (and a buffer) when the file /proc/helloworld is read in the callback @@ -1585,12 +1582,12 @@ functionprocfile_read, and delete the file /proc/helloworld in the functioncleanup_module. -The /proc/helloworld is created when the module is loaded with the function +
The /proc/helloworld is created when the module is loaded with the function
proc_create -. The return value is astruct proc_dir_entry +. The return value is astruct proc_dir_entry, and it will be used to configure the file /proc/helloworld (for example, the owner of this file). A null return value means that the creation has failed. -Each time, everytime the file /proc/helloworld is read, the function +
Each time, everytime the file /proc/helloworld is read, the function
procfile_readis called. Two parameters of this function are very important: the buffer (the second parameter) and the offset (the fourth one). The content of the @@ -1610,85 +1607,85 @@ endlessly. $ cat /proc/helloworld HelloWorld!+
-
1/* -2 * procfs1.c -3 */ ++1/* +2 * procfs1.c +3 */ 4 -5#include <linux/kernel.h> -6#include <linux/module.h> -7#include <linux/proc_fs.h> -8#include <linux/uaccess.h> -9#include <linux/version.h> +5#include <linux/kernel.h> +6#include <linux/module.h> +7#include <linux/proc_fs.h> +8#include <linux/uaccess.h> +9#include <linux/version.h> 10 -11#if LINUX_VERSION_CODE >= KERNEL_VERSION(5, 6, 0) -12#define HAVE_PROC_OPS -13#endif +11#if LINUX_VERSION_CODE >= KERNEL_VERSION(5, 6, 0) +12#define HAVE_PROC_OPS +13#endif 14 -15#define procfs_name "helloworld" +15#define procfs_name "helloworld" 16 -17struct proc_dir_entry *Our_Proc_File; +17struct proc_dir_entry *Our_Proc_File; 18 19 -20ssize_t procfile_read(struct file *filePointer, -21 char *buffer, -22 size_t buffer_length, +20ssize_t procfile_read(struct file *filePointer, +21 char *buffer, +22 size_t buffer_length, 23 loff_t *offset) 24{ -25 char s[13] = "HelloWorld!\n"; -26 int len = sizeof(s); -27 ssize_t ret = len; +25 char s[13] = "HelloWorld!\n"; +26 int len = sizeof(s); +27 ssize_t ret = len; 28 -29 if (*offset >= len || copy_to_user(buffer, s, len)) { -30 pr_info("copy_to_user failed\n"); +29 if (*offset >= len || copy_to_user(buffer, s, len)) { +30 pr_info("copy_to_user failed\n"); 31 ret = 0; -32 } else { -33 pr_info("procfile read %s\n", filePointer->f_path.dentry->d_name.name); +32 } else { +33 pr_info("procfile read %s\n", filePointer->f_path.dentry->d_name.name); 34 *offset += len; 35 } 36 -37 return ret; +37 return ret; 38} 39 -40#ifdef HAVE_PROC_OPS -41static const struct proc_ops proc_file_fops = { +40#ifdef HAVE_PROC_OPS +41static const struct proc_ops proc_file_fops = { 42 .proc_read = procfile_read, 43}; -44#else -45static const struct file_operations proc_file_fops = { +44#else +45static const struct file_operations proc_file_fops = { 46 .read = procfile_read, 47}; -48#endif +48#endif 49 -50static int __init procfs1_init(void) +50static int __init procfs1_init(void) 51{ 52 Our_Proc_File = proc_create(procfs_name, 0644, NULL, &proc_file_fops); -53 if (NULL == Our_Proc_File) { +53 if (NULL == Our_Proc_File) { 54 proc_remove(Our_Proc_File); -55 pr_alert("Error:Could not initialize /proc/%s\n", procfs_name); -56 return -ENOMEM; +55 pr_alert("Error:Could not initialize /proc/%s\n", procfs_name); +56 return -ENOMEM; 57 } 58 -59 pr_info("/proc/%s created\n", procfs_name); -60 return 0; +59 pr_info("/proc/%s created\n", procfs_name); +60 return 0; 61} 62 -63static void __exit procfs1_exit(void) +63static void __exit procfs1_exit(void) 64{ 65 proc_remove(Our_Proc_File); -66 pr_info("/proc/%s removed\n", procfs_name); +66 pr_info("/proc/%s removed\n", procfs_name); 67} 68 69module_init(procfs1_init); 70module_exit(procfs1_exit); 71 -72MODULE_LICENSE("GPL");-
0.7.1 The proc_ops Structure
-The
proc_ops +asThe
proc_opsstructure is defined in include/linux/proc_fs.h in Linux v5.6+. In older kernels, it usedfile_operationsfor custom hooks in /proc file system, but it contains some @@ -1700,10 +1697,10 @@ performance. For example, the file which never disappears in proc_flagPROC_ENTRY_PERMANENTto save 2 atomic ops, 1 allocation, 1 free in per open/read/close sequence. -+
0.7.2 Read and Write a /proc File
-We have seen a very simple example for a /proc file where we only read +
We have seen a very simple example for a /proc file where we only read the file /proc/helloworld. It is also possible to write in a /proc file. It works the same way as read, a function is called when the /proc file is written. But there is a little difference with read, data comes from @@ -1711,7 +1708,7 @@ user, so you have to import data from user space to kernel space (with
copy_from_userorget_user) -The reason for
copy_from_user +The reason for
copy_from_userorget_useris that Linux memory (on Intel architecture, it may be different under some @@ -1722,7 +1719,7 @@ not reference a unique location in memory, only a location in a memory segment, and you need to know which memory segment it is to be able to use it. There is one memory segment for the kernel, and one for each of the processes. -The only memory segment accessible to a process is its own, so when +
The only memory segment accessible to a process is its own, so when writing regular programs to run as processes, there is no need to worry about segments. When you write a kernel module, normally you want to access the kernel memory segment, which is handled automatically by the system. @@ -1740,132 +1737,132 @@ need to import data because it comes from user space, but not for the read funct because data is already in kernel space.
-
1/* -2 * procfs2.c - create a "file" in /proc -3 */ ++1/* +2 * procfs2.c - create a "file" in /proc +3 */ 4 -5#include <linux/kernel.h> /* We're doing kernel work */ -6#include <linux/module.h> /* Specifically, a module */ -7#include <linux/proc_fs.h> /* Necessary because we use the proc fs */ -8#include <linux/uaccess.h> /* for copy_from_user */ -9#include <linux/version.h> +5#include <linux/kernel.h> /* We're doing kernel work */ +6#include <linux/module.h> /* Specifically, a module */ +7#include <linux/proc_fs.h> /* Necessary because we use the proc fs */ +8#include <linux/uaccess.h> /* for copy_from_user */ +9#include <linux/version.h> 10 -11#if LINUX_VERSION_CODE >= KERNEL_VERSION(5, 6, 0) -12#define HAVE_PROC_OPS -13#endif +11#if LINUX_VERSION_CODE >= KERNEL_VERSION(5, 6, 0) +12#define HAVE_PROC_OPS +13#endif 14 -15#define PROCFS_MAX_SIZE 1024 -16#define PROCFS_NAME "buffer1k" +15#define PROCFS_MAX_SIZE 1024 +16#define PROCFS_NAME "buffer1k" 17 -18/* This structure hold information about the /proc file */ -19static struct proc_dir_entry *Our_Proc_File; +18/* This structure hold information about the /proc file */ +19static struct proc_dir_entry *Our_Proc_File; 20 -21/* The buffer used to store character for this module */ -22static char procfs_buffer[PROCFS_MAX_SIZE]; +21/* The buffer used to store character for this module */ +22static char procfs_buffer[PROCFS_MAX_SIZE]; 23 -24/* The size of the buffer */ -25static unsigned long procfs_buffer_size = 0; +24/* The size of the buffer */ +25static unsigned long procfs_buffer_size = 0; 26 -27/* This function is called then the /proc file is read */ -28ssize_t procfile_read(struct file *filePointer, -29 char *buffer, -30 size_t buffer_length, +27/* This function is called then the /proc file is read */ +28ssize_t procfile_read(struct file *filePointer, +29 char *buffer, +30 size_t buffer_length, 31 loff_t *offset) 32{ -33 char s[13] = "HelloWorld!\n"; -34 int len = sizeof(s); -35 ssize_t ret = len; +33 char s[13] = "HelloWorld!\n"; +34 int len = sizeof(s); +35 ssize_t ret = len; 36 -37 if (*offset >= len || copy_to_user(buffer, s, len)) { -38 pr_info("copy_to_user failed\n"); +37 if (*offset >= len || copy_to_user(buffer, s, len)) { +38 pr_info("copy_to_user failed\n"); 39 ret = 0; -40 } else { -41 pr_info("procfile read %s\n", filePointer->f_path.dentry->d_name.name); +40 } else { +41 pr_info("procfile read %s\n", filePointer->f_path.dentry->d_name.name); 42 *offset += len; 43 } 44 -45 return ret; +45 return ret; 46} 47 -48/* This function is called with the /proc file is written. */ -49static ssize_t procfile_write(struct file *file, -50 const char *buff, -51 size_t len, +48/* This function is called with the /proc file is written. */ +49static ssize_t procfile_write(struct file *file, +50 const char *buff, +51 size_t len, 52 loff_t *off) 53{ 54 procfs_buffer_size = len; -55 if (procfs_buffer_size > PROCFS_MAX_SIZE) +55 if (procfs_buffer_size > PROCFS_MAX_SIZE) 56 procfs_buffer_size = PROCFS_MAX_SIZE; 57 -58 if (copy_from_user(procfs_buffer, buff, procfs_buffer_size)) -59 return -EFAULT; +58 if (copy_from_user(procfs_buffer, buff, procfs_buffer_size)) +59 return -EFAULT; 60 -61 procfs_buffer[procfs_buffer_size] = '\0'; -62 return procfs_buffer_size; +61 procfs_buffer[procfs_buffer_size] = '\0'; +62 return procfs_buffer_size; 63} 64 -65#ifdef HAVE_PROC_OPS -66static const struct proc_ops proc_file_fops = { +65#ifdef HAVE_PROC_OPS +66static const struct proc_ops proc_file_fops = { 67 .proc_read = procfile_read, 68 .proc_write = procfile_write, 69}; -70#else -71static const struct file_operations proc_file_fops = { +70#else +71static const struct file_operations proc_file_fops = { 72 .read = procfile_read, 73 .write = procfile_write, 74}; -75#endif +75#endif 76 -77static int __init procfs2_init(void) +77static int __init procfs2_init(void) 78{ 79 Our_Proc_File = proc_create(PROCFS_NAME, 0644, NULL, &proc_file_fops); -80 if (NULL == Our_Proc_File) { +80 if (NULL == Our_Proc_File) { 81 proc_remove(Our_Proc_File); -82 pr_alert("Error:Could not initialize /proc/%s\n", PROCFS_NAME); -83 return -ENOMEM; +82 pr_alert("Error:Could not initialize /proc/%s\n", PROCFS_NAME); +83 return -ENOMEM; 84 } 85 -86 pr_info("/proc/%s created\n", PROCFS_NAME); -87 return 0; +86 pr_info("/proc/%s created\n", PROCFS_NAME); +87 return 0; 88} 89 -90static void __exit procfs2_exit(void) +90static void __exit procfs2_exit(void) 91{ 92 proc_remove(Our_Proc_File); -93 pr_info("/proc/%s removed\n", PROCFS_NAME); +93 pr_info("/proc/%s removed\n", PROCFS_NAME); 94} 95 96module_init(procfs2_init); 97module_exit(procfs2_exit); 98 -99MODULE_LICENSE("GPL");-
0.7.3 Manage /proc file with standard filesystem
-We have seen how to read and write a /proc file with the /proc interface. But it is +
We have seen how to read and write a /proc file with the /proc interface. But it is also possible to manage /proc file with inodes. The main concern is to use advanced functions, like permissions. -
In Linux, there is a standard mechanism for file system registration. +
In Linux, there is a standard mechanism for file system registration. Since every file system has to have its own functions to handle inode and file operations, there is a special structure to hold pointers to all those functions, -
struct inode_operations -, which includes a pointer tostruct proc_ops +struct inode_operations +, which includes a pointer tostruct proc_ops. -The difference between file and inode operations is that file operations deal with +
The difference between file and inode operations is that file operations deal with the file itself whereas inode operations deal with ways of referencing the file, such as creating links to it. -
In /proc, whenever we register a new file, we’re allowed to specify which -
struct inode_operations +In /proc, whenever we register a new file, we’re allowed to specify which +
struct inode_operationswill be used to access to it. This is the mechanism we use, a -struct inode_operations +struct inode_operations -which includes a pointer to astruct proc_ops +which includes a pointer to astruct proc_opswhich includes pointers to ourprocf_readandprocfs_writefunctions. -Another interesting point here is the +
Another interesting point here is the
module_permissionfunction. This function is called whenever a process tries to do something with the /proc file, and it can decide whether to allow access or not. Right now it is only @@ -1874,7 +1871,7 @@ pointer to a structure which includes information on the currently running process), but it could be based on anything we like, such as what other processes are doing with the same file, the time of day, or the last input we received. -It is important to note that the standard roles of read and write are reversed in +
It is important to note that the standard roles of read and write are reversed in the kernel. Read functions are used for output, whereas write functions are used for input. The reason for that is that read and write refer to the user’s point of view — if a process reads something from the kernel, then the kernel needs to output it, and @@ -1882,128 +1879,128 @@ if a process writes something to the kernel, then the kernel receives it as input.
-
1/* -2 * procfs3.c -3 */ ++1/* +2 * procfs3.c +3 */ 4 -5#include <linux/kernel.h> -6#include <linux/module.h> -7#include <linux/proc_fs.h> -8#include <linux/sched.h> -9#include <linux/uaccess.h> -10#include <linux/version.h> +5#include <linux/kernel.h> +6#include <linux/module.h> +7#include <linux/proc_fs.h> +8#include <linux/sched.h> +9#include <linux/uaccess.h> +10#include <linux/version.h> 11 -12#if LINUX_VERSION_CODE >= KERNEL_VERSION(5, 6, 0) -13#define HAVE_PROC_OPS -14#endif +12#if LINUX_VERSION_CODE >= KERNEL_VERSION(5, 6, 0) +13#define HAVE_PROC_OPS +14#endif 15 -16#define PROCFS_MAX_SIZE 2048 -17#define PROCFS_ENTRY_FILENAME "buffer2k" +16#define PROCFS_MAX_SIZE 2048 +17#define PROCFS_ENTRY_FILENAME "buffer2k" 18 -19struct proc_dir_entry *Our_Proc_File; -20static char procfs_buffer[PROCFS_MAX_SIZE]; -21static unsigned long procfs_buffer_size = 0; +19struct proc_dir_entry *Our_Proc_File; +20static char procfs_buffer[PROCFS_MAX_SIZE]; +21static unsigned long procfs_buffer_size = 0; 22 -23static ssize_t procfs_read(struct file *filp, -24 char *buffer, -25 size_t length, +23static ssize_t procfs_read(struct file *filp, +24 char *buffer, +25 size_t length, 26 loff_t *offset) 27{ -28 static int finished = 0; -29 if (finished) { -30 pr_debug("procfs_read: END\n"); +28 static int finished = 0; +29 if (finished) { +30 pr_debug("procfs_read: END\n"); 31 finished = 0; -32 return 0; +32 return 0; 33 } 34 finished = 1; -35 if (copy_to_user(buffer, procfs_buffer, procfs_buffer_size)) -36 return -EFAULT; -37 pr_debug("procfs_read: read %lu bytes\n", procfs_buffer_size); -38 return procfs_buffer_size; +35 if (copy_to_user(buffer, procfs_buffer, procfs_buffer_size)) +36 return -EFAULT; +37 pr_debug("procfs_read: read %lu bytes\n", procfs_buffer_size); +38 return procfs_buffer_size; 39} -40static ssize_t procfs_write(struct file *file, -41 const char *buffer, -42 size_t len, +40static ssize_t procfs_write(struct file *file, +41 const char *buffer, +42 size_t len, 43 loff_t *off) 44{ -45 if (len > PROCFS_MAX_SIZE) +45 if (len > PROCFS_MAX_SIZE) 46 procfs_buffer_size = PROCFS_MAX_SIZE; -47 else +47 else 48 procfs_buffer_size = len; -49 if (copy_from_user(procfs_buffer, buffer, procfs_buffer_size)) -50 return -EFAULT; -51 pr_debug("procfs_write: write %lu bytes\n", procfs_buffer_size); -52 return procfs_buffer_size; +49 if (copy_from_user(procfs_buffer, buffer, procfs_buffer_size)) +50 return -EFAULT; +51 pr_debug("procfs_write: write %lu bytes\n", procfs_buffer_size); +52 return procfs_buffer_size; 53} -54int procfs_open(struct inode *inode, struct file *file) +54int procfs_open(struct inode *inode, struct file *file) 55{ 56 try_module_get(THIS_MODULE); -57 return 0; +57 return 0; 58} -59int procfs_close(struct inode *inode, struct file *file) +59int procfs_close(struct inode *inode, struct file *file) 60{ 61 module_put(THIS_MODULE); -62 return 0; +62 return 0; 63} 64 -65#ifdef HAVE_PROC_OPS -66static struct proc_ops File_Ops_4_Our_Proc_File = { +65#ifdef HAVE_PROC_OPS +66static struct proc_ops File_Ops_4_Our_Proc_File = { 67 .proc_read = procfs_read, 68 .proc_write = procfs_write, 69 .proc_open = procfs_open, 70 .proc_release = procfs_close, 71}; -72#else -73static const struct file_operations File_Ops_4_Our_Proc_File = { +72#else +73static const struct file_operations File_Ops_4_Our_Proc_File = { 74 .read = procfs_read, 75 .write = procfs_write, 76 .open = procfs_open, 77 .release = procfs_close, 78}; -79#endif +79#endif 80 -81static int __init procfs3_init(void) +81static int __init procfs3_init(void) 82{ 83 Our_Proc_File = proc_create(PROCFS_ENTRY_FILENAME, 0644, NULL, 84 &File_Ops_4_Our_Proc_File); -85 if (Our_Proc_File == NULL) { +85 if (Our_Proc_File == NULL) { 86 remove_proc_entry(PROCFS_ENTRY_FILENAME, NULL); -87 pr_debug("Error: Could not initialize /proc/%s\n", +87 pr_debug("Error: Could not initialize /proc/%s\n", 88 PROCFS_ENTRY_FILENAME); -89 return -ENOMEM; +89 return -ENOMEM; 90 } 91 proc_set_size(Our_Proc_File, 80); 92 proc_set_user(Our_Proc_File, GLOBAL_ROOT_UID, GLOBAL_ROOT_GID); 93 -94 pr_debug("/proc/%s created\n", PROCFS_ENTRY_FILENAME); -95 return 0; +94 pr_debug("/proc/%s created\n", PROCFS_ENTRY_FILENAME); +95 return 0; 96} 97 -98static void __exit procfs3_exit(void) +98static void __exit procfs3_exit(void) 99{ 100 remove_proc_entry(PROCFS_ENTRY_FILENAME, NULL); -101 pr_debug("/proc/%s removed\n", PROCFS_ENTRY_FILENAME); +101 pr_debug("/proc/%s removed\n", PROCFS_ENTRY_FILENAME); 102} 103 104module_init(procfs3_init); 105module_exit(procfs3_exit); 106 -107MODULE_LICENSE("GPL");-Still hungry for procfs examples? Well, first of all keep in mind, there are rumors +107MODULE_LICENSE("GPL");
Still hungry for procfs examples? Well, first of all keep in mind, there are rumors around, claiming that procfs is on its way out, consider using sysfs instead. Consider using this mechanism, in case you want to document something kernel related yourself. -
+
0.7.4 Manage /proc file with seq_file
-As we have seen, writing a /proc file may be quite “complex”. +
As we have seen, writing a /proc file may be quite “complex”. So to help people writting /proc file, there is an API named
seq_filethat helps formating a /proc file for output. It is based on sequence, which is composed of 3 functions: start(), next(), and stop(). Theseq_fileAPI starts a sequence when a user read the /proc file. -A sequence begins with the call of the function start(). If the return is a non +
A sequence begins with the call of the function start(). If the return is a non NULL value, the function next() is called. This function is an iterator, the goal is to go through all the data. Each time next() is called, the function show() is also called. It writes data values in the buffer read by the user. The function next() is called until @@ -2012,7 +2009,7 @@ stop() is called. -
BE CAREFUL: when a sequence is finished, another one starts. That means that +
BE CAREFUL: when a sequence is finished, another one starts. That means that at the end of function stop(), the function start() is called again. This loop finishes when the function start() returns NULL. You can see a scheme of this in the Figure 1. @@ -2026,223 +2023,223 @@ Figure 1
+
Figure 1:How seq_file works -The
seq_file +The
-seq_fileprovides basic functions forproc_ops, such asseq_read,seq_lseek, and some others. But nothing to write in the /proc file. Of course, you can still use the same way as in the previous example.1/* -2 * procfs4.c - create a "file" in /proc -3 * This program uses the seq_file library to manage the /proc file. -4 */ ++1/* +2 * procfs4.c - create a "file" in /proc +3 * This program uses the seq_file library to manage the /proc file. +4 */ 5 -6#include <linux/kernel.h> /* We are doing kernel work */ -7#include <linux/module.h> /* Specifically, a module */ -8#include <linux/proc_fs.h> /* Necessary because we use proc fs */ -9#include <linux/seq_file.h> /* for seq_file */ -10#include <linux/version.h> +6#include <linux/kernel.h> /* We are doing kernel work */ +7#include <linux/module.h> /* Specifically, a module */ +8#include <linux/proc_fs.h> /* Necessary because we use proc fs */ +9#include <linux/seq_file.h> /* for seq_file */ +10#include <linux/version.h> 11 -12#if LINUX_VERSION_CODE >= KERNEL_VERSION(5, 6, 0) -13#define HAVE_PROC_OPS -14#endif +12#if LINUX_VERSION_CODE >= KERNEL_VERSION(5, 6, 0) +13#define HAVE_PROC_OPS +14#endif 15 -16#define PROC_NAME "iter" +16#define PROC_NAME "iter" 17 -18/* This function is called at the beginning of a sequence. -19 * ie, when: -20 * - the /proc file is read (first time) -21 * - after the function stop (end of sequence) -22 */ -23static void *my_seq_start(struct seq_file *s, loff_t *pos) +18/* This function is called at the beginning of a sequence. +19 * ie, when: +20 * - the /proc file is read (first time) +21 * - after the function stop (end of sequence) +22 */ +23static void *my_seq_start(struct seq_file *s, loff_t *pos) 24{ -25 static unsigned long counter = 0; +25 static unsigned long counter = 0; 26 -27 /* beginning a new sequence? */ -28 if (*pos == 0) { -29 /* yes => return a non null value to begin the sequence */ -30 return &counter; +27 /* beginning a new sequence? */ +28 if (*pos == 0) { +29 /* yes => return a non null value to begin the sequence */ +30 return &counter; 31 } -32 /* no => it is the end of the sequence, return end to stop reading */ +32 /* no => it is the end of the sequence, return end to stop reading */ 33 *pos = 0; -34 return NULL; +34 return NULL; 35} 36 -37/* This function is called after the beginning of a sequence. -38 * It is called untill the return is NULL (this ends the sequence). -39 */ -40static void *my_seq_next(struct seq_file *s, void *v, loff_t *pos) +37/* This function is called after the beginning of a sequence. +38 * It is called untill the return is NULL (this ends the sequence). +39 */ +40static void *my_seq_next(struct seq_file *s, void *v, loff_t *pos) 41{ -42 unsigned long *tmp_v = (unsigned long *) v; +42 unsigned long *tmp_v = (unsigned long *) v; 43 (*tmp_v)++; 44 (*pos)++; -45 return NULL; +45 return NULL; 46} 47 -48/* This function is called at the end of a sequence. */ -49static void my_seq_stop(struct seq_file *s, void *v) +48/* This function is called at the end of a sequence. */ +49static void my_seq_stop(struct seq_file *s, void *v) 50{ -51 /* nothing to do, we use a static value in start() */ +51 /* nothing to do, we use a static value in start() */ 52} 53 -54/* This function is called for each "step" of a sequence. */ -55static int my_seq_show(struct seq_file *s, void *v) +54/* This function is called for each "step" of a sequence. */ +55static int my_seq_show(struct seq_file *s, void *v) 56{ 57 loff_t *spos = (loff_t *) v; 58 -59 seq_printf(s, "%Ld\n", *spos); -60 return 0; +59 seq_printf(s, "%Ld\n", *spos); +60 return 0; 61} 62 -63/* This structure gather "function" to manage the sequence */ -64static struct seq_operations my_seq_ops = { +63/* This structure gather "function" to manage the sequence */ +64static struct seq_operations my_seq_ops = { 65 .start = my_seq_start, 66 .next = my_seq_next, 67 .stop = my_seq_stop, 68 .show = my_seq_show, 69}; 70 -71/* This function is called when the /proc file is open. */ -72static int my_open(struct inode *inode, struct file *file) +71/* This function is called when the /proc file is open. */ +72static int my_open(struct inode *inode, struct file *file) 73{ -74 return seq_open(file, &my_seq_ops); +74 return seq_open(file, &my_seq_ops); 75}; 76 -77/* This structure gather "function" that manage the /proc file */ -78#ifdef HAVE_PROC_OPS -79static const struct proc_ops my_file_ops = { +77/* This structure gather "function" that manage the /proc file */ +78#ifdef HAVE_PROC_OPS +79static const struct proc_ops my_file_ops = { 80 .proc_open = my_open, 81 .proc_read = seq_read, 82 .proc_lseek = seq_lseek, 83 .proc_release = seq_release, 84}; -85#else -86static const struct file_operations my_file_ops = { +85#else +86static const struct file_operations my_file_ops = { 87 .open = my_open, 88 .read = seq_read, 89 .llseek = seq_lseek, 90 .release = seq_release, 91}; -92#endif +92#endif 93 -94static int __init procfs4_init(void) +94static int __init procfs4_init(void) 95{ -96 struct proc_dir_entry *entry; +96 struct proc_dir_entry *entry; 97 98 entry = proc_create(PROC_NAME, 0, NULL, &my_file_ops); -99 if (entry == NULL) { +99 if (entry == NULL) { 100 remove_proc_entry(PROC_NAME, NULL); -101 pr_debug("Error: Could not initialize /proc/%s\n", PROC_NAME); -102 return -ENOMEM; +101 pr_debug("Error: Could not initialize /proc/%s\n", PROC_NAME); +102 return -ENOMEM; 103 } 104 -105 return 0; +105 return 0; 106} 107 -108static void __exit procfs4_exit(void) +108static void __exit procfs4_exit(void) 109{ 110 remove_proc_entry(PROC_NAME, NULL); -111 pr_debug("/proc/%s removed\n", PROC_NAME); +111 pr_debug("/proc/%s removed\n", PROC_NAME); 112} 113 114module_init(procfs4_init); 115module_exit(procfs4_exit); 116 -117MODULE_LICENSE("GPL");-If you want more information, you can read this web page: +117MODULE_LICENSE("GPL");
If you want more information, you can read this web page:
-You can also read the code of fs/seq_file.c in the linux kernel. +
You can also read the code of fs/seq_file.c in the linux kernel.
