linux/Documentation/SubmittingDrivers
<<
>>
Prefs
   1Submitting Drivers For The Linux Kernel
   2---------------------------------------
   3
   4This document is intended to explain how to submit device drivers to the
   5various kernel trees. Note that if you are interested in video card drivers
   6you should probably talk to XFree86 (http://www.xfree86.org/) and/or X.Org
   7(http://x.org/) instead.
   8
   9Also read the Documentation/SubmittingPatches document.
  10
  11
  12Allocating Device Numbers
  13-------------------------
  14
  15Major and minor numbers for block and character devices are allocated
  16by the Linux assigned name and number authority (currently this is
  17Torben Mathiasen). The site is http://www.lanana.org/. This
  18also deals with allocating numbers for devices that are not going to
  19be submitted to the mainstream kernel.
  20See Documentation/devices.txt for more information on this.
  21
  22If you don't use assigned numbers then when your device is submitted it will
  23be given an assigned number even if that is different from values you may
  24have shipped to customers before.
  25
  26Who To Submit Drivers To
  27------------------------
  28
  29Linux 2.0:
  30        No new drivers are accepted for this kernel tree.
  31
  32Linux 2.2:
  33        No new drivers are accepted for this kernel tree.
  34
  35Linux 2.4:
  36        If the code area has a general maintainer then please submit it to
  37        the maintainer listed in MAINTAINERS in the kernel file. If the
  38        maintainer does not respond or you cannot find the appropriate
  39        maintainer then please contact Marcelo Tosatti
  40        <marcelo.tosatti@cyclades.com>.
  41
  42Linux 2.6:
  43        The same rules apply as 2.4 except that you should follow linux-kernel
  44        to track changes in API's. The final contact point for Linux 2.6
  45        submissions is Andrew Morton <akpm@osdl.org>.
  46
  47What Criteria Determine Acceptance
  48----------------------------------
  49
  50Licensing:      The code must be released to us under the
  51                GNU General Public License. We don't insist on any kind
  52                of exclusive GPL licensing, and if you wish the driver
  53                to be useful to other communities such as BSD you may well
  54                wish to release under multiple licenses.
  55                See accepted licenses at include/linux/module.h
  56
  57Copyright:      The copyright owner must agree to use of GPL.
  58                It's best if the submitter and copyright owner
  59                are the same person/entity. If not, the name of
  60                the person/entity authorizing use of GPL should be
  61                listed in case it's necessary to verify the will of
  62                the copyright owner.
  63
  64Interfaces:     If your driver uses existing interfaces and behaves like
  65                other drivers in the same class it will be much more likely
  66                to be accepted than if it invents gratuitous new ones.
  67                If you need to implement a common API over Linux and NT
  68                drivers do it in userspace.
  69
  70Code:           Please use the Linux style of code formatting as documented
  71                in Documentation/CodingStyle. If you have sections of code
  72                that need to be in other formats, for example because they
  73                are shared with a windows driver kit and you want to
  74                maintain them just once separate them out nicely and note
  75                this fact.
  76
  77Portability:    Pointers are not always 32bits, not all computers are little
  78                endian, people do not all have floating point and you
  79                shouldn't use inline x86 assembler in your driver without
  80                careful thought. Pure x86 drivers generally are not popular.
  81                If you only have x86 hardware it is hard to test portability
  82                but it is easy to make sure the code can easily be made
  83                portable.
  84
  85Clarity:        It helps if anyone can see how to fix the driver. It helps
  86                you because you get patches not bug reports. If you submit a
  87                driver that intentionally obfuscates how the hardware works
  88                it will go in the bitbucket.
  89
  90PM support:     Since Linux is used on many portable and desktop systems, your
  91                driver is likely to be used on such a system and therefore it
  92                should support basic power management by implementing, if
  93                necessary, the .suspend and .resume methods used during the
  94                system-wide suspend and resume transitions.  You should verify
  95                that your driver correctly handles the suspend and resume, but
  96                if you are unable to ensure that, please at least define the
  97                .suspend method returning the -ENOSYS ("Function not
  98                implemented") error.  You should also try to make sure that your
  99                driver uses as little power as possible when it's not doing
 100                anything.  For the driver testing instructions see
 101                Documentation/power/drivers-testing.txt and for a relatively
 102                complete overview of the power management issues related to
 103                drivers see Documentation/power/devices.txt .
 104
 105Control:        In general if there is active maintainance of a driver by
 106                the author then patches will be redirected to them unless
 107                they are totally obvious and without need of checking.
 108                If you want to be the contact and update point for the
 109                driver it is a good idea to state this in the comments,
 110                and include an entry in MAINTAINERS for your driver.
 111
 112What Criteria Do Not Determine Acceptance
 113-----------------------------------------
 114
 115Vendor:         Being the hardware vendor and maintaining the driver is
 116                often a good thing. If there is a stable working driver from
 117                other people already in the tree don't expect 'we are the
 118                vendor' to get your driver chosen. Ideally work with the
 119                existing driver author to build a single perfect driver.
 120
 121Author:         It doesn't matter if a large Linux company wrote the driver,
 122                or you did. Nobody has any special access to the kernel
 123                tree. Anyone who tells you otherwise isn't telling the
 124                whole story.
 125
 126
 127Resources
 128---------
 129
 130Linux kernel master tree:
 131        ftp.??.kernel.org:/pub/linux/kernel/...
 132        ?? == your country code, such as "us", "uk", "fr", etc.
 133
 134Linux kernel mailing list:
 135        linux-kernel@vger.kernel.org
 136        [mail majordomo@vger.kernel.org to subscribe]
 137
 138Linux Device Drivers, Third Edition (covers 2.6.10):
 139        http://lwn.net/Kernel/LDD3/  (free version)
 140
 141LWN.net:
 142        Weekly summary of kernel development activity - http://lwn.net/
 143        2.6 API changes:
 144                http://lwn.net/Articles/2.6-kernel-api/
 145        Porting drivers from prior kernels to 2.6:
 146                http://lwn.net/Articles/driver-porting/
 147
 148KernelTrap:
 149        Occasional Linux kernel articles and developer interviews
 150        http://kerneltrap.org/
 151
 152KernelNewbies:
 153        Documentation and assistance for new kernel programmers
 154        http://kernelnewbies.org/
 155
 156Linux USB project:
 157        http://www.linux-usb.org/
 158
 159How to NOT write kernel driver by Arjan van de Ven:
 160        http://www.fenrus.org/how-to-not-write-a-device-driver-paper.pdf
 161
 162Kernel Janitor:
 163        http://janitor.kernelnewbies.org/
 164
lxr.linux.no kindly hosted by Redpill Linpro AS, provider of Linux consulting and operations services since 1995.