kernel/common.git - Unnamed repository; edit this file 'description' to name the repository.

1 .. _submittingdrivers:
2
3 Submitting Drivers For The Linux Kernel
4 =======================================
5
6 This document is intended to explain how to submit device drivers to the
7 various kernel trees. Note that if you are interested in video card drivers
8 you should probably talk to XFree86 (http://www.xfree86.org/) and/or X.Org
9 (http://x.org/) instead.
10
11 Also read the Documentation/SubmittingPatches document.
12
13
14 Allocating Device Numbers
15 -------------------------
16
17 Major and minor numbers for block and character devices are allocated
18 by the Linux assigned name and number authority (currently this is
19 Torben Mathiasen). The site is http://www.lanana.org/. This
20 also deals with allocating numbers for devices that are not going to
21 be submitted to the mainstream kernel.
22 See Documentation/devices.txt for more information on this.
23
24 If you don't use assigned numbers then when your device is submitted it will
25 be given an assigned number even if that is different from values you may
26 have shipped to customers before.
27
28 Who To Submit Drivers To
29 ------------------------
30
31 Linux 2.0:
32 	No new drivers are accepted for this kernel tree.
33
34 Linux 2.2:
35 	No new drivers are accepted for this kernel tree.
36
37 Linux 2.4:
38 	If the code area has a general maintainer then please submit it to
39 	the maintainer listed in MAINTAINERS in the kernel file. If the
40 	maintainer does not respond or you cannot find the appropriate
41 	maintainer then please contact Willy Tarreau <w@1wt.eu>.
42
43 Linux 2.6 and upper:
44 	The same rules apply as 2.4 except that you should follow linux-kernel
45 	to track changes in API's. The final contact point for Linux 2.6+
46 	submissions is Andrew Morton.
47
48 What Criteria Determine Acceptance
49 ----------------------------------
50
51 Licensing:
52 		The code must be released to us under the
53 		GNU General Public License. We don't insist on any kind
54 		of exclusive GPL licensing, and if you wish the driver
55 		to be useful to other communities such as BSD you may well
56 		wish to release under multiple licenses.
57 		See accepted licenses at include/linux/module.h
58
59 Copyright:
60 		The copyright owner must agree to use of GPL.
61 		It's best if the submitter and copyright owner
62 		are the same person/entity. If not, the name of
63 		the person/entity authorizing use of GPL should be
64 		listed in case it's necessary to verify the will of
65 		the copyright owner.
66
67 Interfaces:
68 		If your driver uses existing interfaces and behaves like
69 		other drivers in the same class it will be much more likely
70 		to be accepted than if it invents gratuitous new ones.
71 		If you need to implement a common API over Linux and NT
72 		drivers do it in userspace.
73
74 Code:
75 		Please use the Linux style of code formatting as documented
76 		in :ref:`Documentation/CodingStyle <codingStyle>`.
77 		If you have sections of code
78 		that need to be in other formats, for example because they
79 		are shared with a windows driver kit and you want to
80 		maintain them just once separate them out nicely and note
81 		this fact.
82
83 Portability:
84 		Pointers are not always 32bits, not all computers are little
85 		endian, people do not all have floating point and you
86 		shouldn't use inline x86 assembler in your driver without
87 		careful thought. Pure x86 drivers generally are not popular.
88 		If you only have x86 hardware it is hard to test portability
89 		but it is easy to make sure the code can easily be made
90 		portable.
91
92 Clarity:
93 		It helps if anyone can see how to fix the driver. It helps
94 		you because you get patches not bug reports. If you submit a
95 		driver that intentionally obfuscates how the hardware works
96 		it will go in the bitbucket.
97
98 PM support:
99 		Since Linux is used on many portable and desktop systems, your
100 		driver is likely to be used on such a system and therefore it
101 		should support basic power management by implementing, if
102 		necessary, the .suspend and .resume methods used during the
103 		system-wide suspend and resume transitions.  You should verify
104 		that your driver correctly handles the suspend and resume, but
105 		if you are unable to ensure that, please at least define the
106 		.suspend method returning the -ENOSYS ("Function not
107 		implemented") error.  You should also try to make sure that your
108 		driver uses as little power as possible when it's not doing
109 		anything.  For the driver testing instructions see
110 		Documentation/power/drivers-testing.txt and for a relatively
111 		complete overview of the power management issues related to
112 		drivers see Documentation/power/devices.txt .