0.8 sysfs: Interacting with your module
-sysfs allows you to interact with the running kernel from userspace by reading or +
sysfs allows you to interact with the running kernel from userspace by reading or setting variables inside of modules. This can be useful for debugging purposes, or just as an interface for applications or scripts. You can find sysfs directories and files -under the sys directory on your system. +under the /sys directory on your system.
1ls -l /sys-An example of a hello world module which includes the creation of a variable +
An example of a hello world module which includes the creation of a variable accessible via sysfs is given below.
-
1/* -2 * hello-sysfs.c sysfs example -3 */ -4#include <linux/fs.h> -5#include <linux/init.h> -6#include <linux/kobject.h> -7#include <linux/module.h> -8#include <linux/string.h> -9#include <linux/sysfs.h> ++1/* +2 * hello-sysfs.c sysfs example +3 */ +4#include <linux/fs.h> +5#include <linux/init.h> +6#include <linux/kobject.h> +7#include <linux/module.h> +8#include <linux/string.h> +9#include <linux/sysfs.h> 10 -11static struct kobject *mymodule; +11static struct kobject *mymodule; 12 -13/* the variable you want to be able to change */ -14static int myvariable = 0; +13/* the variable you want to be able to change */ +14static int myvariable = 0; 15 -16static ssize_t myvariable_show(struct kobject *kobj, -17 struct kobj_attribute *attr, -18 char *buf) +16static ssize_t myvariable_show(struct kobject *kobj, +17 struct kobj_attribute *attr, +18 char *buf) 19{ -20 return sprintf(buf, "%d\n", myvariable); +20 return sprintf(buf, "%d\n", myvariable); 21} 22 -23static ssize_t myvariable_store(struct kobject *kobj, -24 struct kobj_attribute *attr, -25 char *buf, -26 size_t count) +23static ssize_t myvariable_store(struct kobject *kobj, +24 struct kobj_attribute *attr, +25 char *buf, +26 size_t count) 27{ -28 sscanf(buf, "%du", &myvariable); -29 return count; +28 sscanf(buf, "%du", &myvariable); +29 return count; 30} 31 32 -33static struct kobj_attribute myvariable_attribute = -34 __ATTR(myvariable, 0660, myvariable_show, (void *) myvariable_store); +33static struct kobj_attribute myvariable_attribute = +34 __ATTR(myvariable, 0660, myvariable_show, (void *) myvariable_store); 35 -36static int __init mymodule_init(void) +36static int __init mymodule_init(void) 37{ -38 int error = 0; +38 int error = 0; 39 -40 pr_info("mymodule: initialised\n"); +40 pr_info("mymodule: initialised\n"); 41 -42 mymodule = kobject_create_and_add("mymodule", kernel_kobj); -43 if (!mymodule) -44 return -ENOMEM; +42 mymodule = kobject_create_and_add("mymodule", kernel_kobj); +43 if (!mymodule) +44 return -ENOMEM; 45 46 error = sysfs_create_file(mymodule, &myvariable_attribute.attr); -47 if (error) { +47 if (error) { 48 pr_info( -49 "failed to create the myvariable file " -50 "in /sys/kernel/mymodule\n"); +49 "failed to create the myvariable file " +50 "in /sys/kernel/mymodule\n"); 51 } 52 -53 return error; +53 return error; 54} 55 -56static void __exit mymodule_exit(void) +56static void __exit mymodule_exit(void) 57{ -58 pr_info("mymodule: Exit success\n"); +58 pr_info("mymodule: Exit success\n"); 59 kobject_put(mymodule); 60} 61 62module_init(mymodule_init); 63module_exit(mymodule_exit); 64 -65MODULE_LICENSE("GPL");-Make and install the module:
1make @@ -2250,36 +2247,36 @@ accessible via sysfs is given below. --Check that it exists: +
Check that it exists:
1sudo lsmod | grep hello_sysfs-What is the current value of
myvariable +will still try to callWhat is the current value of
myvariable?
1cat /sys/kernel/mymodule/myvariable-Set the value of
myvariable +right before you do theSet the value of
myvariableand check that it changed.-
1echo "32" > /sys/kernel/mymodule/myvariable +1echo "32" > /sys/kernel/mymodule/myvariable 2cat /sys/kernel/mymodule/myvariable-Finally, remove the test module: +
Finally, remove the test module:
1sudo rmmod hello_sysfs-+
0.9 Talking To Device Files
-Device files are supposed to represent physical devices. Most physical devices are +
Device files are supposed to represent physical devices. Most physical devices are used for output as well as input, so there has to be some mechanism for device drivers in the kernel to get the output to send to the device from processes. This is done by opening the device file for output and writing to it, just like writing to a file. In the following example, this is implemented by
device_write. -This is not always enough. Imagine you had a serial port connected to a modem +
This is not always enough. Imagine you had a serial port connected to a modem (even if you have an internal modem, it is still implemented from the CPU’s perspective as a serial port connected to a modem, so you don’t have to tax your imagination too hard). The natural thing to do would be to use the @@ -2292,7 +2289,7 @@ received. -
The answer in Unix is to use a special function called +
The answer in Unix is to use a special function called
ioctl(short for Input Output ConTroL). Every device can have its own ioctl commands, which can be read ioctl’s (to send information from a process to the kernel), @@ -2300,12 +2297,12 @@ write ioctl’s (to return information to a process), both or neither. Notice here the roles of read and write are reversed again, so in ioctl’s read is to send information to the kernel and write is to receive information from the kernel. -The ioctl function is called with three parameters: the file descriptor of the +
The ioctl function is called with three parameters: the file descriptor of the appropriate device file, the ioctl number, and a parameter, which is of type long so you can use a cast to use it to pass anything. You will not be able to pass a structure this way, but you will be able to pass a pointer to the structure. -
The ioctl number encodes the major device number, the type of the ioctl, the +
The ioctl number encodes the major device number, the type of the ioctl, the command, and the type of the parameter. This ioctl number is usually created by a macro call (
_IO,_IOR @@ -2316,213 +2313,213 @@ included both by the programs which will use ioctl (so they can generate the appropriate ioctl’s) and by the kernel module (so it can understand it). In the example below, the header file is chardev.h and the program which uses it is ioctl.c. -If you want to use ioctls in your own kernel modules, it is best to receive an +
If you want to use ioctls in your own kernel modules, it is best to receive an official ioctl assignment, so if you accidentally get somebody else’s ioctls, or if they get yours, you’ll know something is wrong. For more information, consult the kernel -source tree at Documentation/ioctl-number.txt. +source tree at Documentation/driver-api/ioctl.rst.
-
1/* -2 * chardev2.c - Create an input/output character device -3 */ +1/* +2 * chardev2.c - Create an input/output character device +3 */ 4 -5#include <linux/cdev.h> -6#include <linux/delay.h> -7#include <linux/device.h> -8#include <linux/fs.h> -9#include <linux/init.h> -10#include <linux/irq.h> -11#include <linux/kernel.h> /* We are doing kernel work */ -12#include <linux/module.h> /* Specifically, a module */ -13#include <linux/poll.h> +5#include <linux/cdev.h> +6#include <linux/delay.h> +7#include <linux/device.h> +8#include <linux/fs.h> +9#include <linux/init.h> +10#include <linux/irq.h> +11#include <linux/kernel.h> /* We are doing kernel work */ +12#include <linux/module.h> /* Specifically, a module */ +13#include <linux/poll.h> 14 -15#include "chardev.h" -16#define SUCCESS 0 -17#define DEVICE_NAME "char_dev" -18#define BUF_LEN 80 +15#include "chardev.h" +16#define SUCCESS 0 +17#define DEVICE_NAME "char_dev" +18#define BUF_LEN 80 19 -20/* Is the device open right now? Used to prevent concurent access into -21 * the same device -22 */ -23static int Device_Open = 0; +20/* Is the device open right now? Used to prevent concurent access into +21 * the same device +22 */ +23static int Device_Open = 0; 24 -25/* The message the device will give when asked */ -26static char Message[BUF_LEN]; +25/* The message the device will give when asked */ +26static char Message[BUF_LEN]; 27 -28/* How far did the process reading the message get? Useful if the message -29 * is larger than the size of the buffer we get to fill in device_read. -30 */ -31static char *Message_Ptr; +28/* How far did the process reading the message get? Useful if the message +29 * is larger than the size of the buffer we get to fill in device_read. +30 */ +31static char *Message_Ptr; 32 -33/* Major number assigned to our device driver */ -34static int Major; -35static struct class *cls; +33/* Major number assigned to our device driver */ +34static int Major; +35static struct class *cls; 36 -37/* This is called whenever a process attempts to open the device file */ -38static int device_open(struct inode *inode, struct file *file) +37/* This is called whenever a process attempts to open the device file */ +38static int device_open(struct inode *inode, struct file *file) 39{ -40 pr_info("device_open(%p)\n", file); +40 pr_info("device_open(%p)\n", file); 41 -42 /* We don't want to talk to two processes at the same time. */ -43 if (Device_Open) -44 return -EBUSY; +42 /* We don't want to talk to two processes at the same time. */ +43 if (Device_Open) +44 return -EBUSY; 45 46 Device_Open++; -47 /* Initialize the message */ +47 /* Initialize the message */ 48 Message_Ptr = Message; 49 try_module_get(THIS_MODULE); -50 return SUCCESS; +50 return SUCCESS; 51} 52 -53static int device_release(struct inode *inode, struct file *file) +53static int device_release(struct inode *inode, struct file *file) 54{ -55 pr_info("device_release(%p,%p)\n", inode, file); +55 pr_info("device_release(%p,%p)\n", inode, file); 56 -57 /* We're now ready for our next caller */ +57 /* We're now ready for our next caller */ 58 Device_Open--; 59 60 module_put(THIS_MODULE); -61 return SUCCESS; +61 return SUCCESS; 62} 63 -64/* This function is called whenever a process which has already opened the -65 * device file attempts to read from it. -66 */ -67static ssize_t device_read(struct file *file, /* see include/linux/fs.h */ -68 char __user *buffer, /* buffer to be filled */ -69 size_t length, /* length of the buffer */ +64/* This function is called whenever a process which has already opened the +65 * device file attempts to read from it. +66 */ +67static ssize_t device_read(struct file *file, /* see include/linux/fs.h */ +68 char __user *buffer, /* buffer to be filled */ +69 size_t length, /* length of the buffer */ 70 loff_t *offset) 71{ -72 /* Number of bytes actually written to the buffer */ -73 int bytes_read = 0; +72 /* Number of bytes actually written to the buffer */ +73 int bytes_read = 0; 74 -75 pr_info("device_read(%p,%p,%ld)\n", file, buffer, length); +75 pr_info("device_read(%p,%p,%ld)\n", file, buffer, length); 76 -77 /* If at the end of message, return 0 (which signifies end of file). */ -78 if (*Message_Ptr == 0) -79 return 0; +77 /* If at the end of message, return 0 (which signifies end of file). */ +78 if (*Message_Ptr == 0) +79 return 0; 80 -81 /* Actually put the data into the buffer */ -82 while (length && *Message_Ptr) { -83 /* Because the buffer is in the user data segment, not the kernel -84 * data segment, assignment would not work. Instead, we have to -85 * use put_user which copies data from the kernel data segment to -86 * the user data segment. -87 */ +81 /* Actually put the data into the buffer */ +82 while (length && *Message_Ptr) { +83 /* Because the buffer is in the user data segment, not the kernel +84 * data segment, assignment would not work. Instead, we have to +85 * use put_user which copies data from the kernel data segment to +86 * the user data segment. +87 */ 88 put_user(*(Message_Ptr++), buffer++); 89 length--; 90 bytes_read++; 91 } 92 -93 pr_info("Read %d bytes, %ld left\n", bytes_read, length); +93 pr_info("Read %d bytes, %ld left\n", bytes_read, length); 94 -95 /* Read functions are supposed to return the number of bytes actually -96 * inserted into the buffer. -97 */ -98 return bytes_read; +95 /* Read functions are supposed to return the number of bytes actually +96 * inserted into the buffer. +97 */ +98 return bytes_read; 99} 100 -101/* called when somebody tries to write into our device file. */ -102static ssize_t device_write(struct file *file, -103 const char __user *buffer, -104 size_t length, +101/* called when somebody tries to write into our device file. */ +102static ssize_t device_write(struct file *file, +103 const char __user *buffer, +104 size_t length, 105 loff_t *offset) 106{ -107 int i; +107 int i; 108 -109 pr_info("device_write(%p,%s,%ld)", file, buffer, length); +109 pr_info("device_write(%p,%s,%ld)", file, buffer, length); 110 -111 for (i = 0; i < length && i < BUF_LEN; i++) +111 for (i = 0; i < length && i < BUF_LEN; i++) 112 get_user(Message[i], buffer + i); 113 114 Message_Ptr = Message; 115 -116 /* Again, return the number of input characters used. */ -117 return i; +116 /* Again, return the number of input characters used. */ +117 return i; 118} 119 -120/* This function is called whenever a process tries to do an ioctl on our -121 * device file. We get two extra parameters (additional to the inode and file -122 * structures, which all device functions get): the number of the ioctl called -123 * and the parameter given to the ioctl function. -124 * -125 * If the ioctl is write or read/write (meaning output is returned to the -126 * calling process), the ioctl call returns the output of this function. -127 */ -128long device_ioctl(struct file *file, /* ditto */ -129 unsigned int ioctl_num, /* number and param for ioctl */ -130 unsigned long ioctl_param) +120/* This function is called whenever a process tries to do an ioctl on our +121 * device file. We get two extra parameters (additional to the inode and file +122 * structures, which all device functions get): the number of the ioctl called +123 * and the parameter given to the ioctl function. +124 * +125 * If the ioctl is write or read/write (meaning output is returned to the +126 * calling process), the ioctl call returns the output of this function. +127 */ +128long device_ioctl(struct file *file, /* ditto */ +129 unsigned int ioctl_num, /* number and param for ioctl */ +130 unsigned long ioctl_param) 131{ -132 int i; -133 char *temp; -134 char ch; +132 int i; +133 char *temp; +134 char ch; 135 -136 /* Switch according to the ioctl called */ -137 switch (ioctl_num) { -138 case IOCTL_SET_MSG: -139 /* Receive a pointer to a message (in user space) and set that to -140 * be the device's message. Get the parameter given to ioctl by -141 * the process. -142 */ -143 temp = (char *) ioctl_param; +136 /* Switch according to the ioctl called */ +137 switch (ioctl_num) { +138 case IOCTL_SET_MSG: +139 /* Receive a pointer to a message (in user space) and set that to +140 * be the device's message. Get the parameter given to ioctl by +141 * the process. +142 */ +143 temp = (char *) ioctl_param; 144 -145 /* Find the length of the message */ +145 /* Find the length of the message */ 146 get_user(ch, temp); -147 for (i = 0; ch && i < BUF_LEN; i++, temp++) +147 for (i = 0; ch && i < BUF_LEN; i++, temp++) 148 get_user(ch, temp); 149 -150 device_write(file, (char *) ioctl_param, i, 0); -151 break; +150 device_write(file, (char *) ioctl_param, i, 0); +151 break; 152 -153 case IOCTL_GET_MSG: -154 /* Give the current message to the calling process - the parameter -155 * we got is a pointer, fill it. -156 */ -157 i = device_read(file, (char *) ioctl_param, 99, 0); +153 case IOCTL_GET_MSG: +154 /* Give the current message to the calling process - the parameter +155 * we got is a pointer, fill it. +156 */ +157 i = device_read(file, (char *) ioctl_param, 99, 0); 158 -159 /* Put a zero at the end of the buffer, so it will be properly -160 * terminated. -161 */ -162 put_user('\0', (char *) ioctl_param + i); -163 break; +159 /* Put a zero at the end of the buffer, so it will be properly +160 * terminated. +161 */ +162 put_user('\0', (char *) ioctl_param + i); +163 break; 164 -165 case IOCTL_GET_NTH_BYTE: -166 /* This ioctl is both input (ioctl_param) and output (the return -167 * value of this function). -168 */ -169 return Message[ioctl_param]; -170 break; +165 case IOCTL_GET_NTH_BYTE: +166 /* This ioctl is both input (ioctl_param) and output (the return +167 * value of this function). +168 */ +169 return Message[ioctl_param]; +170 break; 171 } 172 -173 return SUCCESS; +173 return SUCCESS; 174} 175 -176/* Module Declarations */ +176/* Module Declarations */ 177 -178/* This structure will hold the functions to be called when a process does -179 * something to the device we created. Since a pointer to this structure -180 * is kept in the devices table, it can't be local to init_module. NULL is -181 * for unimplemented functions. -182 */ -183struct file_operations Fops = { +178/* This structure will hold the functions to be called when a process does +179 * something to the device we created. Since a pointer to this structure +180 * is kept in the devices table, it can't be local to init_module. NULL is +181 * for unimplemented functions. +182 */ +183struct file_operations Fops = { 184 .read = device_read, 185 .write = device_write, 186 .unlocked_ioctl = device_ioctl, 187 .open = device_open, -188 .release = device_release, /* a.k.a. close */ +188 .release = device_release, /* a.k.a. close */ 189}; 190 -191/* Initialize the module - Register the character device */ -192static int __init chardev2_init(void) +191/* Initialize the module - Register the character device */ +192static int __init chardev2_init(void) 193{ -194 /* Register the character device (atleast try) */ -195 int ret_val = register_chrdev(MAJOR_NUM, DEVICE_NAME, &Fops); +194 /* Register the character device (atleast try) */ +195 int ret_val = register_chrdev(MAJOR_NUM, DEVICE_NAME, &Fops); 196 -197 /* Negative values signify an error */ -198 if (ret_val < 0) { -199 pr_alert("%s failed with %d\n", -200 "Sorry, registering the character device ", ret_val); -201 return ret_val; +197 /* Negative values signify an error */ +198 if (ret_val < 0) { +199 pr_alert("%s failed with %d\n", +200 "Sorry, registering the character device ", ret_val); +201 return ret_val; 202 } 203 204 Major = ret_val; @@ -2530,223 +2527,223 @@ source tree at Documentation/ioctl-number.txt. 206 cls = class_create(THIS_MODULE, DEVICE_FILE_NAME); 207 device_create(cls, NULL, MKDEV(Major, MAJOR_NUM), NULL, DEVICE_FILE_NAME); 208 -209 pr_info("Device created on /dev/%s\n", DEVICE_FILE_NAME); +209 pr_info("Device created on /dev/%s\n", DEVICE_FILE_NAME); 210 -211 return 0; +211 return 0; 212} 213 -214/* Cleanup - unregister the appropriate file from /proc */ -215static void __exit chardev2_exit(void) +214/* Cleanup - unregister the appropriate file from /proc */ +215static void __exit chardev2_exit(void) 216{ 217 device_destroy(cls, MKDEV(Major, 0)); 218 class_destroy(cls); 219 -220 /* Unregister the device */ +220 /* Unregister the device */ 221 unregister_chrdev(Major, DEVICE_NAME); 222} 223 224module_init(chardev2_init); 225module_exit(chardev2_exit); 226 -227MODULE_LICENSE("GPL");+227MODULE_LICENSE("GPL");-
1/* -2 * chardev.h - the header file with the ioctl definitions. -3 * -4 * The declarations here have to be in a header file, because they need -5 * to be known both to the kernel module (in chardev.c) and the process -6 * calling ioctl (ioctl.c). -7 */ +1/* +2 * chardev.h - the header file with the ioctl definitions. +3 * +4 * The declarations here have to be in a header file, because they need +5 * to be known both to the kernel module (in chardev.c) and the process +6 * calling ioctl (ioctl.c). +7 */ 8 -9#ifndef CHARDEV_H -10#define CHARDEV_H +9#ifndef CHARDEV_H +10#define CHARDEV_H 11 -12#include <linux/ioctl.h> +12#include <linux/ioctl.h> 13 -14/* The major device number. We can not rely on dynamic registration -15 * any more, because ioctls need to know it. -16 */ -17#define MAJOR_NUM 100 +14/* The major device number. We can not rely on dynamic registration +15 * any more, because ioctls need to know it. +16 */ +17#define MAJOR_NUM 100 18 -19/* Set the message of the device driver */ -20#define IOCTL_SET_MSG _IOW(MAJOR_NUM, 0, char *) -21/* _IOW means that we are creating an ioctl command number for passing -22 * information from a user process to the kernel module. -23 * -24 * The first arguments, MAJOR_NUM, is the major device number we are using. -25 * -26 * The second argument is the number of the command (there could be several -27 * with different meanings). -28 * -29 * The third argument is the type we want to get from the process to the -30 * kernel. -31 */ +19/* Set the message of the device driver */ +20#define IOCTL_SET_MSG _IOW(MAJOR_NUM, 0, char *) +21/* _IOW means that we are creating an ioctl command number for passing +22 * information from a user process to the kernel module. +23 * +24 * The first arguments, MAJOR_NUM, is the major device number we are using. +25 * +26 * The second argument is the number of the command (there could be several +27 * with different meanings). +28 * +29 * The third argument is the type we want to get from the process to the +30 * kernel. +31 */ 32 -33/* Get the message of the device driver */ -34#define IOCTL_GET_MSG _IOR(MAJOR_NUM, 1, char *) -35/* This IOCTL is used for output, to get the message of the device driver. -36 * However, we still need the buffer to place the message in to be input, -37 * as it is allocated by the process. -38 */ +33/* Get the message of the device driver */ +34#define IOCTL_GET_MSG _IOR(MAJOR_NUM, 1, char *) +35/* This IOCTL is used for output, to get the message of the device driver. +36 * However, we still need the buffer to place the message in to be input, +37 * as it is allocated by the process. +38 */ 39 -40/* Get the n'th byte of the message */ -41#define IOCTL_GET_NTH_BYTE _IOWR(MAJOR_NUM, 2, int) -42/* The IOCTL is used for both input and output. It receives from the user -43 * a number, n, and returns Message[n]. -44 */ +40/* Get the n'th byte of the message */ +41#define IOCTL_GET_NTH_BYTE _IOWR(MAJOR_NUM, 2, int) +42/* The IOCTL is used for both input and output. It receives from the user +43 * a number, n, and returns Message[n]. +44 */ 45 -46/* The name of the device file */ -47#define DEVICE_FILE_NAME "char_dev" +46/* The name of the device file */ +47#define DEVICE_FILE_NAME "char_dev" 48 -49#endif+49#endif-
