1How to Get Your Patch Accepted Into the Hwmon Subsystem
2=======================================================
3
4This text is a collection of suggestions for people writing patches or
5drivers for the hwmon subsystem. Following these suggestions will greatly
6increase the chances of your change being accepted.
7
8
91. General
10----------
11
12* It should be unnecessary to mention, but please read and follow:
13
14    - Documentation/process/submit-checklist.rst
15    - Documentation/process/submitting-drivers.rst
16    - Documentation/process/submitting-patches.rst
17    - Documentation/process/coding-style.rst
18
19* Please run your patch through 'checkpatch --strict'. There should be no
20  errors, no warnings, and few if any check messages. If there are any
21  messages, please be prepared to explain.
22
23* If your patch generates checkpatch errors, warnings, or check messages,
24  please refrain from explanations such as "I prefer that coding style".
25  Keep in mind that each unnecessary message helps hiding a real problem,
26  and a consistent coding style makes it easier for others to understand
27  and review the code.
28
29* Please test your patch thoroughly. We are not your test group.
30  Sometimes a patch can not or not completely be tested because of missing
31  hardware. In such cases, you should test-build the code on at least one
32  architecture. If run-time testing was not achieved, it should be written
33  explicitly below the patch header.
34
35* If your patch (or the driver) is affected by configuration options such as
36  CONFIG_SMP, make sure it compiles for all configuration variants.
37
38
392. Adding functionality to existing drivers
40-------------------------------------------
41
42* Make sure the documentation in Documentation/hwmon/<driver_name>.rst is up to
43  date.
44
45* Make sure the information in Kconfig is up to date.
46
47* If the added functionality requires some cleanup or structural changes, split
48  your patch into a cleanup part and the actual addition. This makes it easier
49  to review your changes, and to bisect any resulting problems.
50
51* Never mix bug fixes, cleanup, and functional enhancements in a single patch.
52
53
543. New drivers
55--------------
56
57* Running your patch or driver file(s) through checkpatch does not mean its
58  formatting is clean. If unsure about formatting in your new driver, run it
59  through Lindent. Lindent is not perfect, and you may have to do some minor
60  cleanup, but it is a good start.
61
62* Consider adding yourself to MAINTAINERS.
63
64* Document the driver in Documentation/hwmon/<driver_name>.rst.
65
66* Add the driver to Kconfig and Makefile in alphabetical order.
67
68* Make sure that all dependencies are listed in Kconfig.
69
70* Please list include files in alphabetic order.
71
72* Please align continuation lines with '(' on the previous line.
73
74* Avoid forward declarations if you can. Rearrange the code if necessary.
75
76* Avoid macros to generate groups of sensor attributes. It not only confuses
77  checkpatch, but also makes it more difficult to review the code.
78
79* Avoid calculations in macros and macro-generated functions. While such macros
80  may save a line or so in the source, it obfuscates the code and makes code
81  review more difficult. It may also result in code which is more complicated
82  than necessary. Use inline functions or just regular functions instead.
83
84* Limit the number of kernel log messages. In general, your driver should not
85  generate an error message just because a runtime operation failed. Report
86  errors to user space instead, using an appropriate error code. Keep in mind
87  that kernel error log messages not only fill up the kernel log, but also are
88  printed synchronously, most likely with interrupt disabled, often to a serial
89  console. Excessive logging can seriously affect system performance.
90
91* Use devres functions whenever possible to allocate resources. For rationale
92  and supported functions, please see Documentation/driver-api/driver-model/devres.rst.
93  If a function is not supported by devres, consider using devm_add_action().
94
95* If the driver has a detect function, make sure it is silent. Debug messages
96  and messages printed after a successful detection are acceptable, but it
97  must not print messages such as "Chip XXX not found/supported".
98
99  Keep in mind that the detect function will run for all drivers supporting an
100  address if a chip is detected on that address. Unnecessary messages will just
101  pollute the kernel log and not provide any value.
102
103* Provide a detect function if and only if a chip can be detected reliably.
104
105* Only the following I2C addresses shall be probed: 0x18-0x1f, 0x28-0x2f,
106  0x48-0x4f, 0x58, 0x5c, 0x73 and 0x77. Probing other addresses is strongly
107  discouraged as it is known to cause trouble with other (non-hwmon) I2C
108  chips. If your chip lives at an address which can't be probed then the
109  device will have to be instantiated explicitly (which is always better
110  anyway.)
111
112* Avoid writing to chip registers in the detect function. If you have to write,
113  only do it after you have already gathered enough data to be certain that the
114  detection is going to be successful.
115
116  Keep in mind that the chip might not be what your driver believes it is, and
117  writing to it might cause a bad misconfiguration.
118
119* Make sure there are no race conditions in the probe function. Specifically,
120  completely initialize your chip and your driver first, then register with
121  the hwmon subsystem.
122
123* Use devm_hwmon_device_register_with_groups() or, if your driver needs a remove
124  function, hwmon_device_register_with_groups() to register your driver with the
125  hwmon subsystem. Try using devm_add_action() instead of a remove function if
126  possible. Do not use hwmon_device_register().
127
128* Your driver should be buildable as module. If not, please be prepared to
129  explain why it has to be built into the kernel.
130
131* Do not provide support for deprecated sysfs attributes.
132
133* Do not create non-standard attributes unless really needed. If you have to use
134  non-standard attributes, or you believe you do, discuss it on the mailing list
135  first. Either case, provide a detailed explanation why you need the
136  non-standard attribute(s).
137  Standard attributes are specified in Documentation/hwmon/sysfs-interface.rst.
138
139* When deciding which sysfs attributes to support, look at the chip's
140  capabilities. While we do not expect your driver to support everything the
141  chip may offer, it should at least support all limits and alarms.
142
143* Last but not least, please check if a driver for your chip already exists
144  before starting to write a new driver. Especially for temperature sensors,
145  new chips are often variants of previously released chips. In some cases,
146  a presumably new chip may simply have been relabeled.
147