113
114 Control:
115 		In general if there is active maintenance of a driver by
116 		the author then patches will be redirected to them unless
117 		they are totally obvious and without need of checking.
118 		If you want to be the contact and update point for the
119 		driver it is a good idea to state this in the comments,
120 		and include an entry in MAINTAINERS for your driver.
121
122 What Criteria Do Not Determine Acceptance
123 -----------------------------------------
124
125 Vendor:
126 		Being the hardware vendor and maintaining the driver is
127 		often a good thing. If there is a stable working driver from
128 		other people already in the tree don't expect 'we are the
129 		vendor' to get your driver chosen. Ideally work with the
130 		existing driver author to build a single perfect driver.
131
132 Author:
133 		It doesn't matter if a large Linux company wrote the driver,
134 		or you did. Nobody has any special access to the kernel
135 		tree. Anyone who tells you otherwise isn't telling the
136 		whole story.
137
138
139 Resources
140 ---------
141
142 Linux kernel master tree:
143 	ftp.\ *country_code*\ .kernel.org:/pub/linux/kernel/...
144
145 	where *country_code* == your country code, such as
146 	**us**, **uk**, **fr**, etc.
147
148 	http://git.kernel.org/?p=linux/kernel/git/torvalds/linux.git
149
150 Linux kernel mailing list:
151 	linux-kernel@vger.kernel.org
152 	[mail majordomo@vger.kernel.org to subscribe]
153
154 Linux Device Drivers, Third Edition (covers 2.6.10):
155 	http://lwn.net/Kernel/LDD3/  (free version)
156
157 LWN.net:
158 	Weekly summary of kernel development activity - http://lwn.net/
159
160 	2.6 API changes:
161
162 		http://lwn.net/Articles/2.6-kernel-api/
163
164 	Porting drivers from prior kernels to 2.6:
165
166 		http://lwn.net/Articles/driver-porting/
167
168 KernelNewbies:
169 	Documentation and assistance for new kernel programmers
170
171 		http://kernelnewbies.org/
172
173 Linux USB project:
174 	http://www.linux-usb.org/
175
176 How to NOT write kernel driver by Arjan van de Ven:
177 	http://www.fenrus.org/how-to-not-write-a-device-driver-paper.pdf
178
179 Kernel Janitor:
180 	http://kernelnewbies.org/KernelJanitors
181
182 GIT, Fast Version Control System:
183 	http://git-scm.com/
184
1	.. _submittingdrivers:
2
3	Submitting Drivers For The Linux Kernel
4	=======================================
5
6	This document is intended to explain how to submit device drivers to the
7	various kernel trees. Note that if you are interested in video card drivers
8	you should probably talk to XFree86 (http://www.xfree86.org/) and/or X.Org
9	(http://x.org/) instead.
10
11	Also read the Documentation/SubmittingPatches document.
12
13
14	Allocating Device Numbers
15	-------------------------
16
17	Major and minor numbers for block and character devices are allocated
18	by the Linux assigned name and number authority (currently this is
19	Torben Mathiasen). The site is http://www.lanana.org/. This
20	also deals with allocating numbers for devices that are not going to
21	be submitted to the mainstream kernel.
22	See Documentation/devices.txt for more information on this.
23
24	If you don't use assigned numbers then when your device is submitted it will
25	be given an assigned number even if that is different from values you may
26	have shipped to customers before.
27
28	Who To Submit Drivers To
29	------------------------
30
31	Linux 2.0:
32	No new drivers are accepted for this kernel tree.
33
34	Linux 2.2:
35	No new drivers are accepted for this kernel tree.
36
37	Linux 2.4:
38	If the code area has a general maintainer then please submit it to
39	the maintainer listed in MAINTAINERS in the kernel file. If the
40	maintainer does not respond or you cannot find the appropriate
41	maintainer then please contact Willy Tarreau <w@1wt.eu>.
42
43	Linux 2.6 and upper:
44	The same rules apply as 2.4 except that you should follow linux-kernel
45	to track changes in API's. The final contact point for Linux 2.6+
46	submissions is Andrew Morton.
47
48	What Criteria Determine Acceptance
49	----------------------------------
50
51	Licensing:
52	The code must be released to us under the
53	GNU General Public License. We don't insist on any kind
54	of exclusive GPL licensing, and if you wish the driver
55	to be useful to other communities such as BSD you may well
56	wish to release under multiple licenses.