1/* -2 * ioctl.c -3 */ -4#include <linux/cdev.h> -5#include <linux/fs.h> -6#include <linux/init.h> -7#include <linux/ioctl.h> -8#include <linux/module.h> -9#include <linux/slab.h> -10#include <linux/uaccess.h> ++1/* +2 * ioctl.c +3 */ +4#include <linux/cdev.h> +5#include <linux/fs.h> +6#include <linux/init.h> +7#include <linux/ioctl.h> +8#include <linux/module.h> +9#include <linux/slab.h> +10#include <linux/uaccess.h> 11 -12struct ioctl_arg { -13 unsigned int reg; -14 unsigned int val; +12struct ioctl_arg { +13 unsigned int reg; +14 unsigned int val; 15}; 16 -17/* Documentation/ioctl/ioctl-number.txt */ -18#define IOC_MAGIC '\x66' +17/* Documentation/ioctl/ioctl-number.txt */ +18#define IOC_MAGIC '\x66' 19 -20#define IOCTL_VALSET _IOW(IOC_MAGIC, 0, struct ioctl_arg) -21#define IOCTL_VALGET _IOR(IOC_MAGIC, 1, struct ioctl_arg) -22#define IOCTL_VALGET_NUM _IOR(IOC_MAGIC, 2, int) -23#define IOCTL_VALSET_NUM _IOW(IOC_MAGIC, 3, int) +20#define IOCTL_VALSET _IOW(IOC_MAGIC, 0, struct ioctl_arg) +21#define IOCTL_VALGET _IOR(IOC_MAGIC, 1, struct ioctl_arg) +22#define IOCTL_VALGET_NUM _IOR(IOC_MAGIC, 2, int) +23#define IOCTL_VALSET_NUM _IOW(IOC_MAGIC, 3, int) 24 -25#define IOCTL_VAL_MAXNR 3 -26#define DRIVER_NAME "ioctltest" +25#define IOCTL_VAL_MAXNR 3 +26#define DRIVER_NAME "ioctltest" 27 -28static unsigned int test_ioctl_major = 0; -29static unsigned int num_of_dev = 1; -30static struct cdev test_ioctl_cdev; -31static int ioctl_num = 0; +28static unsigned int test_ioctl_major = 0; +29static unsigned int num_of_dev = 1; +30static struct cdev test_ioctl_cdev; +31static int ioctl_num = 0; 32 -33struct test_ioctl_data { -34 unsigned char val; +33struct test_ioctl_data { +34 unsigned char val; 35 rwlock_t lock; 36}; 37 -38static long test_ioctl_ioctl(struct file *filp, -39 unsigned int cmd, -40 unsigned long arg) +38static long test_ioctl_ioctl(struct file *filp, +39 unsigned int cmd, +40 unsigned long arg) 41{ -42 struct test_ioctl_data *ioctl_data = filp->private_data; -43 int retval = 0; -44 unsigned char val; -45 struct ioctl_arg data; -46 memset(&data, 0, sizeof(data)); +42 struct test_ioctl_data *ioctl_data = filp->private_data; +43 int retval = 0; +44 unsigned char val; +45 struct ioctl_arg data; +46 memset(&data, 0, sizeof(data)); 47 -48 switch (cmd) { -49 case IOCTL_VALSET: -50 if (copy_from_user(&data, (int __user *) arg, sizeof(data))) { +48 switch (cmd) { +49 case IOCTL_VALSET: +50 if (copy_from_user(&data, (int __user *) arg, sizeof(data))) { 51 retval = -EFAULT; -52 goto done; +52 goto done; 53 } 54 -55 pr_alert("IOCTL set val:%x .\n", data.val); +55 pr_alert("IOCTL set val:%x .\n", data.val); 56 write_lock(&ioctl_data->lock); 57 ioctl_data->val = data.val; 58 write_unlock(&ioctl_data->lock); -59 break; +59 break; 60 -61 case IOCTL_VALGET: +61 case IOCTL_VALGET: 62 read_lock(&ioctl_data->lock); 63 val = ioctl_data->val; 64 read_unlock(&ioctl_data->lock); 65 data.val = val; 66 -67 if (copy_to_user((int __user *) arg, &data, sizeof(data))) { +67 if (copy_to_user((int __user *) arg, &data, sizeof(data))) { 68 retval = -EFAULT; -69 goto done; +69 goto done; 70 } 71 -72 break; +72 break; 73 -74 case IOCTL_VALGET_NUM: -75 retval = __put_user(ioctl_num, (int __user *) arg); -76 break; +74 case IOCTL_VALGET_NUM: +75 retval = __put_user(ioctl_num, (int __user *) arg); +76 break; 77 -78 case IOCTL_VALSET_NUM: +78 case IOCTL_VALSET_NUM: 79 ioctl_num = arg; -80 break; +80 break; 81 -82 default: +82 default: 83 retval = -ENOTTY; 84 } 85 86done: -87 return retval; +87 return retval; 88} 89 -90ssize_t test_ioctl_read(struct file *filp, -91 char __user *buf, -92 size_t count, +90ssize_t test_ioctl_read(struct file *filp, +91 char __user *buf, +92 size_t count, 93 loff_t *f_pos) 94{ -95 struct test_ioctl_data *ioctl_data = filp->private_data; -96 unsigned char val; -97 int retval; -98 int i = 0; +95 struct test_ioctl_data *ioctl_data = filp->private_data; +96 unsigned char val; +97 int retval; +98 int i = 0; 99 read_lock(&ioctl_data->lock); 100 val = ioctl_data->val; 101 read_unlock(&ioctl_data->lock); 102 -103 for (; i < count; i++) { -104 if (copy_to_user(&buf[i], &val, 1)) { +103 for (; i < count; i++) { +104 if (copy_to_user(&buf[i], &val, 1)) { 105 retval = -EFAULT; -106 goto out; +106 goto out; 107 } 108 } 109 110 retval = count; 111out: -112 return retval; +112 return retval; 113} 114 -115static int test_ioctl_close(struct inode *inode, struct file *filp) +115static int test_ioctl_close(struct inode *inode, struct file *filp) 116{ -117 pr_alert("%s call.\n", __func__); +117 pr_alert("%s call.\n", __func__); 118 -119 if (filp->private_data) { +119 if (filp->private_data) { 120 kfree(filp->private_data); 121 filp->private_data = NULL; 122 } 123 -124 return 0; +124 return 0; 125} 126 -127static int test_ioctl_open(struct inode *inode, struct file *filp) +127static int test_ioctl_open(struct inode *inode, struct file *filp) 128{ -129 struct test_ioctl_data *ioctl_data; -130 pr_alert("%s call.\n", __func__); -131 ioctl_data = kmalloc(sizeof(struct test_ioctl_data), GFP_KERNEL); +129 struct test_ioctl_data *ioctl_data; +130 pr_alert("%s call.\n", __func__); +131 ioctl_data = kmalloc(sizeof(struct test_ioctl_data), GFP_KERNEL); 132 -133 if (ioctl_data == NULL) { -134 return -ENOMEM; +133 if (ioctl_data == NULL) { +134 return -ENOMEM; 135 } 136 137 rwlock_init(&ioctl_data->lock); 138 ioctl_data->val = 0xFF; 139 filp->private_data = ioctl_data; -140 return 0; +140 return 0; 141} 142 -143struct file_operations fops = { +143struct file_operations fops = { 144 .owner = THIS_MODULE, 145 .open = test_ioctl_open, 146 .release = test_ioctl_close, @@ -2754,66 +2751,66 @@ source tree at Documentation/ioctl-number.txt. 148 .unlocked_ioctl = test_ioctl_ioctl, 149}; 150 -151static int ioctl_init(void) +151static int ioctl_init(void) 152{ -153 dev_t dev = MKDEV(test_ioctl_major, 0); -154 int alloc_ret = 0; -155 int cdev_ret = 0; +153 dev_t dev = MKDEV(test_ioctl_major, 0); +154 int alloc_ret = 0; +155 int cdev_ret = 0; 156 alloc_ret = alloc_chrdev_region(&dev, 0, num_of_dev, DRIVER_NAME); 157 -158 if (alloc_ret) { -159 goto error; +158 if (alloc_ret) { +159 goto error; 160 } 161 162 test_ioctl_major = MAJOR(dev); 163 cdev_init(&test_ioctl_cdev, &fops); 164 cdev_ret = cdev_add(&test_ioctl_cdev, dev, num_of_dev); 165 -166 if (cdev_ret) { -167 goto error; +166 if (cdev_ret) { +167 goto error; 168 } 169 -170 pr_alert("%s driver(major: %d) installed.\n", DRIVER_NAME, +170 pr_alert("%s driver(major: %d) installed.\n", DRIVER_NAME, 171 test_ioctl_major); -172 return 0; +172 return 0; 173error: 174 -175 if (cdev_ret == 0) { +175 if (cdev_ret == 0) { 176 cdev_del(&test_ioctl_cdev); 177 } 178 -179 if (alloc_ret == 0) { +179 if (alloc_ret == 0) { 180 unregister_chrdev_region(dev, num_of_dev); 181 } 182 -183 return -1; +183 return -1; 184} 185 -186static void ioctl_exit(void) +186static void ioctl_exit(void) 187{ -188 dev_t dev = MKDEV(test_ioctl_major, 0); +188 dev_t dev = MKDEV(test_ioctl_major, 0); 189 cdev_del(&test_ioctl_cdev); 190 unregister_chrdev_region(dev, num_of_dev); -191 pr_alert("%s driver removed.\n", DRIVER_NAME); +191 pr_alert("%s driver removed.\n", DRIVER_NAME); 192} 193 194module_init(ioctl_init); 195module_exit(ioctl_exit); 196 -197MODULE_LICENSE("GPL"); -198MODULE_DESCRIPTION("This is test_ioctl module");-+197MODULE_LICENSE("GPL"); +198MODULE_DESCRIPTION("This is test_ioctl module");
0.10 System Calls
-So far, the only thing we’ve done was to use well defined kernel mechanisms to +
So far, the only thing we’ve done was to use well defined kernel mechanisms to register /proc files and device handlers. This is fine if you want to do something the kernel programmers thought you’d want, such as write a device driver. But what if you want to do something unusual, to change the behavior of the system in some way? Then, you are mostly on your own. -
If you are not being sensible and using a virtual machine then this is where kernel +
If you are not being sensible and using a virtual machine then this is where kernel programming can become hazardous. While writing the example below, I killed the
open()system call. This meant I could not open any files, I could not run any @@ -2825,7 +2822,7 @@ ensure you do not lose any files, even within a test environment, please runinsmodand thermmod. -Forget about /proc files, forget about device files. They are just minor details. +
Forget about /proc files, forget about device files. They are just minor details. Minutiae in the vast expanse of the universe. The real process to kernel communication mechanism, the one used by all processes, is system calls. When a process requests a service from the kernel (such as opening a file, forking to a new @@ -2834,11 +2831,11 @@ change the behaviour of the kernel in interesting ways, this is the place to do it. By the way, if you want to see which system calls a program uses, run
strace <arguments>. -In general, a process is not supposed to be able to access the kernel. It can not +
In general, a process is not supposed to be able to access the kernel. It can not access kernel memory and it can’t call kernel functions. The hardware of the CPU enforces this (that is the reason why it is called “protected mode” or “page protection”). -
System calls are an exception to this general rule. What happens is that the +
System calls are an exception to this general rule. What happens is that the process fills the registers with the appropriate values and then calls a special instruction which jumps to a previously defined location in the kernel (of course, that location is readable by user processes, it is not writable by them). Under Intel CPUs, @@ -2846,7 +2843,7 @@ this is done by means of interrupt 0x80. The hardware knows that once you jump t this location, you are no longer running in restricted user mode, but as the operating system kernel — and therefore you’re allowed to do whatever you want. -
The location in the kernel a process can jump to is called system_call. The +
The location in the kernel a process can jump to is called system_call. The procedure at that location checks the system call number, which tells the kernel what service the process requested. Then, it looks at the table of system calls (
sys_call_table @@ -2859,7 +2856,7 @@ different process, if the process time ran out). If you want to read this code, at the source file arch/$(architecture)/kernel/entry.S, after the lineENTRY(system_call). -So, if we want to change the way a certain system call works, what we need to do +
So, if we want to change the way a certain system call works, what we need to do is to write our own function to implement it (usually by adding a bit of our own code, and then calling the original function) and then change the pointer at
sys_call_table @@ -2867,7 +2864,7 @@ code, and then calling the original function) and then change the pointer at don’t want to leave the system in an unstable state, it’s important forcleanup_moduleto restore the table to its original state. -The source code here is an example of such a kernel module. We want to “spy” on a certain +
The source code here is an example of such a kernel module. We want to “spy” on a certain user, and to
pr_info()a message whenever that user opens a file. Towards this end, we replace the system call to open a file with our own function, called @@ -2877,7 +2874,7 @@ spy on, it callspr_info()to display the name of the file to be opened. Then, either way, it calls the original open() function with the same parameters, to actually open the file. -The
init_module +The
init_modulefunction replaces the appropriate location insys_call_tableand keeps the original pointer in a variable. The @@ -2895,7 +2892,7 @@ withB_open, which will call what it thinks is the original system call,A_open, when it’s done. -Now, if B is removed first, everything will be well — it will simply restore the system +
Now, if B is removed first, everything will be well — it will simply restore the system call to
A_open, which calls the original. However, if A is removed and then B is removed, the system will crash. A’s removal will restore the system call to the original, @@ -2918,7 +2915,7 @@ problem. When A is removed, it sees that the system call was changed toA_openwhich is no longer there, so that even without removing B the system would crash. -Note that all the related problems make syscall stealing unfeasiable for +
Note that all the related problems make syscall stealing unfeasiable for production use. In order to keep people from doing potential harmful things
sys_call_tableis no longer exported. This means, if you want to do something more than a mere @@ -2932,134 +2929,134 @@ patch and the README. Depending on your kernel version, you might even need to hand apply the patch.-
1/* -2 * syscall.c -3 * -4 * System call "stealing" sample. -5 * -6 * Disables page protection at a processor level by changing the 16th bit -7 * in the cr0 register (could be Intel specific). -8 * -9 * Based on example by Peter Jay Salzman and -10 * https://bbs.archlinux.org/viewtopic.php?id=139406 -11 */ ++1/* +2 * syscall.c +3 * +4 * System call "stealing" sample. +5 * +6 * Disables page protection at a processor level by changing the 16th bit +7 * in the cr0 register (could be Intel specific). +8 * +9 * Based on example by Peter Jay Salzman and +10 * https://bbs.archlinux.org/viewtopic.php?id=139406 +11 */ 12 -13#include <linux/delay.h> -14#include <linux/kernel.h> -15#include <linux/module.h> -16#include <linux/moduleparam.h> /* which will have params */ -17#include <linux/syscalls.h> -18#include <linux/unistd.h> /* The list of system calls */ +13#include <linux/delay.h> +14#include <linux/kernel.h> +15#include <linux/module.h> +16#include <linux/moduleparam.h> /* which will have params */ +17#include <linux/syscalls.h> +18#include <linux/unistd.h> /* The list of system calls */ 19 -20/* For the current (process) structure, we need this to know who the -21 * current user is. -22 */ -23#include <linux/sched.h> -24#include <linux/uaccess.h> +20/* For the current (process) structure, we need this to know who the +21 * current user is. +22 */ +23#include <linux/sched.h> +24#include <linux/uaccess.h> 25 -26unsigned long **sys_call_table; -27unsigned long original_cr0; +26unsigned long **sys_call_table; +27unsigned long original_cr0; 28 -29/* UID we want to spy on - will be filled from the command line. */ -30static int uid; -31module_param(uid, int, 0644); +29/* UID we want to spy on - will be filled from the command line. */ +30static int uid; +31module_param(uid, int, 0644); 32 -33/* A pointer to the original system call. The reason we keep this, rather -34 * than call the original function (sys_open), is because somebody else -35 * might have replaced the system call before us. Note that this is not -36 * 100% safe, because if another module replaced sys_open before us, -37 * then when we are inserted, we will call the function in that module - -38 * and it might be removed before we are. -39 * -40 * Another reason for this is that we can not get sys_open. -41 * It is a static variable, so it is not exported. -42 */ -43asmlinkage int (*original_call)(const char *, int, int); +33/* A pointer to the original system call. The reason we keep this, rather +34 * than call the original function (sys_open), is because somebody else +35 * might have replaced the system call before us. Note that this is not +36 * 100% safe, because if another module replaced sys_open before us, +37 * then when we are inserted, we will call the function in that module - +38 * and it might be removed before we are. +39 * +40 * Another reason for this is that we can not get sys_open. +41 * It is a static variable, so it is not exported. +42 */ +43asmlinkage int (*original_call)(const char *, int, int); 44 -45/* The function we will replace sys_open (the function called when you -46 * call the open system call) with. To find the exact prototype, with -47 * the number and type of arguments, we find the original function first -48 * (it is at fs/open.c). -49 * -50 * In theory, this means that we are tied to the current version of the -51 * kernel. In practice, the system calls almost never change (it would -52 * wreck havoc and require programs to be recompiled, since the system -53 * calls are the interface between the kernel and the processes). -54 */ -55asmlinkage int our_sys_open(const char *filename, int flags, int mode) +45/* The function we will replace sys_open (the function called when you +46 * call the open system call) with. To find the exact prototype, with +47 * the number and type of arguments, we find the original function first +48 * (it is at fs/open.c). +49 * +50 * In theory, this means that we are tied to the current version of the +51 * kernel. In practice, the system calls almost never change (it would +52 * wreck havoc and require programs to be recompiled, since the system +53 * calls are the interface between the kernel and the processes). +54 */ +55asmlinkage int our_sys_open(const char *filename, int flags, int mode) 56{ -57 int i = 0; -58 char ch; +57 int i = 0; +58 char ch; 59 -60 /* Report the file, if relevant */ -61 pr_info("Opened file by %d: ", uid); -62 do { +60 /* Report the file, if relevant */ +61 pr_info("Opened file by %d: ", uid); +62 do { 63 get_user(ch, filename + i); 64 i++; -65 pr_info("%c", ch); -66 } while (ch != 0); -67 pr_info("\n"); +65 pr_info("%c", ch); +66 } while (ch != 0); +67 pr_info("\n"); 68 -69 /* Call the original sys_open - otherwise, we lose the ability to -70 * open files. -71 */ -72 return original_call(filename, flags, mode); +69 /* Call the original sys_open - otherwise, we lose the ability to +70 * open files. +71 */ +72 return original_call(filename, flags, mode); 73} 74 -75static unsigned long **aquire_sys_call_table(void) +75static unsigned long **aquire_sys_call_table(void) 76{ -77 unsigned long int offset = PAGE_OFFSET; -78 unsigned long **sct; +77 unsigned long int offset = PAGE_OFFSET; +78 unsigned long **sct; 79 -80 while (offset < ULLONG_MAX) { -81 sct = (unsigned long **) offset; +80 while (offset < ULLONG_MAX) { +81 sct = (unsigned long **) offset; 82 -83 if (sct[__NR_close] == (unsigned long *) ksys_close) -84 return sct; +83 if (sct[__NR_close] == (unsigned long *) ksys_close) +84 return sct; 85 -86 offset += sizeof(void *); +86 offset += sizeof(void *); 87 } 88 -89 return NULL; +89 return NULL; 90} 91 -92static int __init syscall_start(void) +92static int __init syscall_start(void) 93{ -94 if (!(sys_call_table = aquire_sys_call_table())) -95 return -1; +94 if (!(sys_call_table = aquire_sys_call_table())) +95 return -1; 96 97 original_cr0 = read_cr0(); 98 99 write_cr0(original_cr0 & ~0x00010000); 100 -101 /* keep track of the original open function */ -102 original_call = (void *) sys_call_table[__NR_open]; +101 /* keep track of the original open function */ +102 original_call = (void *) sys_call_table[__NR_open]; 103 -104 /* use our open function instead */ -105 sys_call_table[__NR_open] = (unsigned long *) our_sys_open; +104 /* use our open function instead */ +105 sys_call_table[__NR_open] = (unsigned long *) our_sys_open; 106 107 write_cr0(original_cr0); 108 -109 pr_info("Spying on UID:%d\n", uid); +109 pr_info("Spying on UID:%d\n", uid); 110 -111 return 0; +111 return 0; 112} 113 -114static void __exit syscall_end(void) +114static void __exit syscall_end(void) 115{ -116 if (!sys_call_table) -117 return; +116 if (!sys_call_table) +117 return; 118 -119 /* Return the system call back to normal */ -120 if (sys_call_table[__NR_open] != (unsigned long *) our_sys_open) { -121 pr_alert("Somebody else also played with the "); -122 pr_alert("open system call\n"); -123 pr_alert("The system may be left in "); -124 pr_alert("an unstable state.\n"); +119 /* Return the system call back to normal */ +120 if (sys_call_table[__NR_open] != (unsigned long *) our_sys_open) { +121 pr_alert("Somebody else also played with the "); +122 pr_alert("open system call\n"); +123 pr_alert("The system may be left in "); +124 pr_alert("an unstable state.\n"); 125 } 126 127 write_cr0(original_cr0 & ~0x00010000); -128 sys_call_table[__NR_open] = (unsigned long *) original_call; +128 sys_call_table[__NR_open] = (unsigned long *) original_call; 129 write_cr0(original_cr0); 130 131 msleep(2000); @@ -3068,14 +3065,14 @@ hand apply the patch. 134module_init(syscall_start); 135module_exit(syscall_end); 136 -137MODULE_LICENSE("GPL");-
