Loading...
1Copyright 2009 Jonathan Corbet <corbet@lwn.net>
2
3Debugfs exists as a simple way for kernel developers to make information
4available to user space. Unlike /proc, which is only meant for information
5about a process, or sysfs, which has strict one-value-per-file rules,
6debugfs has no rules at all. Developers can put any information they want
7there. The debugfs filesystem is also intended to not serve as a stable
8ABI to user space; in theory, there are no stability constraints placed on
9files exported there. The real world is not always so simple, though [1];
10even debugfs interfaces are best designed with the idea that they will need
11to be maintained forever.
12
13Debugfs is typically mounted with a command like:
14
15 mount -t debugfs none /sys/kernel/debug
16
17(Or an equivalent /etc/fstab line).
18The debugfs root directory is accessible only to the root user by
19default. To change access to the tree the "uid", "gid" and "mode" mount
20options can be used.
21
22Note that the debugfs API is exported GPL-only to modules.
23
24Code using debugfs should include <linux/debugfs.h>. Then, the first order
25of business will be to create at least one directory to hold a set of
26debugfs files:
27
28 struct dentry *debugfs_create_dir(const char *name, struct dentry *parent);
29
30This call, if successful, will make a directory called name underneath the
31indicated parent directory. If parent is NULL, the directory will be
32created in the debugfs root. On success, the return value is a struct
33dentry pointer which can be used to create files in the directory (and to
34clean it up at the end). An ERR_PTR(-ERROR) return value indicates that
35something went wrong. If ERR_PTR(-ENODEV) is returned, that is an
36indication that the kernel has been built without debugfs support and none
37of the functions described below will work.
38
39The most general way to create a file within a debugfs directory is with:
40
41 struct dentry *debugfs_create_file(const char *name, umode_t mode,
42 struct dentry *parent, void *data,
43 const struct file_operations *fops);
44
45Here, name is the name of the file to create, mode describes the access
46permissions the file should have, parent indicates the directory which
47should hold the file, data will be stored in the i_private field of the
48resulting inode structure, and fops is a set of file operations which
49implement the file's behavior. At a minimum, the read() and/or write()
50operations should be provided; others can be included as needed. Again,
51the return value will be a dentry pointer to the created file,
52ERR_PTR(-ERROR) on error, or ERR_PTR(-ENODEV) if debugfs support is
53missing.
54
55Create a file with an initial size, the following function can be used
56instead:
57
58 struct dentry *debugfs_create_file_size(const char *name, umode_t mode,
59 struct dentry *parent, void *data,
60 const struct file_operations *fops,
61 loff_t file_size);
62
63file_size is the initial file size. The other parameters are the same
64as the function debugfs_create_file.
65
66In a number of cases, the creation of a set of file operations is not
67actually necessary; the debugfs code provides a number of helper functions
68for simple situations. Files containing a single integer value can be
69created with any of:
70
71 struct dentry *debugfs_create_u8(const char *name, umode_t mode,
72 struct dentry *parent, u8 *value);
73 struct dentry *debugfs_create_u16(const char *name, umode_t mode,
74 struct dentry *parent, u16 *value);
75 struct dentry *debugfs_create_u32(const char *name, umode_t mode,
76 struct dentry *parent, u32 *value);
77 struct dentry *debugfs_create_u64(const char *name, umode_t mode,
78 struct dentry *parent, u64 *value);
79
80These files support both reading and writing the given value; if a specific
81file should not be written to, simply set the mode bits accordingly. The
82values in these files are in decimal; if hexadecimal is more appropriate,
83the following functions can be used instead:
84
85 struct dentry *debugfs_create_x8(const char *name, umode_t mode,
86 struct dentry *parent, u8 *value);
87 struct dentry *debugfs_create_x16(const char *name, umode_t mode,
88 struct dentry *parent, u16 *value);
89 struct dentry *debugfs_create_x32(const char *name, umode_t mode,
90 struct dentry *parent, u32 *value);
91 struct dentry *debugfs_create_x64(const char *name, umode_t mode,
92 struct dentry *parent, u64 *value);
93
94These functions are useful as long as the developer knows the size of the
95value to be exported. Some types can have different widths on different
96architectures, though, complicating the situation somewhat. There is a
97function meant to help out in one special case:
98
99 struct dentry *debugfs_create_size_t(const char *name, umode_t mode,
100 struct dentry *parent,
101 size_t *value);
102
103As might be expected, this function will create a debugfs file to represent
104a variable of type size_t.
