1.. SPDX-License-Identifier: GPL-2.0
2
3.. _deprecated:
4
5=====================================================================
6Deprecated Interfaces, Language Features, Attributes, and Conventions
7=====================================================================
8
9In a perfect world, it would be possible to convert all instances of
10some deprecated API into the new API and entirely remove the old API in
11a single development cycle. However, due to the size of the kernel, the
12maintainership hierarchy, and timing, it's not always feasible to do these
13kinds of conversions at once. This means that new instances may sneak into
14the kernel while old ones are being removed, only making the amount of
15work to remove the API grow. In order to educate developers about what
16has been deprecated and why, this list has been created as a place to
17point when uses of deprecated things are proposed for inclusion in the
18kernel.
19
20__deprecated
21------------
22While this attribute does visually mark an interface as deprecated,
23it `does not produce warnings during builds any more
24<https://git.kernel.org/linus/771c035372a036f83353eef46dbb829780330234>`_
25because one of the standing goals of the kernel is to build without
26warnings and no one was actually doing anything to remove these deprecated
27interfaces. While using `__deprecated` is nice to note an old API in
28a header file, it isn't the full solution. Such interfaces must either
29be fully removed from the kernel, or added to this file to discourage
30others from using them in the future.
31
32BUG() and BUG_ON()
33------------------
34Use WARN() and WARN_ON() instead, and handle the "impossible"
35error condition as gracefully as possible. While the BUG()-family
36of APIs were originally designed to act as an "impossible situation"
37assert and to kill a kernel thread "safely", they turn out to just be
38too risky. (e.g. "In what order do locks need to be released? Have
39various states been restored?") Very commonly, using BUG() will
40destabilize a system or entirely break it, which makes it impossible
41to debug or even get viable crash reports. Linus has `very strong
42<https://lore.kernel.org/lkml/CA+55aFy6jNLsywVYdGp83AMrXBo_P-pkjkphPGrO=82SPKCpLQ@mail.gmail.com/>`_
43feelings `about this
44<https://lore.kernel.org/lkml/CAHk-=whDHsbK3HTOpTF=ue_o04onRwTEaK_ZoJp_fjbqq4+=Jw@mail.gmail.com/>`_.
45
46Note that the WARN()-family should only be used for "expected to
47be unreachable" situations. If you want to warn about "reachable
48but undesirable" situations, please use the pr_warn()-family of
49functions. System owners may have set the *panic_on_warn* sysctl,
50to make sure their systems do not continue running in the face of
51"unreachable" conditions. (For example, see commits like `this one
52<https://git.kernel.org/linus/d4689846881d160a4d12a514e991a740bcb5d65a>`_.)
53
54open-coded arithmetic in allocator arguments
55--------------------------------------------
56Dynamic size calculations (especially multiplication) should not be
57performed in memory allocator (or similar) function arguments due to the
58risk of them overflowing. This could lead to values wrapping around and a
59smaller allocation being made than the caller was expecting. Using those
60allocations could lead to linear overflows of heap memory and other
61misbehaviors. (One exception to this is literal values where the compiler
62can warn if they might overflow. However, the preferred way in these
63cases is to refactor the code as suggested below to avoid the open-coded
64arithmetic.)
65
66For example, do not use ``count * size`` as an argument, as in::
67
68	foo = kmalloc(count * size, GFP_KERNEL);
69
70Instead, the 2-factor form of the allocator should be used::
71
72	foo = kmalloc_array(count, size, GFP_KERNEL);
73
74Specifically, kmalloc() can be replaced with kmalloc_array(), and
75kzalloc() can be replaced with kcalloc().
76
77If no 2-factor form is available, the saturate-on-overflow helpers should
78be used::
79
80	bar = vmalloc(array_size(count, size));
81
82Another common case to avoid is calculating the size of a structure with
83a trailing array of others structures, as in::
84
85	header = kzalloc(sizeof(*header) + count * sizeof(*header->item),
86			 GFP_KERNEL);
87
88Instead, use the helper::
89
90	header = kzalloc(struct_size(header, item, count), GFP_KERNEL);
91
92.. note:: If you are using struct_size() on a structure containing a zero-length
93        or a one-element array as a trailing array member, please refactor such
94        array usage and switch to a `flexible array member
95        <#zero-length-and-one-element-arrays>`_ instead.