0.11 Blocking Processes and threads
-+
0.11.1 Sleep
-What do you do when somebody asks you for something you can not do right +
What do you do when somebody asks you for something you can not do right away? If you are a human being and you are bothered by a human being, the only thing you can say is: "Not right now, I’m busy. Go away!". But if you are a kernel module and you are bothered by a process, you have another @@ -3086,21 +3083,21 @@ processes are being put to sleep by the kernel and woken up all the time (that is the way multiple processes appear to run on the same time on a single CPU). -
This kernel module is an example of this. The file (called /proc/sleep) can only +
This kernel module is an example of this. The file (called /proc/sleep) can only be opened by a single process at a time. If the file is already open, the kernel module calls
wait_event_interruptible. The easiest way to keep a file open is to open it with:
1tail -f-This function changes the status of the task (a task is the kernel data structure +
This function changes the status of the task (a task is the kernel data structure which holds information about a process and the system call it is in, if any) to
TASK_INTERRUPTIBLE, which means that the task will not run until it is woken up somehow, and adds it to WaitQ, the queue of tasks waiting to access the file. Then, the function calls the scheduler to context switch to a different process, one which has some use for the CPU. -When a process is done with the file, it closes it, and +
When a process is done with the file, it closes it, and
module_closeis called. That function wakes up all the processes in the queue (there’s no mechanism to only wake up one of them). It then returns and the process which just @@ -3110,20 +3107,20 @@ Eventually, one of the processes which was in the queue will be given control of the CPU by the scheduler. It starts at the point right after the call tomodule_interruptible_sleep_on. -This means that the process is still in kernel mode - as far as the process +
This means that the process is still in kernel mode - as far as the process is concerned, it issued the open system call and the system call has not returned yet. The process does not know somebody else used the CPU for most of the time between the moment it issued the call and the moment it returned. -
It can then proceed to set a global variable to tell all the other processes that the +
It can then proceed to set a global variable to tell all the other processes that the file is still open and go on with its life. When the other processes get a piece of the CPU, they’ll see that global variable and go back to sleep. -
So we will use
tail -f +So we will use
tail -fto keep the file open in the background, while trying to access it with another process (again in the background, so that we need not switch to a different vt). As soon as the first background process is killed with kill %1 , the second is woken up, is able to access the file and finally terminates. -To make our life more interesting,
module_close +To make our life more interesting,
module_closedoes not have a monopoly on waking up the processes which wait to access the file. A signal, such as Ctrl +c (SIGINT) can also wake up a process. This is because we usedmodule_interruptible_sleep_on @@ -3133,11 +3130,11 @@ usedmodule_interruptible_sleep_oninstead, but that would have resulted in extremely angry users whose Ctrl+c’s are ignored. -In that case, we want to return with +
In that case, we want to return with
-EINTRimmediately. This is important so users can, for example, kill the process before it receives the file. -There is one more point to remember. Some times processes don’t want to sleep, they want +
There is one more point to remember. Some times processes don’t want to sleep, they want either to get what they want immediately, or to be told it cannot be done. Such processes use the
O_NONBLOCKflag when opening the file. The kernel is supposed to respond by returning with the error @@ -3173,533 +3170,534 @@ $ cat_nonblock /proc/sleep Last input: $+
-
1/* -2 * sleep.c - create a /proc file, and if several processes try to open it -3 * at the same time, put all but one to sleep. -4 */ +1/* +2 * sleep.c - create a /proc file, and if several processes try to open it +3 * at the same time, put all but one to sleep. +4 */ 5 -6#include <linux/kernel.h> /* We're doing kernel work */ -7#include <linux/module.h> /* Specifically, a module */ -8#include <linux/proc_fs.h> /* Necessary because we use proc fs */ -9#include <linux/sched.h> /* For putting processes to sleep and -10 waking them up */ -11#include <linux/uaccess.h> /* for get_user and put_user */ -12#include <linux/version.h> +6#include <linux/kernel.h> /* We're doing kernel work */ +7#include <linux/module.h> /* Specifically, a module */ +8#include <linux/proc_fs.h> /* Necessary because we use proc fs */ +9#include <linux/sched.h> /* For putting processes to sleep and +10 waking them up */ +11#include <linux/uaccess.h> /* for get_user and put_user */ +12#include <linux/version.h> 13 -14#if LINUX_VERSION_CODE >= KERNEL_VERSION(5, 6, 0) -15#define HAVE_PROC_OPS -16#endif +14#if LINUX_VERSION_CODE >= KERNEL_VERSION(5, 6, 0) +15#define HAVE_PROC_OPS +16#endif 17 -18/* Here we keep the last message received, to prove that we can process our -19 * input. -20 */ -21#define MESSAGE_LENGTH 80 -22static char Message[MESSAGE_LENGTH]; +18/* Here we keep the last message received, to prove that we can process our +19 * input. +20 */ +21#define MESSAGE_LENGTH 80 +22static char Message[MESSAGE_LENGTH]; 23 -24static struct proc_dir_entry *Our_Proc_File; -25#define PROC_ENTRY_FILENAME "sleep" +24static struct proc_dir_entry *Our_Proc_File; +25#define PROC_ENTRY_FILENAME "sleep" 26 -27/* Since we use the file operations struct, we can't use the special proc -28 * output provisions - we have to use a standard read function, which is this -29 * function. -30 */ -31static ssize_t module_output(struct file *file, /* see include/linux/fs.h */ -32 char *buf, /* The buffer to put data to -33 (in the user segment) */ -34 size_t len, /* The length of the buffer */ +27/* Since we use the file operations struct, we can't use the special proc +28 * output provisions - we have to use a standard read function, which is this +29 * function. +30 */ +31static ssize_t module_output(struct file *file, /* see include/linux/fs.h */ +32 char *buf, /* The buffer to put data to +33 (in the user segment) */ +34 size_t len, /* The length of the buffer */ 35 loff_t *offset) 36{ -37 static int finished = 0; -38 int i; -39 char message[MESSAGE_LENGTH + 30]; +37 static int finished = 0; +38 int i; +39 char message[MESSAGE_LENGTH + 30]; 40 -41 /* Return 0 to signify end of file - that we have nothing more to say -42 * at this point. -43 */ -44 if (finished) { +41 /* Return 0 to signify end of file - that we have nothing more to say +42 * at this point. +43 */ +44 if (finished) { 45 finished = 0; -46 return 0; +46 return 0; 47 } 48 -49 sprintf(message, "Last input:%s\n", Message); -50 for (i = 0; i < len && message[i]; i++) +49 sprintf(message, "Last input:%s\n", Message); +50 for (i = 0; i < len && message[i]; i++) 51 put_user(message[i], buf + i); 52 53 finished = 1; -54 return i; /* Return the number of bytes "read" */ +54 return i; /* Return the number of bytes "read" */ 55} 56 -57/* This function receives input from the user when the user writes to the -58 * /proc file. -59 */ -60static ssize_t module_input(struct file *file, /* The file itself */ -61 const char *buf, /* The buffer with input */ -62 size_t length, /* The buffer's length */ -63 loff_t *offset) /* offset to file - ignore */ +57/* This function receives input from the user when the user writes to the +58 * /proc file. +59 */ +60static ssize_t module_input(struct file *file, /* The file itself */ +61 const char *buf, /* The buffer with input */ +62 size_t length, /* The buffer's length */ +63 loff_t *offset) /* offset to file - ignore */ 64{ -65 int i; +65 int i; 66 -67 /* Put the input into Message, where module_output will later be able -68 * to use it. -69 */ -70 for (i = 0; i < MESSAGE_LENGTH - 1 && i < length; i++) +67 /* Put the input into Message, where module_output will later be able +68 * to use it. +69 */ +70 for (i = 0; i < MESSAGE_LENGTH - 1 && i < length; i++) 71 get_user(Message[i], buf + i); -72 /* we want a standard, zero terminated string */ -73 Message[i] = '\0'; +72 /* we want a standard, zero terminated string */ +73 Message[i] = '\0'; 74 -75 /* We need to return the number of input characters used */ -76 return i; +75 /* We need to return the number of input characters used */ +76 return i; 77} 78 -79/* 1 if the file is currently open by somebody */ -80int Already_Open = 0; +79/* 1 if the file is currently open by somebody */ +80int Already_Open = 0; 81 -82/* Queue of processes who want our file */ +82/* Queue of processes who want our file */ 83DECLARE_WAIT_QUEUE_HEAD(WaitQ); 84 -85/* Called when the /proc file is opened */ -86static int module_open(struct inode *inode, struct file *file) +85/* Called when the /proc file is opened */ +86static int module_open(struct inode *inode, struct file *file) 87{ -88 /* If the file's flags include O_NONBLOCK, it means the process does not -89 * want to wait for the file. In this case, if the file is already open, -90 * we should fail with -EAGAIN, meaning "you will have to try again", -91 * instead of blocking a process which would rather stay awake. -92 */ -93 if ((file->f_flags & O_NONBLOCK) && Already_Open) -94 return -EAGAIN; +88 /* If the file's flags include O_NONBLOCK, it means the process does not +89 * want to wait for the file. In this case, if the file is already open, +90 * we should fail with -EAGAIN, meaning "you will have to try again", +91 * instead of blocking a process which would rather stay awake. +92 */ +93 if ((file->f_flags & O_NONBLOCK) && Already_Open) +94 return -EAGAIN; 95 -96 /* This is the correct place for try_module_get(THIS_MODULE) because if -97 * a process is in the loop, which is within the kernel module, -98 * the kernel module must not be removed. -99 */ +96 /* This is the correct place for try_module_get(THIS_MODULE) because if +97 * a process is in the loop, which is within the kernel module, +98 * the kernel module must not be removed. +99 */ 100 try_module_get(THIS_MODULE); 101 -102 /* If the file is already open, wait until it is not. */ -103 while (Already_Open) { -104 int i, is_sig = 0; +102 /* If the file is already open, wait until it is not. */ +103 while (Already_Open) { +104 int i, is_sig = 0; 105 -106 /* This function puts the current process, including any system -107 * calls, such as us, to sleep. Execution will be resumed right -108 * after the function call, either because somebody called -109 * wake_up(&WaitQ) (only module_close does that, when the file -110 * is closed) or when a signal, such as Ctrl-C, is sent -111 * to the process -112 */ +106 /* This function puts the current process, including any system +107 * calls, such as us, to sleep. Execution will be resumed right +108 * after the function call, either because somebody called +109 * wake_up(&WaitQ) (only module_close does that, when the file +110 * is closed) or when a signal, such as Ctrl-C, is sent +111 * to the process +112 */ 113 wait_event_interruptible(WaitQ, !Already_Open); 114 -115 /* If we woke up because we got a signal we're not blocking, -116 * return -EINTR (fail the system call). This allows processes -117 * to be killed or stopped. -118 */ -119 for (i = 0; i < _NSIG_WORDS && !is_sig; i++) +115 /* If we woke up because we got a signal we're not blocking, +116 * return -EINTR (fail the system call). This allows processes +117 * to be killed or stopped. +118 */ +119 for (i = 0; i < _NSIG_WORDS && !is_sig; i++) 120 is_sig = current->pending.signal.sig[i] & ~current->blocked.sig[i]; 121 -122 if (is_sig) { -123 /* It is important to put module_put(THIS_MODULE) here, because -124 * for processes where the open is interrupted there will never -125 * be a corresponding close. If we do not decrement the usage -126 * count here, we will be left with a positive usage count -127 * which we will have no way to bring down to zero, giving us -128 * an immortal module, which can only be killed by rebooting -129 * the machine. -130 */ +122 if (is_sig) { +123 /* It is important to put module_put(THIS_MODULE) here, because +124 * for processes where the open is interrupted there will never +125 * be a corresponding close. If we do not decrement the usage +126 * count here, we will be left with a positive usage count +127 * which we will have no way to bring down to zero, giving us +128 * an immortal module, which can only be killed by rebooting +129 * the machine. +130 */ 131 module_put(THIS_MODULE); -132 return -EINTR; +132 return -EINTR; 133 } 134 } 135 -136 /* If we got here, Already_Open must be zero. */ +136 /* If we got here, Already_Open must be zero. */ 137 -138 /* Open the file */ +138 /* Open the file */ 139 Already_Open = 1; -140 return 0; /* Allow the access */ +140 return 0; /* Allow the access */ 141} 142 -143/* Called when the /proc file is closed */ -144int module_close(struct inode *inode, struct file *file) +143/* Called when the /proc file is closed */ +144int module_close(struct inode *inode, struct file *file) 145{ -146 /* Set Already_Open to zero, so one of the processes in the WaitQ will -147 * be able to set Already_Open back to one and to open the file. All -148 * the other processes will be called when Already_Open is back to one, -149 * so they'll go back to sleep. -150 */ +146 /* Set Already_Open to zero, so one of the processes in the WaitQ will +147 * be able to set Already_Open back to one and to open the file. All +148 * the other processes will be called when Already_Open is back to one, +149 * so they'll go back to sleep. +150 */ 151 Already_Open = 0; 152 -153 /* Wake up all the processes in WaitQ, so if anybody is waiting for the -154 * file, they can have it. -155 */ +153 /* Wake up all the processes in WaitQ, so if anybody is waiting for the +154 * file, they can have it. +155 */ 156 wake_up(&WaitQ); 157 158 module_put(THIS_MODULE); 159 -160 return 0; /* success */ +160 return 0; /* success */ 161} 162 -163/* Structures to register as the /proc file, with pointers to all the relevant -164 * functions. -165 */ +163/* Structures to register as the /proc file, with pointers to all the relevant +164 * functions. +165 */ 166 -167/* File operations for our proc file. This is where we place pointers to all -168 * the functions called when somebody tries to do something to our file. NULL -169 * means we don't want to deal with something. -170 */ -171#ifdef HAVE_PROC_OPS -172static const struct proc_ops File_Ops_4_Our_Proc_File = { -173 .proc_read = module_output, /* "read" from the file */ -174 .proc_write = module_input, /* "write" to the file */ -175 .proc_open = module_open, /* called when the /proc file is opened */ -176 .proc_release = module_close, /* called when it's closed */ +167/* File operations for our proc file. This is where we place pointers to all +168 * the functions called when somebody tries to do something to our file. NULL +169 * means we don't want to deal with something. +170 */ +171#ifdef HAVE_PROC_OPS +172static const struct proc_ops File_Ops_4_Our_Proc_File = { +173 .proc_read = module_output, /* "read" from the file */ +174 .proc_write = module_input, /* "write" to the file */ +175 .proc_open = module_open, /* called when the /proc file is opened */ +176 .proc_release = module_close, /* called when it's closed */ 177}; -178#else -179static const struct file_operations File_Ops_4_Our_Proc_File = { +178#else +179static const struct file_operations File_Ops_4_Our_Proc_File = { 180 .read = module_output, 181 .write = module_input, 182 .open = module_open, 183 .release = module_close, 184}; -185#endif +185#endif 186 -187/* Initialize the module - register the proc file */ -188static int __init sleep_init(void) +187/* Initialize the module - register the proc file */ +188static int __init sleep_init(void) 189{ 190 Our_Proc_File = 191 proc_create(PROC_ENTRY_FILENAME, 0644, NULL, &File_Ops_4_Our_Proc_File); -192 if (Our_Proc_File == NULL) { +192 if (Our_Proc_File == NULL) { 193 remove_proc_entry(PROC_ENTRY_FILENAME, NULL); -194 pr_debug("Error: Could not initialize /proc/%s\n", PROC_ENTRY_FILENAME); -195 return -ENOMEM; +194 pr_debug("Error: Could not initialize /proc/%s\n", PROC_ENTRY_FILENAME); +195 return -ENOMEM; 196 } 197 proc_set_size(Our_Proc_File, 80); 198 proc_set_user(Our_Proc_File, GLOBAL_ROOT_UID, GLOBAL_ROOT_GID); 199 -200 pr_info("/proc/test created\n"); +200 pr_info("/proc/test created\n"); 201 -202 return 0; +202 return 0; 203} 204 -205/* Cleanup - unregister our file from /proc. This could get dangerous if -206 * there are still processes waiting in WaitQ, because they are inside our -207 * open function, which will get unloaded. I'll explain how to avoid removal -208 * of a kernel module in such a case in chapter 10. -209 */ -210static void __exit sleep_exit(void) +205/* Cleanup - unregister our file from /proc. This could get dangerous if +206 * there are still processes waiting in WaitQ, because they are inside our +207 * open function, which will get unloaded. I'll explain how to avoid removal +208 * of a kernel module in such a case in chapter 10. +209 */ +210static void __exit sleep_exit(void) 211{ 212 remove_proc_entry(PROC_ENTRY_FILENAME, NULL); -213 pr_debug("/proc/%s removed\n", PROC_ENTRY_FILENAME); +213 pr_debug("/proc/%s removed\n", PROC_ENTRY_FILENAME); 214} 215 216module_init(sleep_init); 217module_exit(sleep_exit); 218 -219MODULE_LICENSE("GPL");+219MODULE_LICENSE("GPL");-
1/* -2 * cat_nonblock.c - open a file and display its contents, but exit rather than -3 * wait for input. -4 */ -5#include <errno.h> /* for errno */ -6#include <fcntl.h> /* for open */ -7#include <stdio.h> /* standard I/O */ -8#include <stdlib.h> /* for exit */ -9#include <unistd.h> /* for read */ +-1/* +2 * cat_nonblock.c - open a file and display its contents, but exit rather than +3 * wait for input. +4 */ +5#include <errno.h> /* for errno */ +6#include <fcntl.h> /* for open */ +7#include <stdio.h> /* standard I/O */ +8#include <stdlib.h> /* for exit */ +9#include <unistd.h> /* for read */ 10 -11#define MAX_BYTES 1024 * 4 +11#define MAX_BYTES 1024 * 4 12 -13int main(int argc, char *argv[]) +13int main(int argc, char *argv[]) 14{ -15 int fd; /* The file descriptor for the file to read */ -16 size_t bytes; /* The number of bytes read */ -17 char buffer[MAX_BYTES]; /* The buffer for the bytes */ +15 int fd; /* The file descriptor for the file to read */ +16 size_t bytes; /* The number of bytes read */ +17 char buffer[MAX_BYTES]; /* The buffer for the bytes */ 18 -19 /* Usage */ -20 if (argc != 2) { -21 printf("Usage: %s <filename>\n", argv[0]); -22 puts("Reads the content of a file, but doesn't wait for input"); +19 /* Usage */ +20 if (argc != 2) { +21 printf("Usage: %s <filename>\n", argv[0]); +22 puts("Reads the content of a file, but doesn't wait for input"); 23 exit(-1); 24 } 25 -26 /* Open the file for reading in non blocking mode */ +26 /* Open the file for reading in non blocking mode */ 27 fd = open(argv[1], O_RDONLY | O_NONBLOCK); 28 -29 /* If open failed */ -30 if (fd == -1) { -31 puts(errno == EAGAIN ? "Open would block" : "Open failed"); +29 /* If open failed */ +30 if (fd == -1) { +31 puts(errno == EAGAIN ? "Open would block" : "Open failed"); 32 exit(-1); 33 } 34 -35 /* Read the file and output its contents */ -36 do { -37 /* Read characters from the file */ +35 /* Read the file and output its contents */ +36 do { +37 /* Read characters from the file */ 38 bytes = read(fd, buffer, MAX_BYTES); 39 -40 /* If there's an error, report it and die */ -41 if (bytes == -1) { -42 if (errno = EAGAIN) -43 puts("Normally I'd block, but you told me not to"); -44 else -45 puts("Another read error"); +40 /* If there's an error, report it and die */ +41 if (bytes == -1) { +42 if (errno = EAGAIN) +43 puts("Normally I'd block, but you told me not to"); +44 else +45 puts("Another read error"); 46 exit(-1); 47 } 48 -49 /* Print the characters */ -50 if (bytes > 0) { -51 for (int i = 0; i < bytes; i++) +49 /* Print the characters */ +50 if (bytes > 0) { +51 for (int i = 0; i < bytes; i++) 52 putchar(buffer[i]); 53 } 54 -55 /* While there are no errors and the file isn't over */ -56 } while (bytes > 0); +55 /* While there are no errors and the file isn't over */ +56 } while (bytes > 0); 57 -58 return 0; +58 return 0; 59}-+
0.11.2 Completions
-Sometimes one thing should happen before another within a module having multiple threads. +
Sometimes one thing should happen before another within a module having multiple threads. Rather than using
/bin/sleepcommands, the kernel has another way to do this which allows timeouts or interrupts to also happen. -In the following example two threads are started, but one needs to start before +
In the following example two threads are started, but one needs to start before another.