105
106Boolean values can be placed in debugfs with:
107
108 struct dentry *debugfs_create_bool(const char *name, umode_t mode,
109 struct dentry *parent, bool *value);
110
111A read on the resulting file will yield either Y (for non-zero values) or
112N, followed by a newline. If written to, it will accept either upper- or
113lower-case values, or 1 or 0. Any other input will be silently ignored.
114
115Also, atomic_t values can be placed in debugfs with:
116
117 struct dentry *debugfs_create_atomic_t(const char *name, umode_t mode,
118 struct dentry *parent, atomic_t *value)
119
120A read of this file will get atomic_t values, and a write of this file
121will set atomic_t values.
122
123Another option is exporting a block of arbitrary binary data, with
124this structure and function:
125
126 struct debugfs_blob_wrapper {
127 void *data;
128 unsigned long size;
129 };
130
131 struct dentry *debugfs_create_blob(const char *name, umode_t mode,
132 struct dentry *parent,
133 struct debugfs_blob_wrapper *blob);
134
135A read of this file will return the data pointed to by the
136debugfs_blob_wrapper structure. Some drivers use "blobs" as a simple way
137to return several lines of (static) formatted text output. This function
138can be used to export binary information, but there does not appear to be
139any code which does so in the mainline. Note that all files created with
140debugfs_create_blob() are read-only.
141
142If you want to dump a block of registers (something that happens quite
143often during development, even if little such code reaches mainline.
144Debugfs offers two functions: one to make a registers-only file, and
145another to insert a register block in the middle of another sequential
146file.
147
148 struct debugfs_reg32 {
149 char *name;
150 unsigned long offset;
151 };
152
153 struct debugfs_regset32 {
154 struct debugfs_reg32 *regs;
155 int nregs;
156 void __iomem *base;
157 };
158
159 struct dentry *debugfs_create_regset32(const char *name, umode_t mode,
160 struct dentry *parent,
161 struct debugfs_regset32 *regset);
162
163 void debugfs_print_regs32(struct seq_file *s, struct debugfs_reg32 *regs,
164 int nregs, void __iomem *base, char *prefix);
165
166The "base" argument may be 0, but you may want to build the reg32 array
167using __stringify, and a number of register names (macros) are actually
168byte offsets over a base for the register block.
169
170If you want to dump an u32 array in debugfs, you can create file with:
171
172 void debugfs_create_u32_array(const char *name, umode_t mode,
173 struct dentry *parent,
174 u32 *array, u32 elements);
175
176The "array" argument provides data, and the "elements" argument is
177the number of elements in the array. Note: Once array is created its
178size can not be changed.
179
180There is a helper function to create device related seq_file:
181
182 struct dentry *debugfs_create_devm_seqfile(struct device *dev,
183 const char *name,
184 struct dentry *parent,
185 int (*read_fn)(struct seq_file *s,
186 void *data));
187
188The "dev" argument is the device related to this debugfs file, and
189the "read_fn" is a function pointer which to be called to print the
190seq_file content.
191
192There are a couple of other directory-oriented helper functions:
193
194 struct dentry *debugfs_rename(struct dentry *old_dir,
195 struct dentry *old_dentry,
196 struct dentry *new_dir,
197 const char *new_name);
198
199 struct dentry *debugfs_create_symlink(const char *name,
200 struct dentry *parent,
201 const char *target);
202
203A call to debugfs_rename() will give a new name to an existing debugfs
204file, possibly in a different directory. The new_name must not exist prior
205to the call; the return value is old_dentry with updated information.
