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