-
1/* -2 * completions.c -3 */ -4#include <linux/completion.h> -5#include <linux/init.h> -6#include <linux/kernel.h> -7#include <linux/kthread.h> -8#include <linux/module.h> ++1/* +2 * completions.c +3 */ +4#include <linux/completion.h> +5#include <linux/init.h> +6#include <linux/kernel.h> +7#include <linux/kthread.h> +8#include <linux/module.h> 9 -10static struct { -11 struct completion crank_comp; -12 struct completion flywheel_comp; +10static struct { +11 struct completion crank_comp; +12 struct completion flywheel_comp; 13} machine; 14 -15static int machine_crank_thread(void *arg) +15static int machine_crank_thread(void *arg) 16{ -17 pr_info("Turn the crank\n"); +17 pr_info("Turn the crank\n"); 18 19 complete_all(&machine.crank_comp); 20 complete_and_exit(&machine.crank_comp, 0); 21} 22 -23static int machine_flywheel_spinup_thread(void *arg) +23static int machine_flywheel_spinup_thread(void *arg) 24{ 25 wait_for_completion(&machine.crank_comp); 26 -27 pr_info("Flywheel spins up\n"); +27 pr_info("Flywheel spins up\n"); 28 29 complete_all(&machine.flywheel_comp); 30 complete_and_exit(&machine.flywheel_comp, 0); 31} 32 -33static int completions_init(void) +33static int completions_init(void) 34{ -35 struct task_struct *crank_thread; -36 struct task_struct *flywheel_thread; +35 struct task_struct *crank_thread; +36 struct task_struct *flywheel_thread; 37 -38 pr_info("completions example\n"); +38 pr_info("completions example\n"); 39 40 init_completion(&machine.crank_comp); 41 init_completion(&machine.flywheel_comp); 42 -43 crank_thread = kthread_create(machine_crank_thread, NULL, "KThread Crank"); -44 if (IS_ERR(crank_thread)) -45 goto ERROR_THREAD_1; +43 crank_thread = kthread_create(machine_crank_thread, NULL, "KThread Crank"); +44 if (IS_ERR(crank_thread)) +45 goto ERROR_THREAD_1; 46 47 flywheel_thread = kthread_create(machine_flywheel_spinup_thread, NULL, -48 "KThread Flywheel"); -49 if (IS_ERR(flywheel_thread)) -50 goto ERROR_THREAD_2; +48 "KThread Flywheel"); +49 if (IS_ERR(flywheel_thread)) +50 goto ERROR_THREAD_2; 51 52 wake_up_process(flywheel_thread); 53 wake_up_process(crank_thread); 54 -55 return 0; +55 return 0; 56 57ERROR_THREAD_2: 58 kthread_stop(crank_thread); 59ERROR_THREAD_1: 60 -61 return -1; +61 return -1; 62} 63 -64void completions_exit(void) +64void completions_exit(void) 65{ 66 wait_for_completion(&machine.crank_comp); 67 wait_for_completion(&machine.flywheel_comp); 68 -69 pr_info("completions exit\n"); +69 pr_info("completions exit\n"); 70} 71 72module_init(completions_init); 73module_exit(completions_exit); 74 -75MODULE_DESCRIPTION("Completions example"); -76MODULE_LICENSE("GPL");-The machine structure stores the completion states for the two threads. At the -exit point of each thread the respective completion state is updated, and +75MODULE_DESCRIPTION("Completions example"); +76MODULE_LICENSE("GPL");
The
machine +structure stores the completion states for the two threads. At the exit +point of each thread the respective completion state is updated, andwait_for_completionis used by the flywheel thread to ensure that it does not begin prematurely. -So even though
flywheel_thread +So even though
flywheel_threadis started first you should notice if you load this module and rundmesgthat turning the crank always happens first because the flywheel thread waits for it to complete. -There are other variations upon the +
There are other variations upon the
wait_for_completionfunction, which include timeouts or being interrupted, but this basic mechanism is enough for many common situations without adding a lot of complexity. -+
0.12 Avoiding Collisions and Deadlocks
-If processes running on different CPUs or in different threads try to access the same +
If processes running on different CPUs or in different threads try to access the same memory, then it is possible that strange things can happen or your system can lock up. To avoid this, various types of mutual exclusion kernel functions are available. These indicate if a section of code is "locked" or "unlocked" so that simultaneous attempts to run it can not happen.
0.12.1 Mutex
-You can use kernel mutexes (mutual exclusions) in much the same manner that you +
You can use kernel mutexes (mutual exclusions) in much the same manner that you might deploy them in userland. This may be all that is needed to avoid collisions in most cases.
-
1/* -2 * example_mutex.c -3 */ -4#include <linux/init.h> -5#include <linux/kernel.h> -6#include <linux/module.h> -7#include <linux/mutex.h> ++1/* +2 * example_mutex.c +3 */ +4#include <linux/init.h> +5#include <linux/kernel.h> +6#include <linux/module.h> +7#include <linux/mutex.h> 8 9DEFINE_MUTEX(mymutex); 10 -11static int example_mutex_init(void) +11static int example_mutex_init(void) 12{ -13 int ret; +13 int ret; 14 -15 pr_info("example_mutex init\n"); +15 pr_info("example_mutex init\n"); 16 17 ret = mutex_trylock(&mymutex); -18 if (ret != 0) { -19 pr_info("mutex is locked\n"); +18 if (ret != 0) { +19 pr_info("mutex is locked\n"); 20 -21 if (mutex_is_locked(&mymutex) == 0) -22 pr_info("The mutex failed to lock!\n"); +21 if (mutex_is_locked(&mymutex) == 0) +22 pr_info("The mutex failed to lock!\n"); 23 24 mutex_unlock(&mymutex); -25 pr_info("mutex is unlocked\n"); -26 } else -27 pr_info("Failed to lock\n"); +25 pr_info("mutex is unlocked\n"); +26 } else +27 pr_info("Failed to lock\n"); 28 -29 return 0; +29 return 0; 30} 31 -32static void example_mutex_exit(void) +32static void example_mutex_exit(void) 33{ -34 pr_info("example_mutex exit\n"); +34 pr_info("example_mutex exit\n"); 35} 36 37module_init(example_mutex_init); 38module_exit(example_mutex_exit); 39 -40MODULE_DESCRIPTION("Mutex example"); -41MODULE_LICENSE("GPL");-+40MODULE_DESCRIPTION("Mutex example"); +41MODULE_LICENSE("GPL");
0.12.2 Spinlocks
-As the name suggests, spinlocks lock up the CPU that the code is running on, +
As the name suggests, spinlocks lock up the CPU that the code is running on, taking 100% of its resources. Because of this you should only use the spinlock -mechanism around code which is likely to take no more than a few milliseconds to +mechanism around code which is likely to take no more than a few milliseconds to run and so will not noticably slow anything down from the user’s point of view. -
The example here is "irq safe" in that if interrupts happen during the lock then +
The example here is "irq safe" in that if interrupts happen during the lock then they will not be forgotten and will activate when the unlock happens, using the
flagsvariable to retain their state.-
1/* -2 * example_spinlock.c -3 */ -4#include <linux/init.h> -5#include <linux/interrupt.h> -6#include <linux/kernel.h> -7#include <linux/module.h> -8#include <linux/spinlock.h> ++1/* +2 * example_spinlock.c +3 */ +4#include <linux/init.h> +5#include <linux/interrupt.h> +6#include <linux/kernel.h> +7#include <linux/module.h> +8#include <linux/spinlock.h> 9 10DEFINE_SPINLOCK(sl_static); 11spinlock_t sl_dynamic; 12 -13static void example_spinlock_static(void) +13static void example_spinlock_static(void) 14{ -15 unsigned long flags; +15 unsigned long flags; 16 17 spin_lock_irqsave(&sl_static, flags); -18 pr_info("Locked static spinlock\n"); +18 pr_info("Locked static spinlock\n"); 19 -20 /* Do something or other safely. Because this uses 100% CPU time, this -21 * code should take no more than a few milliseconds to run. -22 */ +20 /* Do something or other safely. Because this uses 100% CPU time, this +21 * code should take no more than a few milliseconds to run. +22 */ 23 24 spin_unlock_irqrestore(&sl_static, flags); -25 pr_info("Unlocked static spinlock\n"); +25 pr_info("Unlocked static spinlock\n"); 26} 27 -28static void example_spinlock_dynamic(void) +28static void example_spinlock_dynamic(void) 29{ -30 unsigned long flags; +30 unsigned long flags; 31 32 spin_lock_init(&sl_dynamic); 33 spin_lock_irqsave(&sl_dynamic, flags); -34 pr_info("Locked dynamic spinlock\n"); +34 pr_info("Locked dynamic spinlock\n"); 35 -36 /* Do something or other safely. Because this uses 100% CPU time, this -37 * code should take no more than a few milliseconds to run. -38 */ +36 /* Do something or other safely. Because this uses 100% CPU time, this +37 * code should take no more than a few milliseconds to run. +38 */ 39 40 spin_unlock_irqrestore(&sl_dynamic, flags); -41 pr_info("Unlocked dynamic spinlock\n"); +41 pr_info("Unlocked dynamic spinlock\n"); 42} 43 -44static int example_spinlock_init(void) +44static int example_spinlock_init(void) 45{ -46 pr_info("example spinlock started\n"); +46 pr_info("example spinlock started\n"); 47 48 example_spinlock_static(); 49 example_spinlock_dynamic(); 50 -51 return 0; +51 return 0; 52} 53 -54static void example_spinlock_exit(void) +54static void example_spinlock_exit(void) 55{ -56 pr_info("example spinlock exit\n"); +56 pr_info("example spinlock exit\n"); 57} 58 59module_init(example_spinlock_init); 60module_exit(example_spinlock_exit); 61 -62MODULE_DESCRIPTION("Spinlock example"); -63MODULE_LICENSE("GPL");-+62MODULE_DESCRIPTION("Spinlock example"); +63MODULE_LICENSE("GPL");
0.12.3 Read and write locks
-Read and write locks are specialised kinds of spinlocks so that you can exclusively +
Read and write locks are specialised kinds of spinlocks so that you can exclusively read from something or write to something. Like the earlier spinlocks example the one below shows an "irq safe" situation in which if other functions were triggered from irqs which might also read and write to whatever you are concerned with @@ -3709,69 +3707,69 @@ the system and cause users to start revolting against the tyranny of your module.
-
1/* -2 * example_rwlock.c -3 */ -4#include <linux/interrupt.h> -5#include <linux/kernel.h> -6#include <linux/module.h> ++1/* +2 * example_rwlock.c +3 */ +4#include <linux/interrupt.h> +5#include <linux/kernel.h> +6#include <linux/module.h> 7 8DEFINE_RWLOCK(myrwlock); 9 -10static void example_read_lock(void) +10static void example_read_lock(void) 11{ -12 unsigned long flags; +12 unsigned long flags; 13 14 read_lock_irqsave(&myrwlock, flags); -15 pr_info("Read Locked\n"); +15 pr_info("Read Locked\n"); 16 -17 /* Read from something */ +17 /* Read from something */ 18 19 read_unlock_irqrestore(&myrwlock, flags); -20 pr_info("Read Unlocked\n"); +20 pr_info("Read Unlocked\n"); 21} 22 -23static void example_write_lock(void) +23static void example_write_lock(void) 24{ -25 unsigned long flags; +25 unsigned long flags; 26 27 write_lock_irqsave(&myrwlock, flags); -28 pr_info("Write Locked\n"); +28 pr_info("Write Locked\n"); 29 -30 /* Write to something */ +30 /* Write to something */ 31 32 write_unlock_irqrestore(&myrwlock, flags); -33 pr_info("Write Unlocked\n"); +33 pr_info("Write Unlocked\n"); 34} 35 -36static int example_rwlock_init(void) +36static int example_rwlock_init(void) 37{ -38 pr_info("example_rwlock started\n"); +38 pr_info("example_rwlock started\n"); 39 40 example_read_lock(); 41 example_write_lock(); 42 -43 return 0; +43 return 0; 44} 45 -46static void example_rwlock_exit(void) +46static void example_rwlock_exit(void) 47{ -48 pr_info("example_rwlock exit\n"); +48 pr_info("example_rwlock exit\n"); 49} 50 51module_init(example_rwlock_init); 52module_exit(example_rwlock_exit); 53 -54MODULE_DESCRIPTION("Read/Write locks example"); -55MODULE_LICENSE("GPL");-Of course, if you know for sure that there are no functions triggered by irqs +54MODULE_DESCRIPTION("Read/Write locks example"); +55MODULE_LICENSE("GPL");
Of course, if you know for sure that there are no functions triggered by irqs which could possibly interfere with your logic then you can use the simpler
read_lock(&myrwlock)andread_unlock(&myrwlock)or the corresponding write functions.0.12.4 Atomic operations
-If you are doing simple arithmetic: adding, subtracting or bitwise operations then +
If you are doing simple arithmetic: adding, subtracting or bitwise operations then there is another way in the multi-CPU and multi-hyperthreaded world to stop other parts of the system from messing with your mojo. By using atomic operations you can be confident that your addition, subtraction or bit flip did actually happen @@ -3779,236 +3777,236 @@ and was not overwritten by some other shenanigans. An example is shown below.