206Symbolic links can be created with debugfs_create_symlink().
207
208There is one important thing that all debugfs users must take into account:
209there is no automatic cleanup of any directories created in debugfs. If a
210module is unloaded without explicitly removing debugfs entries, the result
211will be a lot of stale pointers and no end of highly antisocial behavior.
212So all debugfs users - at least those which can be built as modules - must
213be prepared to remove all files and directories they create there. A file
214can be removed with:
215
216 void debugfs_remove(struct dentry *dentry);
217
218The dentry value can be NULL or an error value, in which case nothing will
219be removed.
220
221Once upon a time, debugfs users were required to remember the dentry
222pointer for every debugfs file they created so that all files could be
223cleaned up. We live in more civilized times now, though, and debugfs users
224can call:
225
226 void debugfs_remove_recursive(struct dentry *dentry);
227
228If this function is passed a pointer for the dentry corresponding to the
229top-level directory, the entire hierarchy below that directory will be
230removed.
231
232Notes:
233 [1] http://lwn.net/Articles/309298/
1Copyright 2009 Jonathan Corbet <corbet@lwn.net>
2
3Debugfs exists as a simple way for kernel developers to make information
4available to user space. Unlike /proc, which is only meant for information
5about a process, or sysfs, which has strict one-value-per-file rules,
6debugfs has no rules at all. Developers can put any information they want
7there. The debugfs filesystem is also intended to not serve as a stable
8ABI to user space; in theory, there are no stability constraints placed on
9files exported there. The real world is not always so simple, though [1];
10even debugfs interfaces are best designed with the idea that they will need
11to be maintained forever.
12
13Debugfs is typically mounted with a command like:
14
15 mount -t debugfs none /sys/kernel/debug
16
17(Or an equivalent /etc/fstab line).
18
19Note that the debugfs API is exported GPL-only to modules.
20
21Code using debugfs should include <linux/debugfs.h>. Then, the first order
22of business will be to create at least one directory to hold a set of
23debugfs files:
24
25 struct dentry *debugfs_create_dir(const char *name, struct dentry *parent);
26
27This call, if successful, will make a directory called name underneath the
28indicated parent directory. If parent is NULL, the directory will be
29created in the debugfs root. On success, the return value is a struct
30dentry pointer which can be used to create files in the directory (and to
31clean it up at the end). A NULL return value indicates that something went
32wrong. If ERR_PTR(-ENODEV) is returned, that is an indication that the
33kernel has been built without debugfs support and none of the functions
34described below will work.
35
36The most general way to create a file within a debugfs directory is with:
37
38 struct dentry *debugfs_create_file(const char *name, mode_t mode,
39 struct dentry *parent, void *data,
40 const struct file_operations *fops);
41
42Here, name is the name of the file to create, mode describes the access
43permissions the file should have, parent indicates the directory which
44should hold the file, data will be stored in the i_private field of the
45resulting inode structure, and fops is a set of file operations which
46implement the file's behavior. At a minimum, the read() and/or write()
47operations should be provided; others can be included as needed. Again,
48the return value will be a dentry pointer to the created file, NULL for
49error, or ERR_PTR(-ENODEV) if debugfs support is missing.