57	See accepted licenses at include/linux/module.h
58
59	Copyright:
60	The copyright owner must agree to use of GPL.
61	It's best if the submitter and copyright owner
62	are the same person/entity. If not, the name of
63	the person/entity authorizing use of GPL should be
64	listed in case it's necessary to verify the will of
65	the copyright owner.
66
67	Interfaces:
68	If your driver uses existing interfaces and behaves like
69	other drivers in the same class it will be much more likely
70	to be accepted than if it invents gratuitous new ones.
71	If you need to implement a common API over Linux and NT
72	drivers do it in userspace.
73
74	Code:
75	Please use the Linux style of code formatting as documented
76	in :ref:`Documentation/CodingStyle <codingStyle>`.
77	If you have sections of code
78	that need to be in other formats, for example because they
79	are shared with a windows driver kit and you want to
80	maintain them just once separate them out nicely and note
81	this fact.
82
83	Portability:
84	Pointers are not always 32bits, not all computers are little
85	endian, people do not all have floating point and you
86	shouldn't use inline x86 assembler in your driver without
87	careful thought. Pure x86 drivers generally are not popular.
88	If you only have x86 hardware it is hard to test portability
89	but it is easy to make sure the code can easily be made
90	portable.
91
92	Clarity:
93	It helps if anyone can see how to fix the driver. It helps
94	you because you get patches not bug reports. If you submit a
95	driver that intentionally obfuscates how the hardware works
96	it will go in the bitbucket.
97
98	PM support:
99	Since Linux is used on many portable and desktop systems, your
100	driver is likely to be used on such a system and therefore it
101	should support basic power management by implementing, if
102	necessary, the .suspend and .resume methods used during the
103	system-wide suspend and resume transitions. You should verify
104	that your driver correctly handles the suspend and resume, but
105	if you are unable to ensure that, please at least define the
106	.suspend method returning the -ENOSYS ("Function not
107	implemented") error. You should also try to make sure that your
108	driver uses as little power as possible when it's not doing
109	anything. For the driver testing instructions see
110	Documentation/power/drivers-testing.txt and for a relatively
111	complete overview of the power management issues related to
112	drivers see Documentation/power/devices.txt .
113
114	Control:
115	In general if there is active maintenance of a driver by
116	the author then patches will be redirected to them unless
117	they are totally obvious and without need of checking.
118	If you want to be the contact and update point for the
119	driver it is a good idea to state this in the comments,
120	and include an entry in MAINTAINERS for your driver.
121
122	What Criteria Do Not Determine Acceptance
123	-----------------------------------------
124
125	Vendor:
126	Being the hardware vendor and maintaining the driver is
127	often a good thing. If there is a stable working driver from
128	other people already in the tree don't expect 'we are the
129	vendor' to get your driver chosen. Ideally work with the
130	existing driver author to build a single perfect driver.
131
132	Author:
133	It doesn't matter if a large Linux company wrote the driver,
134	or you did. Nobody has any special access to the kernel
135	tree. Anyone who tells you otherwise isn't telling the
136	whole story.
137
138
139	Resources
140	---------
141
142	Linux kernel master tree:
143	ftp.\ country_code\ .kernel.org:/pub/linux/kernel/...
144
145	where country_code == your country code, such as
146	us, uk, fr, etc.
147
148	http://git.kernel.org/?p=linux/kernel/git/torvalds/linux.git
149
150	Linux kernel mailing list:
151	linux-kernel@vger.kernel.org
152	[mail majordomo@vger.kernel.org to subscribe]
153
154	Linux Device Drivers, Third Edition (covers 2.6.10):
155	http://lwn.net/Kernel/LDD3/ (free version)
156
157	LWN.net:
158	Weekly summary of kernel development activity - http://lwn.net/
159
160	2.6 API changes:
161
162	http://lwn.net/Articles/2.6-kernel-api/
163
164	Porting drivers from prior kernels to 2.6:
165
166	http://lwn.net/Articles/driver-porting/
167
168	KernelNewbies:
169	Documentation and assistance for new kernel programmers
170
171	http://kernelnewbies.org/
172
173	Linux USB project:
174	http://www.linux-usb.org/
175
176	How to NOT write kernel driver by Arjan van de Ven:
177	http://www.fenrus.org/how-to-not-write-a-device-driver-paper.pdf
178
179	Kernel Janitor:
180	http://kernelnewbies.org/KernelJanitors
181
182	GIT, Fast Version Control System:
183	http://git-scm.com/
184