-
1/* -2 * example_atomic.c -3 */ -4#include <linux/interrupt.h> -5#include <linux/kernel.h> -6#include <linux/module.h> +-1/* +2 * example_atomic.c +3 */ +4#include <linux/interrupt.h> +5#include <linux/kernel.h> +6#include <linux/module.h> 7 -8#define BYTE_TO_BINARY_PATTERN "%c%c%c%c%c%c%c%c" -9#define BYTE_TO_BINARY(byte) \ -10 (byte & 0x80 ? '1' : '0'), (byte & 0x40 ? '1' : '0'), \ -11 (byte & 0x20 ? '1' : '0'), (byte & 0x10 ? '1' : '0'), \ -12 (byte & 0x08 ? '1' : '0'), (byte & 0x04 ? '1' : '0'), \ -13 (byte & 0x02 ? '1' : '0'), (byte & 0x01 ? '1' : '0') +8#define BYTE_TO_BINARY_PATTERN "%c%c%c%c%c%c%c%c" +9#define BYTE_TO_BINARY(byte) \ +10 (byte & 0x80 ? '1' : '0'), (byte & 0x40 ? '1' : '0'), \ +11 (byte & 0x20 ? '1' : '0'), (byte & 0x10 ? '1' : '0'), \ +12 (byte & 0x08 ? '1' : '0'), (byte & 0x04 ? '1' : '0'), \ +13 (byte & 0x02 ? '1' : '0'), (byte & 0x01 ? '1' : '0') 14 -15static void atomic_add_subtract(void) +15static void atomic_add_subtract(void) 16{ 17 atomic_t debbie; 18 atomic_t chris = ATOMIC_INIT(50); 19 20 atomic_set(&debbie, 45); 21 -22 /* subtract one */ +22 /* subtract one */ 23 atomic_dec(&debbie); 24 25 atomic_add(7, &debbie); 26 -27 /* add one */ +27 /* add one */ 28 atomic_inc(&debbie); 29 -30 pr_info("chris: %d, debbie: %d\n", atomic_read(&chris), +30 pr_info("chris: %d, debbie: %d\n", atomic_read(&chris), 31 atomic_read(&debbie)); 32} 33 -34static void atomic_bitwise(void) +34static void atomic_bitwise(void) 35{ -36 unsigned long word = 0; +36 unsigned long word = 0; 37 -38 pr_info("Bits 0: " BYTE_TO_BINARY_PATTERN, BYTE_TO_BINARY(word)); +38 pr_info("Bits 0: " BYTE_TO_BINARY_PATTERN, BYTE_TO_BINARY(word)); 39 set_bit(3, &word); 40 set_bit(5, &word); -41 pr_info("Bits 1: " BYTE_TO_BINARY_PATTERN, BYTE_TO_BINARY(word)); +41 pr_info("Bits 1: " BYTE_TO_BINARY_PATTERN, BYTE_TO_BINARY(word)); 42 clear_bit(5, &word); -43 pr_info("Bits 2: " BYTE_TO_BINARY_PATTERN, BYTE_TO_BINARY(word)); +43 pr_info("Bits 2: " BYTE_TO_BINARY_PATTERN, BYTE_TO_BINARY(word)); 44 change_bit(3, &word); 45 -46 pr_info("Bits 3: " BYTE_TO_BINARY_PATTERN, BYTE_TO_BINARY(word)); -47 if (test_and_set_bit(3, &word)) -48 pr_info("wrong\n"); -49 pr_info("Bits 4: " BYTE_TO_BINARY_PATTERN, BYTE_TO_BINARY(word)); +46 pr_info("Bits 3: " BYTE_TO_BINARY_PATTERN, BYTE_TO_BINARY(word)); +47 if (test_and_set_bit(3, &word)) +48 pr_info("wrong\n"); +49 pr_info("Bits 4: " BYTE_TO_BINARY_PATTERN, BYTE_TO_BINARY(word)); 50 51 word = 255; -52 pr_info("Bits 5: " BYTE_TO_BINARY_PATTERN, BYTE_TO_BINARY(word)); +52 pr_info("Bits 5: " BYTE_TO_BINARY_PATTERN, BYTE_TO_BINARY(word)); 53} 54 -55static int example_atomic_init(void) +55static int example_atomic_init(void) 56{ -57 pr_info("example_atomic started\n"); +57 pr_info("example_atomic started\n"); 58 59 atomic_add_subtract(); 60 atomic_bitwise(); 61 -62 return 0; +62 return 0; 63} 64 -65static void example_atomic_exit(void) +65static void example_atomic_exit(void) 66{ -67 pr_info("example_atomic exit\n"); +67 pr_info("example_atomic exit\n"); 68} 69 70module_init(example_atomic_init); 71module_exit(example_atomic_exit); 72 -73MODULE_DESCRIPTION("Atomic operations example"); -74MODULE_LICENSE("GPL");+73MODULE_DESCRIPTION("Atomic operations example"); +74MODULE_LICENSE("GPL");+
0.13 Replacing Print Macros
-+
0.13.1 Replacement
-In Section 1.2.1.2, I said that X and kernel module programming do not mix. That is +
In Section 1.2.1.2, I said that X and kernel module programming do not mix. That is true for developing kernel modules, but in actual use, you want to be able to send messages to whichever tty the command to load the module came from. -
"tty" is an abbreviation of teletype: originally a combination keyboard-printer +
"tty" is an abbreviation of teletype: originally a combination keyboard-printer used to communicate with a Unix system, and today an abstraction for the text stream used for a Unix program, whether it is a physical terminal, an xterm on an X display, a network connection used with ssh, etc. -
The way this is done is by using current, a pointer to the currently running task, +
The way this is done is by using current, a pointer to the currently running task, to get the current task’s tty structure. Then, we look inside that tty structure to find a pointer to a string write function, which we use to write a string to the tty.
-
1/* -2 * print_string.c - Send output to the tty we're running on, regardless if -3 * it is through X11, telnet, etc. We do this by printing the string to the -4 * tty associated with the current task. -5 */ -6#include <linux/init.h> -7#include <linux/kernel.h> -8#include <linux/module.h> -9#include <linux/sched.h> /* For current */ -10#include <linux/tty.h> /* For the tty declarations */ ++1/* +2 * print_string.c - Send output to the tty we're running on, regardless if +3 * it is through X11, telnet, etc. We do this by printing the string to the +4 * tty associated with the current task. +5 */ +6#include <linux/init.h> +7#include <linux/kernel.h> +8#include <linux/module.h> +9#include <linux/sched.h> /* For current */ +10#include <linux/tty.h> /* For the tty declarations */ 11 -12static void print_string(char *str) +12static void print_string(char *str) 13{ -14 struct tty_struct *my_tty; -15 const struct tty_operations *ttyops; +14 struct tty_struct *my_tty; +15 const struct tty_operations *ttyops; 16 -17 /* The tty for the current task, for 2.6.6+ kernels */ +17 /* The tty for the current task, for 2.6.6+ kernels */ 18 my_tty = get_current_tty(); 19 ttyops = my_tty->driver->ops; 20 -21 /* If my_tty is NULL, the current task has no tty you can print to (i.e., -22 * if it is a daemon). If so, there is nothing we can do. -23 */ -24 if (my_tty) { -25 /* my_tty->driver is a struct which holds the tty's functions, -26 * one of which (write) is used to write strings to the tty. -27 * It can be used to take a string either from the user's or -28 * kernel's memory segment. -29 * -30 * The function's 1st parameter is the tty to write to, because the -31 * same function would normally be used for all tty's of a certain -32 * type. -33 * The 2nd parameter is a pointer to a string. -34 * The 3rd parameter is the length of the string. -35 * -36 * As you will see below, sometimes it's necessary to use -37 * preprocessor stuff to create code that works for different -38 * kernel versions. The (naive) approach we've taken here does not -39 * scale well. The right way to deal with this is described in -40 * section 2 of -41 * linux/Documentation/SubmittingPatches -42 */ -43 (ttyops->write)(my_tty, /* The tty itself */ -44 str, /* String */ -45 strlen(str)); /* Length */ +21 /* If my_tty is NULL, the current task has no tty you can print to (i.e., +22 * if it is a daemon). If so, there is nothing we can do. +23 */ +24 if (my_tty) { +25 /* my_tty->driver is a struct which holds the tty's functions, +26 * one of which (write) is used to write strings to the tty. +27 * It can be used to take a string either from the user's or +28 * kernel's memory segment. +29 * +30 * The function's 1st parameter is the tty to write to, because the +31 * same function would normally be used for all tty's of a certain +32 * type. +33 * The 2nd parameter is a pointer to a string. +34 * The 3rd parameter is the length of the string. +35 * +36 * As you will see below, sometimes it's necessary to use +37 * preprocessor stuff to create code that works for different +38 * kernel versions. The (naive) approach we've taken here does not +39 * scale well. The right way to deal with this is described in +40 * section 2 of +41 * linux/Documentation/SubmittingPatches +42 */ +43 (ttyops->write)(my_tty, /* The tty itself */ +44 str, /* String */ +45 strlen(str)); /* Length */ 46 -47 /* ttys were originally hardware devices, which (usually) strictly -48 * followed the ASCII standard. In ASCII, to move to a new line you -49 * need two characters, a carriage return and a line feed. On Unix, -50 * the ASCII line feed is used for both purposes - so we can not -51 * just use \n, because it would not have a carriage return and the -52 * next line will start at the column right after the line feed. -53 * -54 * This is why text files are different between Unix and MS Windows. -55 * In CP/M and derivatives, like MS-DOS and MS Windows, the ASCII -56 * standard was strictly adhered to, and therefore a newline requirs -57 * both a LF and a CR. -58 */ -59 (ttyops->write)(my_tty, "\015\012", 2); +47 /* ttys were originally hardware devices, which (usually) strictly +48 * followed the ASCII standard. In ASCII, to move to a new line you +49 * need two characters, a carriage return and a line feed. On Unix, +50 * the ASCII line feed is used for both purposes - so we can not +51 * just use \n, because it would not have a carriage return and the +52 * next line will start at the column right after the line feed. +53 * +54 * This is why text files are different between Unix and MS Windows. +55 * In CP/M and derivatives, like MS-DOS and MS Windows, the ASCII +56 * standard was strictly adhered to, and therefore a newline requirs +57 * both a LF and a CR. +58 */ +59 (ttyops->write)(my_tty, "\015\012", 2); 60 } 61} 62 -63static int __init print_string_init(void) +63static int __init print_string_init(void) 64{ -65 print_string("The module has been inserted. Hello world!"); -66 return 0; +65 print_string("The module has been inserted. Hello world!"); +66 return 0; 67} 68 -69static void __exit print_string_exit(void) +69static void __exit print_string_exit(void) 70{ -71 print_string("The module has been removed. Farewell world!"); +71 print_string("The module has been removed. Farewell world!"); 72} 73 74module_init(print_string_init); 75module_exit(print_string_exit); 76 -77MODULE_LICENSE("GPL");-
0.13.2 Flashing keyboard LEDs
-In certain conditions, you may desire a simpler and more direct way to communicate +
In certain conditions, you may desire a simpler and more direct way to communicate to the external world. Flashing keyboard LEDs can be such a solution: It is an immediate way to attract attention or to display a status condition. Keyboard LEDs are present on every hardware, they are always visible, they do not need any setup, and their use is rather simple and non-intrusive, compared to writing to a tty or a file. -
The following source code illustrates a minimal kernel module which, when +
The following source code illustrates a minimal kernel module which, when loaded, starts blinking the keyboard LEDs until it is unloaded.
-
1/* -2 * kbleds.c - Blink keyboard leds until the module is unloaded. -3 */ ++1/* +2 * kbleds.c - Blink keyboard leds until the module is unloaded. +3 */ 4 -5#include <linux/init.h> -6#include <linux/kd.h> /* For KDSETLED */ -7#include <linux/module.h> -8#include <linux/tty.h> /* For fg_console, MAX_NR_CONSOLES */ -9#include <linux/vt.h> -10#include <linux/vt_kern.h> /* for fg_console */ +5#include <linux/init.h> +6#include <linux/kd.h> /* For KDSETLED */ +7#include <linux/module.h> +8#include <linux/tty.h> /* For fg_console, MAX_NR_CONSOLES */ +9#include <linux/vt.h> +10#include <linux/vt_kern.h> /* for fg_console */ 11 -12#include <linux/console_struct.h> /* For vc_cons */ +12#include <linux/console_struct.h> /* For vc_cons */ 13 -14MODULE_DESCRIPTION("Example module illustrating the use of Keyboard LEDs."); +14MODULE_DESCRIPTION("Example module illustrating the use of Keyboard LEDs."); 15 -16struct timer_list my_timer; -17struct tty_driver *my_driver; -18char kbledstatus = 0; +16struct timer_list my_timer; +17struct tty_driver *my_driver; +18char kbledstatus = 0; 19 -20#define BLINK_DELAY HZ / 5 -21#define ALL_LEDS_ON 0x07 -22#define RESTORE_LEDS 0xFF +20#define BLINK_DELAY HZ / 5 +21#define ALL_LEDS_ON 0x07 +22#define RESTORE_LEDS 0xFF 23 -24/* Function my_timer_func blinks the keyboard LEDs periodically by invoking -25 * command KDSETLED of ioctl() on the keyboard driver. To learn more on virtual -26 * terminal ioctl operations, please see file: -27 * drivers/char/vt_ioctl.c, function vt_ioctl(). -28 * -29 * The argument to KDSETLED is alternatively set to 7 (thus causing the led -30 * mode to be set to LED_SHOW_IOCTL, and all the leds are lit) and to 0xFF -31 * (any value above 7 switches back the led mode to LED_SHOW_FLAGS, thus -32 * the LEDs reflect the actual keyboard status). To learn more on this, -33 * please see file: drivers/char/keyboard.c, function setledstate(). -34 */ +24/* Function my_timer_func blinks the keyboard LEDs periodically by invoking +25 * command KDSETLED of ioctl() on the keyboard driver. To learn more on virtual +26 * terminal ioctl operations, please see file: +27 * drivers/char/vt_ioctl.c, function vt_ioctl(). +28 * +29 * The argument to KDSETLED is alternatively set to 7 (thus causing the led +30 * mode to be set to LED_SHOW_IOCTL, and all the leds are lit) and to 0xFF +31 * (any value above 7 switches back the led mode to LED_SHOW_FLAGS, thus +32 * the LEDs reflect the actual keyboard status). To learn more on this, +33 * please see file: drivers/char/keyboard.c, function setledstate(). +34 */ 35 -36static void my_timer_func(unsigned long ptr) +36static void my_timer_func(unsigned long ptr) 37{ -38 unsigned long *pstatus = (unsigned long *) ptr; -39 struct tty_struct *t = vc_cons[fg_console].d->port.tty; +38 unsigned long *pstatus = (unsigned long *) ptr; +39 struct tty_struct *t = vc_cons[fg_console].d->port.tty; 40 -41 if (*pstatus == ALL_LEDS_ON) +41 if (*pstatus == ALL_LEDS_ON) 42 *pstatus = RESTORE_LEDS; -43 else +43 else 44 *pstatus = ALL_LEDS_ON; 45 46 (my_driver->ops->ioctl)(t, KDSETLED, *pstatus); @@ -4017,35 +4015,35 @@ loaded, starts blinking the keyboard LEDs until it is unloaded. 49 add_timer(&my_timer); 50} 51 -52static int __init kbleds_init(void) +52static int __init kbleds_init(void) 53{ -54 int i; +54 int i; 55 -56 pr_info("kbleds: loading\n"); -57 pr_info("kbleds: fgconsole is %x\n", fg_console); -58 for (i = 0; i < MAX_NR_CONSOLES; i++) { -59 if (!vc_cons[i].d) -60 break; -61 pr_info("poet_atkm: console[%i/%i] #%i, tty %lx\n", i, MAX_NR_CONSOLES, -62 vc_cons[i].d->vc_num, (unsigned long) vc_cons[i].d->port.tty); +56 pr_info("kbleds: loading\n"); +57 pr_info("kbleds: fgconsole is %x\n", fg_console); +58 for (i = 0; i < MAX_NR_CONSOLES; i++) { +59 if (!vc_cons[i].d) +60 break; +61 pr_info("poet_atkm: console[%i/%i] #%i, tty %lx\n", i, MAX_NR_CONSOLES, +62 vc_cons[i].d->vc_num, (unsigned long) vc_cons[i].d->port.tty); 63 } -64 pr_info("kbleds: finished scanning consoles\n"); +64 pr_info("kbleds: finished scanning consoles\n"); 65 66 my_driver = vc_cons[fg_console].d->port.tty->driver; -67 pr_info("kbleds: tty driver magic %x\n", my_driver->magic); +67 pr_info("kbleds: tty driver magic %x\n", my_driver->magic); 68 -69 /* Set up the LED blink timer the first time. */ -70 timer_setup(&my_timer, (void *) &my_timer_func, -71 (unsigned long) &kbledstatus); +69 /* Set up the LED blink timer the first time. */ +70 timer_setup(&my_timer, (void *) &my_timer_func, +71 (unsigned long) &kbledstatus); 72 my_timer.expires = jiffies + BLINK_DELAY; 73 add_timer(&my_timer); 74 -75 return 0; +75 return 0; 76} 77 -78static void __exit kbleds_cleanup(void) +78static void __exit kbleds_cleanup(void) 79{ -80 pr_info("kbleds: unloading...\n"); +80 pr_info("kbleds: unloading...\n"); 81 del_timer(&my_timer); 82 (my_driver->ops->ioctl)(vc_cons[fg_console].d->port.tty, KDSETLED, 83 RESTORE_LEDS); @@ -4054,14 +4052,14 @@ loaded, starts blinking the keyboard LEDs until it is unloaded. 86module_init(kbleds_init); 87module_exit(kbleds_cleanup); 88 -89MODULE_LICENSE("GPL");-If none of the examples in this chapter fit your debugging needs -there might yet be some other tricks to try. Ever wondered what -
CONFIG_LL_DEBUG -in make menuconfig is good for? If you activate that you get low level access to the +89MODULE_LICENSE("GPL");If none of the examples in this chapter fit your debugging needs +there might yet be some other tricks to try. Ever wondered what +
CONFIG_LL_DEBUG +in make menuconfig is good for? If you activate that you get low level access to the serial port. While this might not sound very powerful by itself, you can patch kernel/printk.c or any other essential syscall to print ASCII characters, thus makeing it possible to trace virtually everything what your code does @@ -4069,66 +4067,66 @@ over a serial line. If you find yourself porting the kernel to some new and former unsupported architecture, this is usually amongst the first things that should be implemented. Logging over a netconsole might also be worth a try. -While you have seen lots of stuff that can be used to aid debugging here, there are +
While you have seen lots of stuff that can be used to aid debugging here, there are some things to be aware of. Debugging is almost always intrusive. Adding debug code can change the situation enough to make the bug seem to dissappear. Thus you should try to keep debug code to a minimum and make sure it does not show up in production code. -
+
0.14 Scheduling Tasks
-There are two main ways of running tasks: tasklets and work queues. Tasklets are a +
There are two main ways of running tasks: tasklets and work queues. Tasklets are a quick and easy way of scheduling a single function to be run, for example when triggered from an interrupt, whereas work queues are more complicated but also better suited to running multiple things in a sequence. -
+
0.14.1 Tasklets
-Here is an example tasklet module. The +
Here is an example tasklet module. The
tasklet_fnfunction runs for a few seconds and in the mean time execution of theexample_tasklet_initfunction continues to the exit point.-
1/* -2 * example_tasklet.c -3 */ -4#include <linux/delay.h> -5#include <linux/interrupt.h> -6#include <linux/kernel.h> -7#include <linux/module.h> ++1/* +2 * example_tasklet.c +3 */ +4#include <linux/delay.h> +5#include <linux/interrupt.h> +6#include <linux/kernel.h> +7#include <linux/module.h> 8 -9static void tasklet_fn(unsigned long data) +9static void tasklet_fn(unsigned long data) 10{ -11 pr_info("Example tasklet starts\n"); +11 pr_info("Example tasklet starts\n"); 12 mdelay(5000); -13 pr_info("Example tasklet ends\n"); +13 pr_info("Example tasklet ends\n"); 14} 15 16DECLARE_TASKLET(mytask, tasklet_fn, 0L); 17 -18static int example_tasklet_init(void) +18static int example_tasklet_init(void) 19{ -20 pr_info("tasklet example init\n"); +20 pr_info("tasklet example init\n"); 21 tasklet_schedule(&mytask); 22 mdelay(200); -23 pr_info("Example tasklet init continues...\n"); -24 return 0; +23 pr_info("Example tasklet init continues...\n"); +24 return 0; 25} 26 -27static void example_tasklet_exit(void) +27static void example_tasklet_exit(void) 28{ -29 pr_info("tasklet example exit\n"); +29 pr_info("tasklet example exit\n"); 30 tasklet_kill(&mytask); 31} 32 33module_init(example_tasklet_init); 34module_exit(example_tasklet_exit); 35 -36MODULE_DESCRIPTION("Tasklet example"); -37MODULE_LICENSE("GPL");-So with this example loaded
dmesg +36MODULE_DESCRIPTION("Tasklet example"); +37MODULE_LICENSE("GPL");So with this example loaded
dmesgshould show: @@ -4140,39 +4138,39 @@ Example tasklet starts Example tasklet init continues... Example tasklet ends-
+
+
0.14.2 Work queues
-To add a task to the scheduler we can use a workqueue. The kernel then uses the +
To add a task to the scheduler we can use a workqueue. The kernel then uses the Completely Fair Scheduler (CFS) to execute work within the queue.