50
51In a number of cases, the creation of a set of file operations is not
52actually necessary; the debugfs code provides a number of helper functions
53for simple situations. Files containing a single integer value can be
54created with any of:
55
56 struct dentry *debugfs_create_u8(const char *name, mode_t mode,
57 struct dentry *parent, u8 *value);
58 struct dentry *debugfs_create_u16(const char *name, mode_t mode,
59 struct dentry *parent, u16 *value);
60 struct dentry *debugfs_create_u32(const char *name, mode_t mode,
61 struct dentry *parent, u32 *value);
62 struct dentry *debugfs_create_u64(const char *name, mode_t mode,
63 struct dentry *parent, u64 *value);
64
65These files support both reading and writing the given value; if a specific
66file should not be written to, simply set the mode bits accordingly. The
67values in these files are in decimal; if hexadecimal is more appropriate,
68the following functions can be used instead:
69
70 struct dentry *debugfs_create_x8(const char *name, mode_t mode,
71 struct dentry *parent, u8 *value);
72 struct dentry *debugfs_create_x16(const char *name, mode_t mode,
73 struct dentry *parent, u16 *value);
74 struct dentry *debugfs_create_x32(const char *name, mode_t mode,
75 struct dentry *parent, u32 *value);
76 struct dentry *debugfs_create_x64(const char *name, mode_t mode,
77 struct dentry *parent, u64 *value);
78
79These functions are useful as long as the developer knows the size of the
80value to be exported. Some types can have different widths on different
81architectures, though, complicating the situation somewhat. There is a
82function meant to help out in one special case:
83
84 struct dentry *debugfs_create_size_t(const char *name, mode_t mode,
85 struct dentry *parent,
86 size_t *value);
87
88As might be expected, this function will create a debugfs file to represent
89a variable of type size_t.
90
91Boolean values can be placed in debugfs with:
92
93 struct dentry *debugfs_create_bool(const char *name, mode_t mode,
94 struct dentry *parent, u32 *value);
95
96A read on the resulting file will yield either Y (for non-zero values) or
97N, followed by a newline. If written to, it will accept either upper- or
98lower-case values, or 1 or 0. Any other input will be silently ignored.
99
100Finally, a block of arbitrary binary data can be exported with:
101
102 struct debugfs_blob_wrapper {
103 void *data;
104 unsigned long size;
105 };
106
107 struct dentry *debugfs_create_blob(const char *name, mode_t mode,
108 struct dentry *parent,
109 struct debugfs_blob_wrapper *blob);
110
111A read of this file will return the data pointed to by the
112debugfs_blob_wrapper structure. Some drivers use "blobs" as a simple way
113to return several lines of (static) formatted text output. This function
114can be used to export binary information, but there does not appear to be
115any code which does so in the mainline. Note that all files created with
116debugfs_create_blob() are read-only.
117
118There are a couple of other directory-oriented helper functions:
119
120 struct dentry *debugfs_rename(struct dentry *old_dir,
121 struct dentry *old_dentry,
122 struct dentry *new_dir,
123 const char *new_name);
124
125 struct dentry *debugfs_create_symlink(const char *name,
126 struct dentry *parent,
127 const char *target);
128
129A call to debugfs_rename() will give a new name to an existing debugfs
130file, possibly in a different directory. The new_name must not exist prior
131to the call; the return value is old_dentry with updated information.
132Symbolic links can be created with debugfs_create_symlink().
133
134There is one important thing that all debugfs users must take into account:
135there is no automatic cleanup of any directories created in debugfs. If a
136module is unloaded without explicitly removing debugfs entries, the result
137will be a lot of stale pointers and no end of highly antisocial behavior.
138So all debugfs users - at least those which can be built as modules - must
139be prepared to remove all files and directories they create there. A file
140can be removed with:
141
142 void debugfs_remove(struct dentry *dentry);
143
144The dentry value can be NULL, in which case nothing will be removed.
145
146Once upon a time, debugfs users were required to remember the dentry
147pointer for every debugfs file they created so that all files could be
148cleaned up. We live in more civilized times now, though, and debugfs users
149can call:
150
151 void debugfs_remove_recursive(struct dentry *dentry);
152
153If this function is passed a pointer for the dentry corresponding to the
154top-level directory, the entire hierarchy below that directory will be
155removed.
156
157Notes:
158 [1] http://lwn.net/Articles/309298/