96
97For other calculations, please compose the use of the size_mul(),
98size_add(), and size_sub() helpers. For example, in the case of::
99
100	foo = krealloc(current_size + chunk_size * (count - 3), GFP_KERNEL);
101
102Instead, use the helpers::
103
104	foo = krealloc(size_add(current_size,
105				size_mul(chunk_size,
106					 size_sub(count, 3))), GFP_KERNEL);
107
108For more details, also see array3_size() and flex_array_size(),
109as well as the related check_mul_overflow(), check_add_overflow(),
110check_sub_overflow(), and check_shl_overflow() family of functions.
111
112simple_strtol(), simple_strtoll(), simple_strtoul(), simple_strtoull()
113----------------------------------------------------------------------
114The simple_strtol(), simple_strtoll(),
115simple_strtoul(), and simple_strtoull() functions
116explicitly ignore overflows, which may lead to unexpected results
117in callers. The respective kstrtol(), kstrtoll(),
118kstrtoul(), and kstrtoull() functions tend to be the
119correct replacements, though note that those require the string to be
120NUL or newline terminated.
121
122strcpy()
123--------
124strcpy() performs no bounds checking on the destination buffer. This
125could result in linear overflows beyond the end of the buffer, leading to
126all kinds of misbehaviors. While `CONFIG_FORTIFY_SOURCE=y` and various
127compiler flags help reduce the risk of using this function, there is
128no good reason to add new uses of this function. The safe replacement
129is strscpy(), though care must be given to any cases where the return
130value of strcpy() was used, since strscpy() does not return a pointer to
131the destination, but rather a count of non-NUL bytes copied (or negative
132errno when it truncates).
133
134strncpy() on NUL-terminated strings
135-----------------------------------
136Use of strncpy() does not guarantee that the destination buffer will
137be NUL terminated. This can lead to various linear read overflows and
138other misbehavior due to the missing termination. It also NUL-pads
139the destination buffer if the source contents are shorter than the
140destination buffer size, which may be a needless performance penalty
141for callers using only NUL-terminated strings.
142
143When the destination is required to be NUL-terminated, the replacement is
144strscpy(), though care must be given to any cases where the return value
145of strncpy() was used, since strscpy() does not return a pointer to the
146destination, but rather a count of non-NUL bytes copied (or negative
147errno when it truncates). Any cases still needing NUL-padding should
148instead use strscpy_pad().
149
150If a caller is using non-NUL-terminated strings, strtomem() should be
151used, and the destinations should be marked with the `__nonstring
152<https://gcc.gnu.org/onlinedocs/gcc/Common-Variable-Attributes.html>`_
153attribute to avoid future compiler warnings. For cases still needing
154NUL-padding, strtomem_pad() can be used.
155
156strlcpy()
157---------
158strlcpy() reads the entire source buffer first (since the return value
159is meant to match that of strlen()). This read may exceed the destination
160size limit. This is both inefficient and can lead to linear read overflows
161if a source string is not NUL-terminated. The safe replacement is strscpy(),
162though care must be given to any cases where the return value of strlcpy()
163is used, since strscpy() will return negative errno values when it truncates.
164
165%p format specifier
166-------------------
167Traditionally, using "%p" in format strings would lead to regular address
168exposure flaws in dmesg, proc, sysfs, etc. Instead of leaving these to
169be exploitable, all "%p" uses in the kernel are being printed as a hashed
170value, rendering them unusable for addressing. New uses of "%p" should not
171be added to the kernel. For text addresses, using "%pS" is likely better,
172as it produces the more useful symbol name instead. For nearly everything
173else, just do not add "%p" at all.