-
1/* -2 * sched.c -3 */ -4#include <linux/init.h> -5#include <linux/module.h> -6#include <linux/workqueue.h> ++1/* +2 * sched.c +3 */ +4#include <linux/init.h> +5#include <linux/module.h> +6#include <linux/workqueue.h> 7 -8static struct workqueue_struct *queue = NULL; -9static struct work_struct work; +8static struct workqueue_struct *queue = NULL; +9static struct work_struct work; 10 -11static void work_handler(struct work_struct *data) +11static void work_handler(struct work_struct *data) 12{ -13 pr_info("work handler function.\n"); +13 pr_info("work handler function.\n"); 14} 15 -16static int __init sched_init(void) +16static int __init sched_init(void) 17{ -18 queue = alloc_workqueue("HELLOWORLD", WQ_UNBOUND, 1); +18 queue = alloc_workqueue("HELLOWORLD", WQ_UNBOUND, 1); 19 INIT_WORK(&work, work_handler); 20 schedule_work(&work); 21 -22 return 0; +22 return 0; 23} 24 -25static void __exit sched_exit(void) +25static void __exit sched_exit(void) 26{ 27 destroy_workqueue(queue); 28} @@ -4180,21 +4178,21 @@ Completely Fair Scheduler (CFS) to execute work within the queue. 30module_init(sched_init); 31module_exit(sched_exit); 32 -33MODULE_LICENSE("GPL"); -34MODULE_DESCRIPTION("Workqueue example");-+33MODULE_LICENSE("GPL"); +34MODULE_DESCRIPTION("Workqueue example");
0.15 Interrupt Handlers
-+
0.15.1 Interrupt Handlers
-Except for the last chapter, everything we did in the kernel so far we have done as a +
Except for the last chapter, everything we did in the kernel so far we have done as a response to a process asking for it, either by dealing with a special file, sending an
ioctl(), or issuing a system call. But the job of the kernel is not just to respond to process requests. Another job, which is every bit as important, is to speak to the hardware connected to the machine. -There are two types of interaction between the CPU and the rest of the +
There are two types of interaction between the CPU and the rest of the computer’s hardware. The first type is when the CPU gives orders to the hardware, the other is when the hardware needs to tell the CPU something. The second, called interrupts, is much harder to implement because it has to be dealt with when @@ -4204,14 +4202,14 @@ lost. -
Under Linux, hardware interrupts are called IRQ’s (Interrupt ReQuests). There +
Under Linux, hardware interrupts are called IRQ’s (Interrupt ReQuests). There are two types of IRQ’s, short and long. A short IRQ is one which is expected to take a very short period of time, during which the rest of the machine will be blocked and no other interrupts will be handled. A long IRQ is one which can take longer, and during which other interrupts may occur (but not interrupts from the same device). If at all possible, it is better to declare an interrupt handler to be long. -
When the CPU receives an interrupt, it stops whatever it is doing (unless it is +
When the CPU receives an interrupt, it stops whatever it is doing (unless it is processing a more important interrupt, in which case it will deal with this one only when the more important one is done), saves certain parameters on the stack and calls the interrupt handler. This means that certain things @@ -4223,10 +4221,10 @@ the new information at a later time (this is called the "bottom half") and return. The kernel is then guaranteed to call the bottom half as soon as possible – and when it does, everything allowed in kernel modules will be allowed. -
The way to implement this is to call +
The way to implement this is to call
request_irq()to get your interrupt handler called when the relevant IRQ is received. -In practice IRQ handling can be a bit more complex. Hardware is often +
In practice IRQ handling can be a bit more complex. Hardware is often designed in a way that chains two interrupt controllers, so that all the IRQs from interrupt controller B are cascaded to a certain IRQ from interrupt controller A. Of course, that requires that the kernel finds out which IRQ it @@ -4240,7 +4238,7 @@ need to solve another truckload of problems. It is not enough to know if a certain IRQs has happend, it’s also important for what CPU(s) it was for. People still interested in more details, might want to refer to "APIC" now. -
This function receives the IRQ number, the name of the function, +
This function receives the IRQ number, the name of the function, flags, a name for /proc/interrupts and a parameter to pass to the interrupt handler. Usually there is a certain number of IRQs available. How many IRQs there are is hardware-dependent. The flags can include @@ -4253,129 +4251,129 @@ already a handler on this IRQ, or if you are both willing to share. -
+
0.15.2 Detecting button presses
-Many popular single board computers, such as Raspberry Pi or Beagleboards, have a +
Many popular single board computers, such as Raspberry Pi or Beagleboards, have a bunch of GPIO pins. Attaching buttons to those and then having a button press do something is a classic case in which you might need to use interrupts, so that instead of having the CPU waste time and battery power polling for a change in input state it is better for the input to trigger the CPU to then run a particular handling function. -
Here is an example where buttons are connected to GPIO numbers 17 and 18 and +
Here is an example where buttons are connected to GPIO numbers 17 and 18 and an LED is connected to GPIO 4. You can change those numbers to whatever is appropriate for your board.
-
1/* -2 * intrpt.c - Handling GPIO with interrupts -3 * -4 * Based upon the RPi example by Stefan Wendler (devnull@kaltpost.de) -5 * from: -6 * https://github.com/wendlers/rpi-kmod-samples -7 * -8 * Press one button to turn on a LED and another to turn it off. -9 */ ++1/* +2 * intrpt.c - Handling GPIO with interrupts +3 * +4 * Based upon the RPi example by Stefan Wendler (devnull@kaltpost.de) +5 * from: +6 * https://github.com/wendlers/rpi-kmod-samples +7 * +8 * Press one button to turn on a LED and another to turn it off. +9 */ 10 -11#include <linux/gpio.h> -12#include <linux/interrupt.h> -13#include <linux/kernel.h> -14#include <linux/module.h> +11#include <linux/gpio.h> +12#include <linux/interrupt.h> +13#include <linux/kernel.h> +14#include <linux/module.h> 15 -16static int button_irqs[] = {-1, -1}; +16static int button_irqs[] = {-1, -1}; 17 -18/* Define GPIOs for LEDs. -19 * TODO: Change the numbers for the GPIO on your board. -20 */ -21static struct gpio leds[] = {{4, GPIOF_OUT_INIT_LOW, "LED 1"}}; +18/* Define GPIOs for LEDs. +19 * TODO: Change the numbers for the GPIO on your board. +20 */ +21static struct gpio leds[] = {{4, GPIOF_OUT_INIT_LOW, "LED 1"}}; 22 -23/* Define GPIOs for BUTTONS -24 * TODO: Change the numbers for the GPIO on your board. -25 */ -26static struct gpio buttons[] = {{17, GPIOF_IN, "LED 1 ON BUTTON"}, -27 {18, GPIOF_IN, "LED 1 OFF BUTTON"}}; +23/* Define GPIOs for BUTTONS +24 * TODO: Change the numbers for the GPIO on your board. +25 */ +26static struct gpio buttons[] = {{17, GPIOF_IN, "LED 1 ON BUTTON"}, +27 {18, GPIOF_IN, "LED 1 OFF BUTTON"}}; 28 -29/* interrupt function triggered when a button is pressed. */ -30static irqreturn_t button_isr(int irq, void *data) +29/* interrupt function triggered when a button is pressed. */ +30static irqreturn_t button_isr(int irq, void *data) 31{ -32 /* first button */ -33 if (irq == button_irqs[0] && !gpio_get_value(leds[0].gpio)) +32 /* first button */ +33 if (irq == button_irqs[0] && !gpio_get_value(leds[0].gpio)) 34 gpio_set_value(leds[0].gpio, 1); -35 /* second button */ -36 else if (irq == button_irqs[1] && gpio_get_value(leds[0].gpio)) +35 /* second button */ +36 else if (irq == button_irqs[1] && gpio_get_value(leds[0].gpio)) 37 gpio_set_value(leds[0].gpio, 0); 38 -39 return IRQ_HANDLED; +39 return IRQ_HANDLED; 40} 41 -42static int __init intrpt_init(void) +42static int __init intrpt_init(void) 43{ -44 int ret = 0; +44 int ret = 0; 45 -46 pr_info("%s\n", __func__); +46 pr_info("%s\n", __func__); 47 -48 /* register LED gpios */ +48 /* register LED gpios */ 49 ret = gpio_request_array(leds, ARRAY_SIZE(leds)); 50 -51 if (ret) { -52 pr_err("Unable to request GPIOs for LEDs: %d\n", ret); -53 return ret; +51 if (ret) { +52 pr_err("Unable to request GPIOs for LEDs: %d\n", ret); +53 return ret; 54 } 55 -56 /* register BUTTON gpios */ +56 /* register BUTTON gpios */ 57 ret = gpio_request_array(buttons, ARRAY_SIZE(buttons)); 58 -59 if (ret) { -60 pr_err("Unable to request GPIOs for BUTTONs: %d\n", ret); -61 goto fail1; +59 if (ret) { +60 pr_err("Unable to request GPIOs for BUTTONs: %d\n", ret); +61 goto fail1; 62 } 63 -64 pr_info("Current button1 value: %d\n", gpio_get_value(buttons[0].gpio)); +64 pr_info("Current button1 value: %d\n", gpio_get_value(buttons[0].gpio)); 65 66 ret = gpio_to_irq(buttons[0].gpio); 67 -68 if (ret < 0) { -69 pr_err("Unable to request IRQ: %d\n", ret); -70 goto fail2; +68 if (ret < 0) { +69 pr_err("Unable to request IRQ: %d\n", ret); +70 goto fail2; 71 } 72 73 button_irqs[0] = ret; 74 -75 pr_info("Successfully requested BUTTON1 IRQ # %d\n", button_irqs[0]); +75 pr_info("Successfully requested BUTTON1 IRQ # %d\n", button_irqs[0]); 76 77 ret = request_irq(button_irqs[0], button_isr, 78 IRQF_TRIGGER_RISING | IRQF_TRIGGER_FALLING, -79 "gpiomod#button1", NULL); +79 "gpiomod#button1", NULL); 80 -81 if (ret) { -82 pr_err("Unable to request IRQ: %d\n", ret); -83 goto fail2; +81 if (ret) { +82 pr_err("Unable to request IRQ: %d\n", ret); +83 goto fail2; 84 } 85 86 87 ret = gpio_to_irq(buttons[1].gpio); 88 -89 if (ret < 0) { -90 pr_err("Unable to request IRQ: %d\n", ret); -91 goto fail2; +89 if (ret < 0) { +90 pr_err("Unable to request IRQ: %d\n", ret); +91 goto fail2; 92 } 93 94 button_irqs[1] = ret; 95 -96 pr_info("Successfully requested BUTTON2 IRQ # %d\n", button_irqs[1]); +96 pr_info("Successfully requested BUTTON2 IRQ # %d\n", button_irqs[1]); 97 98 ret = request_irq(button_irqs[1], button_isr, 99 IRQF_TRIGGER_RISING | IRQF_TRIGGER_FALLING, -100 "gpiomod#button2", NULL); +100 "gpiomod#button2", NULL); 101 -102 if (ret) { -103 pr_err("Unable to request IRQ: %d\n", ret); -104 goto fail3; +102 if (ret) { +103 pr_err("Unable to request IRQ: %d\n", ret); +104 goto fail3; 105 } 106 -107 return 0; +107 return 0; 108 -109/* cleanup what has been setup so far */ +109/* cleanup what has been setup so far */ 110fail3: 111 free_irq(button_irqs[0], NULL); 112 @@ -4385,24 +4383,24 @@ appropriate for your board. 116fail1: 117 gpio_free_array(leds, ARRAY_SIZE(leds)); 118 -119 return ret; +119 return ret; 120} 121 -122static void __exit intrpt_exit(void) +122static void __exit intrpt_exit(void) 123{ -124 int i; +124 int i; 125 -126 pr_info("%s\n", __func__); +126 pr_info("%s\n", __func__); 127 -128 /* free irqs */ +128 /* free irqs */ 129 free_irq(button_irqs[0], NULL); 130 free_irq(button_irqs[1], NULL); 131 -132 /* turn all LEDs off */ -133 for (i = 0; i < ARRAY_SIZE(leds); i++) +132 /* turn all LEDs off */ +133 for (i = 0; i < ARRAY_SIZE(leds); i++) 134 gpio_set_value(leds[i].gpio, 0); 135 -136 /* unregister */ +136 /* unregister */ 137 gpio_free_array(leds, ARRAY_SIZE(leds)); 138 gpio_free_array(buttons, ARRAY_SIZE(buttons)); 139} @@ -4410,144 +4408,144 @@ appropriate for your board. 141module_init(intrpt_init); 142module_exit(intrpt_exit); 143 -144MODULE_LICENSE("GPL"); -145MODULE_DESCRIPTION("Handle some GPIO interrupts");-+144MODULE_LICENSE("GPL"); +145MODULE_DESCRIPTION("Handle some GPIO interrupts");
0.15.3 Bottom Half
-Suppose you want to do a bunch of stuff inside of an interrupt routine. A common +
Suppose you want to do a bunch of stuff inside of an interrupt routine. A common way to do that without rendering the interrupt unavailable for a significant duration is to combine it with a tasklet. This pushes the bulk of the work off into the scheduler. -
The example below modifies the previous example to also run an additional task +
The example below modifies the previous example to also run an additional task when an interrupt is triggered.
-
1/* -2 * bottomhalf.c - Top and bottom half interrupt handling -3 * -4 * Based upon the RPi example by Stefan Wendler (devnull@kaltpost.de) -5 * from: -6 * https://github.com/wendlers/rpi-kmod-samples -7 * -8 * Press one button to turn on a LED and another to turn it off -9 */ ++1/* +2 * bottomhalf.c - Top and bottom half interrupt handling +3 * +4 * Based upon the RPi example by Stefan Wendler (devnull@kaltpost.de) +5 * from: +6 * https://github.com/wendlers/rpi-kmod-samples +7 * +8 * Press one button to turn on a LED and another to turn it off +9 */ 10 -11#include <linux/delay.h> -12#include <linux/gpio.h> -13#include <linux/interrupt.h> -14#include <linux/kernel.h> -15#include <linux/module.h> +11#include <linux/delay.h> +12#include <linux/gpio.h> +13#include <linux/interrupt.h> +14#include <linux/kernel.h> +15#include <linux/module.h> 16 -17static int button_irqs[] = {-1, -1}; +17static int button_irqs[] = {-1, -1}; 18 -19/* Define GPIOs for LEDs. -20 * TODO: Change the numbers for the GPIO on your board. -21 */ -22static struct gpio leds[] = {{4, GPIOF_OUT_INIT_LOW, "LED 1"}}; +19/* Define GPIOs for LEDs. +20 * TODO: Change the numbers for the GPIO on your board. +21 */ +22static struct gpio leds[] = {{4, GPIOF_OUT_INIT_LOW, "LED 1"}}; 23 -24/* Define GPIOs for BUTTONS -25 * TODO: Change the numbers for the GPIO on your board. -26 */ -27static struct gpio buttons[] = { -28 {17, GPIOF_IN, "LED 1 ON BUTTON"}, -29 {18, GPIOF_IN, "LED 1 OFF BUTTON"}, +24/* Define GPIOs for BUTTONS +25 * TODO: Change the numbers for the GPIO on your board. +26 */ +27static struct gpio buttons[] = { +28 {17, GPIOF_IN, "LED 1 ON BUTTON"}, +29 {18, GPIOF_IN, "LED 1 OFF BUTTON"}, 30}; 31 -32/* Tasklet containing some non-trivial amount of processing */ -33static void bottomhalf_tasklet_fn(unsigned long data) +32/* Tasklet containing some non-trivial amount of processing */ +33static void bottomhalf_tasklet_fn(unsigned long data) 34{ -35 pr_info("Bottom half tasklet starts\n"); -36 /* do something which takes a while */ +35 pr_info("Bottom half tasklet starts\n"); +36 /* do something which takes a while */ 37 mdelay(500); -38 pr_info("Bottom half tasklet ends\n"); +38 pr_info("Bottom half tasklet ends\n"); 39} 40 41DECLARE_TASKLET(buttontask, bottomhalf_tasklet_fn, 0L); 42 -43/* interrupt function triggered when a button is pressed */ -44static irqreturn_t button_isr(int irq, void *data) +43/* interrupt function triggered when a button is pressed */ +44static irqreturn_t button_isr(int irq, void *data) 45{ -46 /* Do something quickly right now */ -47 if (irq == button_irqs[0] && !gpio_get_value(leds[0].gpio)) +46 /* Do something quickly right now */ +47 if (irq == button_irqs[0] && !gpio_get_value(leds[0].gpio)) 48 gpio_set_value(leds[0].gpio, 1); -49 else if (irq == button_irqs[1] && gpio_get_value(leds[0].gpio)) +49 else if (irq == button_irqs[1] && gpio_get_value(leds[0].gpio)) 50 gpio_set_value(leds[0].gpio, 0); 51 -52 /* Do the rest at leisure via the scheduler */ +52 /* Do the rest at leisure via the scheduler */ 53 tasklet_schedule(&buttontask); 54 -55 return IRQ_HANDLED; +55 return IRQ_HANDLED; 56} 57 -58static int __init bottomhalf_init(void) +58static int __init bottomhalf_init(void) 59{ -60 int ret = 0; +60 int ret = 0; 61 -62 pr_info("%s\n", __func__); +62 pr_info("%s\n", __func__); 63 -64 /* register LED gpios */ +64 /* register LED gpios */ 65 ret = gpio_request_array(leds, ARRAY_SIZE(leds)); 66 -67 if (ret) { -68 pr_err("Unable to request GPIOs for LEDs: %d\n", ret); -69 return ret; +67 if (ret) { +68 pr_err("Unable to request GPIOs for LEDs: %d\n", ret); +69 return ret; 70 } 71 -72 /* register BUTTON gpios */ +72 /* register BUTTON gpios */ 73 ret = gpio_request_array(buttons, ARRAY_SIZE(buttons)); 74 -75 if (ret) { -76 pr_err("Unable to request GPIOs for BUTTONs: %d\n", ret); -77 goto fail1; +75 if (ret) { +76 pr_err("Unable to request GPIOs for BUTTONs: %d\n", ret); +77 goto fail1; 78 } 79 -80 pr_info("Current button1 value: %d\n", gpio_get_value(buttons[0].gpio)); +80 pr_info("Current button1 value: %d\n", gpio_get_value(buttons[0].gpio)); 81 82 ret = gpio_to_irq(buttons[0].gpio); 83 -84 if (ret < 0) { -85 pr_err("Unable to request IRQ: %d\n", ret); -86 goto fail2; +84 if (ret < 0) { +85 pr_err("Unable to request IRQ: %d\n", ret); +86 goto fail2; 87 } 88 89 button_irqs[0] = ret; 90 -91 pr_info("Successfully requested BUTTON1 IRQ # %d\n", button_irqs[0]); +91 pr_info("Successfully requested BUTTON1 IRQ # %d\n", button_irqs[0]); 92 93 ret = request_irq(button_irqs[0], button_isr, 94 IRQF_TRIGGER_RISING | IRQF_TRIGGER_FALLING, -95 "gpiomod#button1", NULL); +95 "gpiomod#button1", NULL); 96 -97 if (ret) { -98 pr_err("Unable to request IRQ: %d\n", ret); -99 goto fail2; +97 if (ret) { +98 pr_err("Unable to request IRQ: %d\n", ret); +99 goto fail2; 100 } 101 102 103 ret = gpio_to_irq(buttons[1].gpio); 104 -105 if (ret < 0) { -106 pr_err("Unable to request IRQ: %d\n", ret); -107 goto fail2; +105 if (ret < 0) { +106 pr_err("Unable to request IRQ: %d\n", ret); +107 goto fail2; 108 } 109 110 button_irqs[1] = ret; 111 -112 pr_info("Successfully requested BUTTON2 IRQ # %d\n", button_irqs[1]); +112 pr_info("Successfully requested BUTTON2 IRQ # %d\n", button_irqs[1]); 113 114 ret = request_irq(button_irqs[1], button_isr, 115 IRQF_TRIGGER_RISING | IRQF_TRIGGER_FALLING, -116 "gpiomod#button2", NULL); +116 "gpiomod#button2", NULL); 117 -118 if (ret) { -119 pr_err("Unable to request IRQ: %d\n", ret); -120 goto fail3; +118 if (ret) { +119 pr_err("Unable to request IRQ: %d\n", ret); +120 goto fail3; 121 } 122 -123 return 0; +123 return 0; 124 -125/* cleanup what has been setup so far */ +125/* cleanup what has been setup so far */ 126fail3: 127 free_irq(button_irqs[0], NULL); 128 @@ -4557,24 +4555,24 @@ when an interrupt is triggered. 132fail1: 133 gpio_free_array(leds, ARRAY_SIZE(leds)); 134 -135 return ret; +135 return ret; 136} 137 -138static void __exit bottomhalf_exit(void) +138static void __exit bottomhalf_exit(void) 139{ -140 int i; +140 int i; 141 -142 pr_info("%s\n", __func__); +142 pr_info("%s\n", __func__); 143 -144 /* free irqs */ +144 /* free irqs */ 145 free_irq(button_irqs[0], NULL); 146 free_irq(button_irqs[1], NULL); 147 -148 /* turn all LEDs off */ -149 for (i = 0; i < ARRAY_SIZE(leds); i++) +148 /* turn all LEDs off */ +149 for (i = 0; i < ARRAY_SIZE(leds); i++) 150 gpio_set_value(leds[i].gpio, 0); 151 -152 /* unregister */ +152 /* unregister */ 153 gpio_free_array(leds, ARRAY_SIZE(leds)); 154 gpio_free_array(buttons, ARRAY_SIZE(buttons)); 155} @@ -4582,12 +4580,12 @@ when an interrupt is triggered. 157module_init(bottomhalf_init); 158module_exit(bottomhalf_exit); 159 -160MODULE_LICENSE("GPL"); -161MODULE_DESCRIPTION("Interrupt with top and bottom half");-+160MODULE_LICENSE("GPL"); +161MODULE_DESCRIPTION("Interrupt with top and bottom half");
0.16 Crypto
-At the dawn of the internet everybody trusted everybody completely…but that did +
At the dawn of the internet everybody trusted everybody completely…but that did not work out so well. When this guide was originally written it was a more innocent era in which almost nobody actually gave a damn about crypto - least of all kernel developers. That is certainly no longer the case now. To handle crypto stuff the @@ -4596,269 +4594,269 @@ favourite hash functions. -
+
0.16.1 Hash functions
-Calculating and checking the hashes of things is a common operation. Here is a +
Calculating and checking the hashes of things is a common operation. Here is a demonstration of how to calculate a sha256 hash within a kernel module.
