cachepc-linux

Fork of AMDESE/linux with modifications for CachePC side-channel attack
git clone https://git.sinitax.com/sinitax/cachepc-linux
Log | Files | Refs | README | LICENSE | sfeed.txt

submitting-patches.rst (6880B)


      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* Please use the standard multi-line comment style. Do not mix C and C++
     24  style comments in a single driver (with the exception of the SPDX license
     25  identifier).
     26
     27* If your patch generates checkpatch errors, warnings, or check messages,
     28  please refrain from explanations such as "I prefer that coding style".
     29  Keep in mind that each unnecessary message helps hiding a real problem,
     30  and a consistent coding style makes it easier for others to understand
     31  and review the code.
     32
     33* Please test your patch thoroughly. We are not your test group.
     34  Sometimes a patch can not or not completely be tested because of missing
     35  hardware. In such cases, you should test-build the code on at least one
     36  architecture. If run-time testing was not achieved, it should be written
     37  explicitly below the patch header.
     38
     39* If your patch (or the driver) is affected by configuration options such as
     40  CONFIG_SMP, make sure it compiles for all configuration variants.
     41
     42
     432. Adding functionality to existing drivers
     44-------------------------------------------
     45
     46* Make sure the documentation in Documentation/hwmon/<driver_name>.rst is up to
     47  date.
     48
     49* Make sure the information in Kconfig is up to date.
     50
     51* If the added functionality requires some cleanup or structural changes, split
     52  your patch into a cleanup part and the actual addition. This makes it easier
     53  to review your changes, and to bisect any resulting problems.
     54
     55* Never mix bug fixes, cleanup, and functional enhancements in a single patch.
     56
     57
     583. New drivers
     59--------------
     60
     61* Running your patch or driver file(s) through checkpatch does not mean its
     62  formatting is clean. If unsure about formatting in your new driver, run it
     63  through Lindent. Lindent is not perfect, and you may have to do some minor
     64  cleanup, but it is a good start.
     65
     66* Consider adding yourself to MAINTAINERS.
     67
     68* Document the driver in Documentation/hwmon/<driver_name>.rst.
     69
     70* Add the driver to Kconfig and Makefile in alphabetical order.
     71
     72* Make sure that all dependencies are listed in Kconfig.
     73
     74* Please list include files in alphabetic order.
     75
     76* Please align continuation lines with '(' on the previous line.
     77
     78* Avoid forward declarations if you can. Rearrange the code if necessary.
     79
     80* Avoid macros to generate groups of sensor attributes. It not only confuses
     81  checkpatch, but also makes it more difficult to review the code.
     82
     83* Avoid calculations in macros and macro-generated functions. While such macros
     84  may save a line or so in the source, it obfuscates the code and makes code
     85  review more difficult. It may also result in code which is more complicated
     86  than necessary. Use inline functions or just regular functions instead.
     87
     88* Limit the number of kernel log messages. In general, your driver should not
     89  generate an error message just because a runtime operation failed. Report
     90  errors to user space instead, using an appropriate error code. Keep in mind
     91  that kernel error log messages not only fill up the kernel log, but also are
     92  printed synchronously, most likely with interrupt disabled, often to a serial
     93  console. Excessive logging can seriously affect system performance.
     94
     95* Use devres functions whenever possible to allocate resources. For rationale
     96  and supported functions, please see Documentation/driver-api/driver-model/devres.rst.
     97  If a function is not supported by devres, consider using devm_add_action().
     98
     99* If the driver has a detect function, make sure it is silent. Debug messages
    100  and messages printed after a successful detection are acceptable, but it
    101  must not print messages such as "Chip XXX not found/supported".
    102
    103  Keep in mind that the detect function will run for all drivers supporting an
    104  address if a chip is detected on that address. Unnecessary messages will just
    105  pollute the kernel log and not provide any value.
    106
    107* Provide a detect function if and only if a chip can be detected reliably.
    108
    109* Only the following I2C addresses shall be probed: 0x18-0x1f, 0x28-0x2f,
    110  0x48-0x4f, 0x58, 0x5c, 0x73 and 0x77. Probing other addresses is strongly
    111  discouraged as it is known to cause trouble with other (non-hwmon) I2C
    112  chips. If your chip lives at an address which can't be probed then the
    113  device will have to be instantiated explicitly (which is always better
    114  anyway.)
    115
    116* Avoid writing to chip registers in the detect function. If you have to write,
    117  only do it after you have already gathered enough data to be certain that the
    118  detection is going to be successful.
    119
    120  Keep in mind that the chip might not be what your driver believes it is, and
    121  writing to it might cause a bad misconfiguration.
    122
    123* Make sure there are no race conditions in the probe function. Specifically,
    124  completely initialize your chip and your driver first, then register with
    125  the hwmon subsystem.
    126
    127* Use devm_hwmon_device_register_with_info() or, if your driver needs a remove
    128  function, hwmon_device_register_with_info() to register your driver with the
    129  hwmon subsystem. Try using devm_add_action() instead of a remove function if
    130  possible. Do not use hwmon_device_register().
    131
    132* Your driver should be buildable as module. If not, please be prepared to
    133  explain why it has to be built into the kernel.
    134
    135* Do not provide support for deprecated sysfs attributes.
    136
    137* Do not create non-standard attributes unless really needed. If you have to use
    138  non-standard attributes, or you believe you do, discuss it on the mailing list
    139  first. Either case, provide a detailed explanation why you need the
    140  non-standard attribute(s).
    141  Standard attributes are specified in Documentation/hwmon/sysfs-interface.rst.
    142
    143* When deciding which sysfs attributes to support, look at the chip's
    144  capabilities. While we do not expect your driver to support everything the
    145  chip may offer, it should at least support all limits and alarms.
    146
    147* Last but not least, please check if a driver for your chip already exists
    148  before starting to write a new driver. Especially for temperature sensors,
    149  new chips are often variants of previously released chips. In some cases,
    150  a presumably new chip may simply have been relabeled.