174
175Paraphrasing Linus's current `guidance <https://lore.kernel.org/lkml/CA+55aFwQEd_d40g4mUCSsVRZzrFPUJt74vc6PPpb675hYNXcKw@mail.gmail.com/>`_:
176
177- If the hashed "%p" value is pointless, ask yourself whether the pointer
178  itself is important. Maybe it should be removed entirely?
179- If you really think the true pointer value is important, why is some
180  system state or user privilege level considered "special"? If you think
181  you can justify it (in comments and commit log) well enough to stand
182  up to Linus's scrutiny, maybe you can use "%px", along with making sure
183  you have sensible permissions.
184
185If you are debugging something where "%p" hashing is causing problems,
186you can temporarily boot with the debug flag "`no_hash_pointers
187<https://git.kernel.org/linus/5ead723a20e0447bc7db33dc3070b420e5f80aa6>`_".
188
189Variable Length Arrays (VLAs)
190-----------------------------
191Using stack VLAs produces much worse machine code than statically
192sized stack arrays. While these non-trivial `performance issues
193<https://git.kernel.org/linus/02361bc77888>`_ are reason enough to
194eliminate VLAs, they are also a security risk. Dynamic growth of a stack
195array may exceed the remaining memory in the stack segment. This could
196lead to a crash, possible overwriting sensitive contents at the end of the
197stack (when built without `CONFIG_THREAD_INFO_IN_TASK=y`), or overwriting
198memory adjacent to the stack (when built without `CONFIG_VMAP_STACK=y`)
199
200Implicit switch case fall-through
201---------------------------------
202The C language allows switch cases to fall through to the next case
203when a "break" statement is missing at the end of a case. This, however,
204introduces ambiguity in the code, as it's not always clear if the missing
205break is intentional or a bug. For example, it's not obvious just from
206looking at the code if `STATE_ONE` is intentionally designed to fall
207through into `STATE_TWO`::
208
209	switch (value) {
210	case STATE_ONE:
211		do_something();
212	case STATE_TWO:
213		do_other();
214		break;
215	default:
216		WARN("unknown state");
217	}
218
219As there have been a long list of flaws `due to missing "break" statements
220<https://cwe.mitre.org/data/definitions/484.html>`_, we no longer allow
221implicit fall-through. In order to identify intentional fall-through
222cases, we have adopted a pseudo-keyword macro "fallthrough" which
223expands to gcc's extension `__attribute__((__fallthrough__))
224<https://gcc.gnu.org/onlinedocs/gcc/Statement-Attributes.html>`_.
225(When the C17/C18  `[[fallthrough]]` syntax is more commonly supported by
226C compilers, static analyzers, and IDEs, we can switch to using that syntax
227for the macro pseudo-keyword.)
228
229All switch/case blocks must end in one of:
230
231* break;
232* fallthrough;
233* continue;
234* goto <label>;
235* return [expression];
236
237Zero-length and one-element arrays
238----------------------------------
239There is a regular need in the kernel to provide a way to declare having
240a dynamically sized set of trailing elements in a structure. Kernel code
241should always use `"flexible array members" <https://en.wikipedia.org/wiki/Flexible_array_member>`_
242for these cases. The older style of one-element or zero-length arrays should
243no longer be used.
244
245In older C code, dynamically sized trailing elements were done by specifying
246a one-element array at the end of a structure::
247
248        struct something {
249                size_t count;
250                struct foo items[1];
251        };
252
253This led to fragile size calculations via sizeof() (which would need to
254remove the size of the single trailing element to get a correct size of
255the "header"). A `GNU C extension <https://gcc.gnu.org/onlinedocs/gcc/Zero-Length.html>`_
256was introduced to allow for zero-length arrays, to avoid these kinds of
257size problems::
258
259        struct something {
260                size_t count;
261                struct foo items[0];
262        };
263
264But this led to other problems, and didn't solve some problems shared by
265both styles, like not being able to detect when such an array is accidentally
266being used _not_ at the end of a structure (which could happen directly, or
267when such a struct was in unions, structs of structs, etc).
