]>
Commit | Line | Data |
---|---|---|
83d290c5 | 1 | # SPDX-License-Identifier: GPL-2.0+ |
0f605c15 SG |
2 | # |
3 | # (C) Copyright 2014 Google, Inc | |
4 | # Simon Glass <[email protected]> | |
0f605c15 | 5 | |
0f605c15 SG |
6 | Background |
7 | ---------- | |
8 | ||
89b199c3 SG |
9 | U-Boot traditionally had a board.c file for each architecture. This introduced |
10 | quite a lot of duplication, with each architecture tending to do | |
0f605c15 | 11 | initialisation slightly differently. To address this, a new 'generic board |
89b199c3 | 12 | init' feature was introduced in March 2013 (further motivation is |
0f605c15 SG |
13 | provided in the cover letter below). |
14 | ||
89b199c3 SG |
15 | All boards and architectures have moved to this as of mid 2016. |
16 | ||
0f605c15 SG |
17 | |
18 | What has changed? | |
19 | ----------------- | |
20 | ||
89b199c3 | 21 | The main change is that the arch/<arch>/lib/board.c file is removed in |
0f605c15 SG |
22 | favour of common/board_f.c (for pre-relocation init) and common/board_r.c |
23 | (for post-relocation init). | |
24 | ||
bb5a2cf9 | 25 | Related to this, the global_data and bd_info structures now have a core set of |
0f605c15 SG |
26 | fields which are common to all architectures. Architecture-specific fields |
27 | have been moved to separate structures. | |
28 | ||
29 | ||
0f605c15 SG |
30 | Further Background |
31 | ------------------ | |
32 | ||
33 | The full text of the original generic board series is reproduced below. | |
34 | ||
35 | --8<------------- | |
36 | ||
37 | This series creates a generic board.c implementation which contains | |
38 | the essential functions of the major arch/xxx/lib/board.c files. | |
39 | ||
40 | What is the motivation for this change? | |
41 | ||
42 | 1. There is a lot of repeated code in the board.c files. Any change to | |
43 | things like setting up the baud rate requires a change in 10 separate | |
44 | places. | |
45 | ||
46 | 2. Since there are 10 separate files, adding a new feature which requires | |
47 | initialisation is painful since it must be independently added in 10 | |
48 | places. | |
49 | ||
a560ad70 RR |
50 | 3. As time goes by the architectures naturally diverge since there is limited |
51 | pressure to compare features or even CONFIG options against similar things | |
0f605c15 SG |
52 | in other board.c files. |
53 | ||
54 | 4. New architectures must implement all the features all over again, and | |
a560ad70 | 55 | sometimes in subtle different ways. This places an unfair burden on getting |
0f605c15 SG |
56 | a new architecture fully functional and running with U-Boot. |
57 | ||
58 | 5. While it is a bit of a tricky change, I believe it is worthwhile and | |
59 | achievable. There is no requirement that all code be common, only that | |
60 | the code that is common should be located in common/board.c rather than | |
61 | arch/xxx/lib/board.c. | |
62 | ||
63 | All the functions of board_init_f() and board_init_r() are broken into | |
64 | separate function calls so that they can easily be included or excluded | |
65 | for a particular architecture. It also makes it easier to adopt Graeme's | |
66 | initcall proposal when it is ready. | |
67 | ||
68 | http://lists.denx.de/pipermail/u-boot/2012-January/114499.html | |
69 | ||
70 | This series removes the dependency on generic relocation. So relocation | |
71 | happens as one big chunk and is still completely arch-specific. See the | |
72 | relocation series for a proposed solution to this for ARM: | |
73 | ||
74 | http://lists.denx.de/pipermail/u-boot/2011-December/112928.html | |
75 | ||
76 | or Graeme's recent x86 series v2: | |
77 | ||
78 | http://lists.denx.de/pipermail/u-boot/2012-January/114467.html | |
79 | ||
80 | Instead of moving over a whole architecture, this series takes the approach | |
81 | of simply enabling generic board support for an architecture. It is then up | |
82 | to each board to opt in by defining CONFIG_SYS_GENERIC_BOARD in the board | |
83 | config file. If this is not done, then the code will be generated as | |
84 | before. This allows both sets of code to co-exist until we are comfortable | |
85 | with the generic approach, and enough boards run. | |
86 | ||
87 | ARM is a relatively large board.c file and one which I can test, therefore | |
88 | I think it is a good target for this series. On the other hand, x86 is | |
89 | relatively small and simple, but different enough that it introduces a | |
90 | few issues to be solved. So I have chosen both ARM and x86 for this series. | |
91 | After a suggestion from Wolfgang I have added PPC also. This is the | |
92 | largest and most feature-full board, so hopefully we have all bases | |
93 | covered in this RFC. | |
94 | ||
95 | A generic global_data structure is also required. This might upset a few | |
96 | people. Here is my basic reasoning: most fields are the same, all | |
97 | architectures include and need it, most global_data.h files already have | |
98 | #ifdefs to select fields for a particular SOC, so it is hard to | |
99 | see why architecures are different in this area. We can perhaps add a | |
100 | way to put architecture-specific fields into a separate header file, but | |
101 | for now I have judged that to be counter-productive. | |
102 | ||
103 | Similarly we need a generic bd_info structure, since generic code will | |
104 | be accessing it. I have done this in the same way as global_data and the | |
105 | same comments apply. | |
106 | ||
107 | There was dicussion on the list about passing gd_t around as a parameter | |
108 | to pre-relocation init functions. I think this makes sense, but it can | |
109 | be done as a separate change, and this series does not require it. | |
110 | ||
111 | While this series needs to stand on its own (as with the link script | |
112 | cleanup series and the generic relocation series) the goal is the | |
113 | unification of the board init code. So I hope we can address issues with | |
114 | this in mind, rather than focusing too narrowly on particular ARM, x86 or | |
115 | PPC issues. | |
116 | ||
117 | I have run-tested ARM on Tegra Seaboard only. To try it out, define | |
118 | CONFIG_SYS_GENERIC_BOARD in your board file and rebuild. Most likely on | |
119 | x86 and PPC at least it will hang, but if you are lucky it will print | |
120 | something first :-) | |
121 | ||
122 | I have run this though MAKEALL with CONFIG_SYS_GENERIC_BOARD on for all | |
123 | ARM, PPC and x86 boards. There are a few failures due to errors in | |
124 | the board config, which I have sent patches for. The main issue is | |
125 | just the difference between __bss_end and __bss_end__. | |
126 | ||
127 | Note: the first group of commits are required for this series to build, | |
128 | but could be separated out if required. I have included them here for | |
129 | convenience. | |
130 | ||
131 | ------------->8-- | |
132 | ||
133 | Simon Glass, [email protected] | |
134 | March 2014 | |
89b199c3 | 135 | Updated after final removal, May 2016 |