-
1/* -2 * cryptosha256.c -3 */ -4#include <crypto/internal/hash.h> -5#include <linux/module.h> ++1/* +2 * cryptosha256.c +3 */ +4#include <crypto/internal/hash.h> +5#include <linux/module.h> 6 -7#define SHA256_LENGTH 32 +7#define SHA256_LENGTH 32 8 -9static void show_hash_result(char *plaintext, char *hash_sha256) +9static void show_hash_result(char *plaintext, char *hash_sha256) 10{ -11 int i; -12 char str[SHA256_LENGTH * 2 + 1]; +11 int i; +12 char str[SHA256_LENGTH * 2 + 1]; 13 -14 pr_info("sha256 test for string: \"%s\"\n", plaintext); -15 for (i = 0; i < SHA256_LENGTH; i++) -16 sprintf(&str[i * 2], "%02x", (unsigned char) hash_sha256[i]); +14 pr_info("sha256 test for string: \"%s\"\n", plaintext); +15 for (i = 0; i < SHA256_LENGTH; i++) +16 sprintf(&str[i * 2], "%02x", (unsigned char) hash_sha256[i]); 17 str[i * 2] = 0; -18 pr_info("%s\n", str); +18 pr_info("%s\n", str); 19} 20 -21int cryptosha256_init(void) +21int cryptosha256_init(void) 22{ -23 char *plaintext = "This is a test"; -24 char hash_sha256[SHA256_LENGTH]; -25 struct crypto_shash *sha256; -26 struct shash_desc *shash; +23 char *plaintext = "This is a test"; +24 char hash_sha256[SHA256_LENGTH]; +25 struct crypto_shash *sha256; +26 struct shash_desc *shash; 27 -28 sha256 = crypto_alloc_shash("sha256", 0, 0); -29 if (IS_ERR(sha256)) -30 return -1; +28 sha256 = crypto_alloc_shash("sha256", 0, 0); +29 if (IS_ERR(sha256)) +30 return -1; 31 -32 shash = kmalloc(sizeof(struct shash_desc) + crypto_shash_descsize(sha256), +32 shash = kmalloc(sizeof(struct shash_desc) + crypto_shash_descsize(sha256), 33 GFP_KERNEL); -34 if (!shash) -35 return -ENOMEM; +34 if (!shash) +35 return -ENOMEM; 36 37 shash->tfm = sha256; 38 -39 if (crypto_shash_init(shash)) -40 return -1; +39 if (crypto_shash_init(shash)) +40 return -1; 41 -42 if (crypto_shash_update(shash, plaintext, strlen(plaintext))) -43 return -1; +42 if (crypto_shash_update(shash, plaintext, strlen(plaintext))) +43 return -1; 44 -45 if (crypto_shash_final(shash, hash_sha256)) -46 return -1; +45 if (crypto_shash_final(shash, hash_sha256)) +46 return -1; 47 48 kfree(shash); 49 crypto_free_shash(sha256); 50 51 show_hash_result(plaintext, hash_sha256); 52 -53 return 0; +53 return 0; 54} 55 -56void cryptosha256_exit(void) {} +56void cryptosha256_exit(void) {} 57 58module_init(cryptosha256_init); 59module_exit(cryptosha256_exit); 60 -61MODULE_DESCRIPTION("sha256 hash test"); -62MODULE_LICENSE("GPL");-Make and install the module: +61MODULE_DESCRIPTION("sha256 hash test"); +62MODULE_LICENSE("GPL");
Make and install the module:
1make 2sudo insmod cryptosha256.ko 3dmesg-And you should see that the hash was calculated for the test string. -
Finally, remove the test module: +
And you should see that the hash was calculated for the test string. +
Finally, remove the test module:
1sudo rmmod cryptosha256-+
0.16.2 Symmetric key encryption
-Here is an example of symmetrically encrypting a string using the AES algorithm +
Here is an example of symmetrically encrypting a string using the AES algorithm and a password.
-
1/* -2 * cryptosk.c -3 */ -4#include <crypto/internal/skcipher.h> -5#include <linux/crypto.h> -6#include <linux/module.h> ++1/* +2 * cryptosk.c +3 */ +4#include <crypto/internal/skcipher.h> +5#include <linux/crypto.h> +6#include <linux/module.h> 7 -8#define SYMMETRIC_KEY_LENGTH 32 -9#define CIPHER_BLOCK_SIZE 16 +8#define SYMMETRIC_KEY_LENGTH 32 +9#define CIPHER_BLOCK_SIZE 16 10 -11struct tcrypt_result { -12 struct completion completion; -13 int err; +11struct tcrypt_result { +12 struct completion completion; +13 int err; 14}; 15 -16struct skcipher_def { -17 struct scatterlist sg; -18 struct crypto_skcipher *tfm; -19 struct skcipher_request *req; -20 struct tcrypt_result result; -21 char *scratchpad; -22 char *ciphertext; -23 char *ivdata; +16struct skcipher_def { +17 struct scatterlist sg; +18 struct crypto_skcipher *tfm; +19 struct skcipher_request *req; +20 struct tcrypt_result result; +21 char *scratchpad; +22 char *ciphertext; +23 char *ivdata; 24}; 25 -26static struct skcipher_def sk; +26static struct skcipher_def sk; 27 -28static void test_skcipher_finish(struct skcipher_def *sk) +28static void test_skcipher_finish(struct skcipher_def *sk) 29{ -30 if (sk->tfm) +30 if (sk->tfm) 31 crypto_free_skcipher(sk->tfm); -32 if (sk->req) +32 if (sk->req) 33 skcipher_request_free(sk->req); -34 if (sk->ivdata) +34 if (sk->ivdata) 35 kfree(sk->ivdata); -36 if (sk->scratchpad) +36 if (sk->scratchpad) 37 kfree(sk->scratchpad); -38 if (sk->ciphertext) +38 if (sk->ciphertext) 39 kfree(sk->ciphertext); 40} 41 -42static int test_skcipher_result(struct skcipher_def *sk, int rc) +42static int test_skcipher_result(struct skcipher_def *sk, int rc) 43{ -44 switch (rc) { -45 case 0: -46 break; -47 case -EINPROGRESS || -EBUSY: +44 switch (rc) { +45 case 0: +46 break; +47 case -EINPROGRESS || -EBUSY: 48 rc = wait_for_completion_interruptible(&sk->result.completion); -49 if (!rc && !sk->result.err) { +49 if (!rc && !sk->result.err) { 50 reinit_completion(&sk->result.completion); -51 break; +51 break; 52 } -53 pr_info("skcipher encrypt returned with %d result %d\n", rc, +53 pr_info("skcipher encrypt returned with %d result %d\n", rc, 54 sk->result.err); -55 break; -56 default: -57 pr_info("skcipher encrypt returned with %d result %d\n", rc, +55 break; +56 default: +57 pr_info("skcipher encrypt returned with %d result %d\n", rc, 58 sk->result.err); -59 break; +59 break; 60 } 61 62 init_completion(&sk->result.completion); 63 -64 return rc; +64 return rc; 65} 66 -67static void test_skcipher_callback(struct crypto_async_request *req, int error) +67static void test_skcipher_callback(struct crypto_async_request *req, int error) 68{ -69 struct tcrypt_result *result = req->data; +69 struct tcrypt_result *result = req->data; 70 -71 if (error == -EINPROGRESS) -72 return; +71 if (error == -EINPROGRESS) +72 return; 73 74 result->err = error; 75 complete(&result->completion); -76 pr_info("Encryption finished successfully\n"); +76 pr_info("Encryption finished successfully\n"); 77 -78 /* decrypt data */ -79#if 0 -80 memset((void*)sk.scratchpad, '-', CIPHER_BLOCK_SIZE); -81 ret = crypto_skcipher_decrypt(sk.req); -82 ret = test_skcipher_result(&sk, ret); -83 if (ret) -84 return; +78 /* decrypt data */ +79#if 0 +80 memset((void*)sk.scratchpad, '-', CIPHER_BLOCK_SIZE); +81 ret = crypto_skcipher_decrypt(sk.req); +82 ret = test_skcipher_result(&sk, ret); +83 if (ret) +84 return; 85 -86 sg_copy_from_buffer(&sk.sg, 1, sk.scratchpad, CIPHER_BLOCK_SIZE); -87 sk.scratchpad[CIPHER_BLOCK_SIZE-1] = 0; +86 sg_copy_from_buffer(&sk.sg, 1, sk.scratchpad, CIPHER_BLOCK_SIZE); +87 sk.scratchpad[CIPHER_BLOCK_SIZE-1] = 0; 88 -89 pr_info("Decryption request successful\n"); -90 pr_info("Decrypted: %s\n", sk.scratchpad); -91#endif +89 pr_info("Decryption request successful\n"); +90 pr_info("Decrypted: %s\n", sk.scratchpad); +91#endif 92} 93 -94static int test_skcipher_encrypt(char *plaintext, -95 char *password, -96 struct skcipher_def *sk) +94static int test_skcipher_encrypt(char *plaintext, +95 char *password, +96 struct skcipher_def *sk) 97{ -98 int ret = -EFAULT; -99 unsigned char key[SYMMETRIC_KEY_LENGTH]; +98 int ret = -EFAULT; +99 unsigned char key[SYMMETRIC_KEY_LENGTH]; 100 -101 if (!sk->tfm) { -102 sk->tfm = crypto_alloc_skcipher("cbc-aes-aesni", 0, 0); -103 if (IS_ERR(sk->tfm)) { -104 pr_info("could not allocate skcipher handle\n"); -105 return PTR_ERR(sk->tfm); +101 if (!sk->tfm) { +102 sk->tfm = crypto_alloc_skcipher("cbc-aes-aesni", 0, 0); +103 if (IS_ERR(sk->tfm)) { +104 pr_info("could not allocate skcipher handle\n"); +105 return PTR_ERR(sk->tfm); 106 } 107 } 108 -109 if (!sk->req) { +109 if (!sk->req) { 110 sk->req = skcipher_request_alloc(sk->tfm, GFP_KERNEL); -111 if (!sk->req) { -112 pr_info("could not allocate skcipher request\n"); +111 if (!sk->req) { +112 pr_info("could not allocate skcipher request\n"); 113 ret = -ENOMEM; -114 goto out; +114 goto out; 115 } 116 } 117 118 skcipher_request_set_callback(sk->req, CRYPTO_TFM_REQ_MAY_BACKLOG, 119 test_skcipher_callback, &sk->result); 120 -121 /* clear the key */ -122 memset((void *) key, '\0', SYMMETRIC_KEY_LENGTH); +121 /* clear the key */ +122 memset((void *) key, '\0', SYMMETRIC_KEY_LENGTH); 123 -124 /* Use the world's favourite password */ -125 sprintf((char *) key, "%s", password); +124 /* Use the world's favourite password */ +125 sprintf((char *) key, "%s", password); 126 -127 /* AES 256 with given symmetric key */ -128 if (crypto_skcipher_setkey(sk->tfm, key, SYMMETRIC_KEY_LENGTH)) { -129 pr_info("key could not be set\n"); +127 /* AES 256 with given symmetric key */ +128 if (crypto_skcipher_setkey(sk->tfm, key, SYMMETRIC_KEY_LENGTH)) { +129 pr_info("key could not be set\n"); 130 ret = -EAGAIN; -131 goto out; +131 goto out; 132 } -133 pr_info("Symmetric key: %s\n", key); -134 pr_info("Plaintext: %s\n", plaintext); +133 pr_info("Symmetric key: %s\n", key); +134 pr_info("Plaintext: %s\n", plaintext); 135 -136 if (!sk->ivdata) { -137 /* see https://en.wikipedia.org/wiki/Initialization_vector */ +136 if (!sk->ivdata) { +137 /* see https://en.wikipedia.org/wiki/Initialization_vector */ 138 sk->ivdata = kmalloc(CIPHER_BLOCK_SIZE, GFP_KERNEL); -139 if (!sk->ivdata) { -140 pr_info("could not allocate ivdata\n"); -141 goto out; +139 if (!sk->ivdata) { +140 pr_info("could not allocate ivdata\n"); +141 goto out; 142 } 143 get_random_bytes(sk->ivdata, CIPHER_BLOCK_SIZE); 144 } 145 -146 if (!sk->scratchpad) { -147 /* The text to be encrypted */ +146 if (!sk->scratchpad) { +147 /* The text to be encrypted */ 148 sk->scratchpad = kmalloc(CIPHER_BLOCK_SIZE, GFP_KERNEL); -149 if (!sk->scratchpad) { -150 pr_info("could not allocate scratchpad\n"); -151 goto out; +149 if (!sk->scratchpad) { +150 pr_info("could not allocate scratchpad\n"); +151 goto out; 152 } 153 } -154 sprintf((char *) sk->scratchpad, "%s", plaintext); +154 sprintf((char *) sk->scratchpad, "%s", plaintext); 155 156 sg_init_one(&sk->sg, sk->scratchpad, CIPHER_BLOCK_SIZE); 157 skcipher_request_set_crypt(sk->req, &sk->sg, &sk->sg, CIPHER_BLOCK_SIZE, 158 sk->ivdata); 159 init_completion(&sk->result.completion); 160 -161 /* encrypt data */ +161 /* encrypt data */ 162 ret = crypto_skcipher_encrypt(sk->req); 163 ret = test_skcipher_result(sk, ret); -164 if (ret) -165 goto out; +164 if (ret) +165 goto out; 166 -167 pr_info("Encryption request successful\n"); +167 pr_info("Encryption request successful\n"); 168 169out: -170 return ret; +170 return ret; 171} 172 -173int cryptoapi_init(void) +173int cryptoapi_init(void) 174{ -175 /* The world's favorite password */ -176 char *password = "password123"; +175 /* The world's favorite password */ +176 char *password = "password123"; 177 178 sk.tfm = NULL; 179 sk.req = NULL; @@ -4866,11 +4864,11 @@ and a password. 181 sk.ciphertext = NULL; 182 sk.ivdata = NULL; 183 -184 test_skcipher_encrypt("Testing", password, &sk); -185 return 0; +184 test_skcipher_encrypt("Testing", password, &sk); +185 return 0; 186} 187 -188void cryptoapi_exit(void) +188void cryptoapi_exit(void) 189{ 190 test_skcipher_finish(&sk); 191} @@ -4878,12 +4876,12 @@ and a password. 193module_init(cryptoapi_init); 194module_exit(cryptoapi_exit); 195 -196MODULE_DESCRIPTION("Symmetric key encryption example"); -197MODULE_LICENSE("GPL");-+196MODULE_DESCRIPTION("Symmetric key encryption example"); +197MODULE_LICENSE("GPL");
0.17 Standardizing the interfaces: The Device Model
-Up to this point we have seen all kinds of modules doing all kinds of things, but there +
Up to this point we have seen all kinds of modules doing all kinds of things, but there was no consistency in their interfaces with the rest of the kernel. To impose some consistency such that there is at minimum a standardized way to start, suspend and resume a device a device model was added. An example is show below, and you can @@ -4891,59 +4889,59 @@ use this as a template to add your own suspend, resume or other interface functions.
-
1/* -2 * devicemodel.c -3 */ -4#include <linux/kernel.h> -5#include <linux/module.h> -6#include <linux/platform_device.h> +-1/* +2 * devicemodel.c +3 */ +4#include <linux/kernel.h> +5#include <linux/module.h> +6#include <linux/platform_device.h> 7 -8struct devicemodel_data { -9 char *greeting; -10 int number; +8struct devicemodel_data { +9 char *greeting; +10 int number; 11}; 12 -13static int devicemodel_probe(struct platform_device *dev) +13static int devicemodel_probe(struct platform_device *dev) 14{ -15 struct devicemodel_data *pd = -16 (struct devicemodel_data *) (dev->dev.platform_data); +15 struct devicemodel_data *pd = +16 (struct devicemodel_data *) (dev->dev.platform_data); 17 -18 pr_info("devicemodel probe\n"); -19 pr_info("devicemodel greeting: %s; %d\n", pd->greeting, pd->number); +18 pr_info("devicemodel probe\n"); +19 pr_info("devicemodel greeting: %s; %d\n", pd->greeting, pd->number); 20 -21 /* Your device initialization code */ +21 /* Your device initialization code */ 22 -23 return 0; +23 return 0; 24} 25 -26static int devicemodel_remove(struct platform_device *dev) +26static int devicemodel_remove(struct platform_device *dev) 27{ -28 pr_info("devicemodel example removed\n"); +28 pr_info("devicemodel example removed\n"); 29 -30 /* Your device removal code */ +30 /* Your device removal code */ 31 -32 return 0; +32 return 0; 33} 34 -35static int devicemodel_suspend(struct device *dev) +35static int devicemodel_suspend(struct device *dev) 36{ -37 pr_info("devicemodel example suspend\n"); +37 pr_info("devicemodel example suspend\n"); 38 -39 /* Your device suspend code */ +39 /* Your device suspend code */ 40 -41 return 0; +41 return 0; 42} 43 -44static int devicemodel_resume(struct device *dev) +44static int devicemodel_resume(struct device *dev) 45{ -46 pr_info("devicemodel example resume\n"); +46 pr_info("devicemodel example resume\n"); 47 -48 /* Your device resume code */ +48 /* Your device resume code */ 49 -50 return 0; +50 return 0; 51} 52 -53static const struct dev_pm_ops devicemodel_pm_ops = { +53static const struct dev_pm_ops devicemodel_pm_ops = { 54 .suspend = devicemodel_suspend, 55 .resume = devicemodel_resume, 56 .poweroff = devicemodel_suspend, @@ -4952,10 +4950,10 @@ functions. 59 .restore = devicemodel_resume, 60}; 61 -62static struct platform_driver devicemodel_driver = { +62static struct platform_driver devicemodel_driver = { 63 .driver = 64 { -65 .name = "devicemodel_example", +65 .name = "devicemodel_example", 66 .owner = THIS_MODULE, 67 .pm = &devicemodel_pm_ops, 68 }, @@ -4963,43 +4961,43 @@ functions. 70 .remove = devicemodel_remove, 71}; 72 -73static int devicemodel_init(void) +73static int devicemodel_init(void) 74{ -75 int ret; +75 int ret; 76 -77 pr_info("devicemodel init\n"); +77 pr_info("devicemodel init\n"); 78 79 ret = platform_driver_register(&devicemodel_driver); 80 -81 if (ret) { -82 pr_err("Unable to register driver\n"); -83 return ret; +81 if (ret) { +82 pr_err("Unable to register driver\n"); +83 return ret; 84 } 85 -86 return 0; +86 return 0; 87} 88 -89static void devicemodel_exit(void) +89static void devicemodel_exit(void) 90{ -91 pr_info("devicemodel exit\n"); +91 pr_info("devicemodel exit\n"); 92 platform_driver_unregister(&devicemodel_driver); 93} 94 95module_init(devicemodel_init); 96module_exit(devicemodel_exit); 97 -98MODULE_LICENSE("GPL"); -99MODULE_DESCRIPTION("Linux Device Model example");+98MODULE_LICENSE("GPL"); +99MODULE_DESCRIPTION("Linux Device Model example");+
0.18 Optimizations
-+
0.18.1 Likely and Unlikely conditions
-Sometimes you might want your code to run as quickly as possible, +
Sometimes you might want your code to run as quickly as possible, especially if it is handling an interrupt or doing something which might cause noticible latency. If your code contains boolean conditions and if you know that the conditions are almost always likely to evaluate as either @@ -5009,60 +5007,60 @@ know that the conditions are almost always likely to evaluate as either
likelyandunlikelymacros. -For example, when allocating memory you are almost always expecting this to +
For example, when allocating memory you are almost always expecting this to succeed.
1bvl = bvec_alloc(gfp_mask, nr_iovecs, &idx); -2if (unlikely(!bvl)) { +2if (unlikely(!bvl)) { 3 mempool_free(bio, bio_pool); 4 bio = NULL; -5 goto out; +5 goto out; 6}-When the
unlikely +When the
unlikelymacro is used, the compiler alters its machine instruction output, so that it continues along the false branch and only jumps if the condition is true. That avoids flushing the processor pipeline. The opposite happens if you use thelikelymacro. -+
0.19 Common Pitfalls
-+
0.19.1 Using standard libraries
-You can not do that. In a kernel module you can only use kernel functions, which are +
You can not do that. In a kernel module you can only use kernel functions, which are the functions you can see in /proc/kallsyms. -
+
0.19.2 Disabling interrupts
-You might need to do this for a short time and that is OK, but if you do not enable +
You might need to do this for a short time and that is OK, but if you do not enable them afterwards, your system will be stuck and you will have to power it off. -
+
0.20 Where To Go From Here?
-For people seriously interested in kernel programming, I recommend kernelnewbies.org -and the Documentation subdirectory within the kernel source code which is not +
For people seriously interested in kernel programming, I recommend kernelnewbies.org +and the Documentation subdirectory within the kernel source code which is not always easy to understand but can be a starting point for further investigation. Also, as Linus Torvalds said, the best way to learn the kernel is to read the source code yourself. -
If you are interested in more examples of short kernel modules then searching on +
If you are interested in more examples of short kernel modules then searching on sites such as Github and Gitlab is a good way to start, although there is a lot of duplication of older LKMPG examples which may not compile with newer kernel versions. You will also be able to find examples of the use of kernel modules to attack or compromise systems or exfiltrate data and those can be useful for thinking about how to defend systems and learning about existing security mechanisms within the kernel. -
I hope I have helped you in your quest to become a better programmer, or at +
I hope I have helped you in your quest to become a better programmer, or at least to have fun through technology. And, if you do write useful kernel modules, I hope you publish them under the GPL, so I can use them too. -
If you would like to contribute to this guide or notice anything glaringly wrong, +
If you would like to contribute to this guide or notice anything glaringly wrong, please create an issue at https://github.com/sysprog21/lkmpg. -
Happy hacking! +
Happy hacking!