268
269C99 introduced "flexible array members", which lacks a numeric size for
270the array declaration entirely::
271
272        struct something {
273                size_t count;
274                struct foo items[];
275        };
276
277This is the way the kernel expects dynamically sized trailing elements
278to be declared. It allows the compiler to generate errors when the
279flexible array does not occur last in the structure, which helps to prevent
280some kind of `undefined behavior
281<https://git.kernel.org/linus/76497732932f15e7323dc805e8ea8dc11bb587cf>`_
282bugs from being inadvertently introduced to the codebase. It also allows
283the compiler to correctly analyze array sizes (via sizeof(),
284`CONFIG_FORTIFY_SOURCE`, and `CONFIG_UBSAN_BOUNDS`). For instance,
285there is no mechanism that warns us that the following application of the
286sizeof() operator to a zero-length array always results in zero::
287
288        struct something {
289                size_t count;
290                struct foo items[0];
291        };
292
293        struct something *instance;
294
295        instance = kmalloc(struct_size(instance, items, count), GFP_KERNEL);
296        instance->count = count;
297
298        size = sizeof(instance->items) * instance->count;
299        memcpy(instance->items, source, size);
300
301At the last line of code above, ``size`` turns out to be ``zero``, when one might
302have thought it represents the total size in bytes of the dynamic memory recently
303allocated for the trailing array ``items``. Here are a couple examples of this
304issue: `link 1
305<https://git.kernel.org/linus/f2cd32a443da694ac4e28fbf4ac6f9d5cc63a539>`_,
306`link 2
307<https://git.kernel.org/linus/ab91c2a89f86be2898cee208d492816ec238b2cf>`_.
308Instead, `flexible array members have incomplete type, and so the sizeof()
309operator may not be applied <https://gcc.gnu.org/onlinedocs/gcc/Zero-Length.html>`_,
310so any misuse of such operators will be immediately noticed at build time.
311
312With respect to one-element arrays, one has to be acutely aware that `such arrays
313occupy at least as much space as a single object of the type
314<https://gcc.gnu.org/onlinedocs/gcc/Zero-Length.html>`_,
315hence they contribute to the size of the enclosing structure. This is prone
316to error every time people want to calculate the total size of dynamic memory
317to allocate for a structure containing an array of this kind as a member::
318
319        struct something {
320                size_t count;
321                struct foo items[1];
322        };
323
324        struct something *instance;
325
326        instance = kmalloc(struct_size(instance, items, count - 1), GFP_KERNEL);
327        instance->count = count;
328
329        size = sizeof(instance->items) * instance->count;
330        memcpy(instance->items, source, size);
331
332In the example above, we had to remember to calculate ``count - 1`` when using
333the struct_size() helper, otherwise we would have --unintentionally-- allocated
334memory for one too many ``items`` objects. The cleanest and least error-prone way
335to implement this is through the use of a `flexible array member`, together with
336struct_size() and flex_array_size() helpers::
337
338        struct something {
339                size_t count;
340                struct foo items[];
341        };
342
343        struct something *instance;
344
345        instance = kmalloc(struct_size(instance, items, count), GFP_KERNEL);
346        instance->count = count;
347
348        memcpy(instance->items, source, flex_array_size(instance, items, instance->count));
349
350There are two special cases of replacement where the DECLARE_FLEX_ARRAY()
351helper needs to be used. (Note that it is named __DECLARE_FLEX_ARRAY() for
352use in UAPI headers.) Those cases are when the flexible array is either
353alone in a struct or is part of a union. These are disallowed by the C99
354specification, but for no technical reason (as can be seen by both the
355existing use of such arrays in those places and the work-around that
356DECLARE_FLEX_ARRAY() uses). For example, to convert this::
357
358	struct something {
359		...
360		union {
361			struct type1 one[0];
362			struct type2 two[0];
363		};
364	};
365
366The helper must be used::
367
368	struct something {
369		...
370		union {
371			DECLARE_FLEX_ARRAY(struct type1, one);
372			DECLARE_FLEX_ARRAY(struct type2, two);
373		};
374	};
375