/
TODO
224 lines (156 loc) · 6.7 KB
/
TODO
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
Background
==========
- Priority scale: High, Medium and Low
- Complexity scale: C1, C2, C4 and C8. The complexity scale is exponential,
with complexity 1 being the lowest complexity. Complexity is a function
of both task 'complexity' and task 'scope'.
The general rule of thumb is that a complexity 1 task should take 1-2 weeks
for a person very familiar with BlueZ codebase. Higher complexity tasks
require more time and have higher uncertainty.
Higher complexity tasks should be refined into several lower complexity tasks
once the task is better understood.
General
=======
- UUID handling: Use the new functions created for UUID handling in all parts
of BlueZ code. Currently, the new bt_uuid_* functions are being used by
GATT-related code only.
Priority: high
Complexity: C4
- Update PBAP client/server implementation to 1.2 and create necessary APIs for
new features it introduces.
Priority: Medium
Complexity: C4
- Create GOEP unit tests based on its test specification:
https://www.bluetooth.org/docman/handlers/DownloadDoc.ashx?doc_id=230559
Priority: Medium
Complexity: C2
- Function in src/adapter.c to convert old storage files to new ini-file format
should be removed 6-8 months after first BlueZ 5 release.
Priority: Low
Complexity: C1
- Remove usage of symlinks for drivers, such as profiles/input/suspend.c and
profiles/sap/sap.c. Instead, select drivers at runtime by using config
options or probing for running D-Bus services (using e.g.
g_dbus_add_service_watch()). Idea first mentioned on
http://thread.gmane.org/gmane.linux.bluez.kernel/30175/focus=30190.
- Reuse connection handling code of src/profile.c also for built-in profiles
so plugins would only need to register their btd_profile and the core takes
care of the rest including listen to the right channel and manages the sdp
record. Once btd_profile manages the connection it can also notify about
their state, this probably remove the need of having callbacks to
.connect/.disconnect since their state can be tracked, it also enables any
plugin to track any profile state change which can be useful for e.g.
a connection policy plugin in case one is needed.
Priority: Low
Complexity: C2
- Add queueing support for src/agent.c, currently if there is any request
pending the code fail with error EBUSY which is very inconvenient.
Priority: Low
Complexity: C2
Low Energy
==========
- Connection modes. Adapter interface needs to be changed to manage
connection modes and adapter type. See Volume 3, Part C, section 9.3.
1. Mode management: Peripheral / Central
Priority: Medium
Complexity: C2
- Advertising data. The D-Bus interface needs to be updated to enable setting
scan response data, and to read the advertising and scan response data which
has been broadcast from other LE devices.
Priority: Medium
Complexity: C2
- Static random address setup and storage. Once this address is written
in a given remote, the address can not be changed anymore.
Priority: Low
Complexity: C1
- Device Name Characteristic is a GAP characteristic for Low Energy. This
characteristic shall be integrated/used in the discovery procedure. The
idea is to report the value of this characteristic using DeviceFound signals.
Discussion with the community is needed before to start this task. Other GAP
characteristics for LE needs to follow a similar approach. It is not clear
if all GAP characteristics can be exposed using properties instead of a primary
service characteristics.
See Volume 3, Part C, section 12.1 for more information.
Priority: Low
Complexity: C2
ATT/GATT (new shared stack)
===========================
- Add complete GATT test coverage in unit/test-gatt following the GATT test
spec. This could use shared/gatt-client and shared/gatt-server at the same
time to test both against eachother. We should definitely have tests for
gatt-server and gatt-client simultaneously on one side of the connection.
Priority: High
Complexity: C4
- Write an example using client D-Bus API using C.
Priority: High
Complexity: C2
- Write an example using client D-Bus API using python.
Priority: High
Complexity: C2
- Define packed structs for ATT protocol PDUs in shared/att-types to improve
readability. We should probably do this once there are extensive unit tests
for gatt-client/gatt-server so that we don't accidentally break working code.
Priority: Medium
Complexity: C2
- Use struct iovec to pass around byte buffers that will be sent over the wire,
instead of passing uint8_t and size_t parameters everywhere.
Priority: Medium
Complexity: C1
- Move all daemon plugins and profiles that are GATT based to use
shared/gatt-client instead of attrib/*. This is a complicated task that
potentially needs a new plugin/profile probing interface and a lot of
rewriting that can cause regressions in existing functionality.
Priority: Medium
Complexity: C4
- Introduce a way for shared/gatt-server to check security permissions on the
current connection through bt_att.
Priority: Medium
Complexity: C2
- Implement other low-priority ATT protocol operations for shared/gatt-server:
Read Multiple Request
Priority: Low
Complexity: C1
- Send out indications from the "Service Changed" characteristic upon
reconnection if a bonded device is not connected when the local database is
modified.
Priority: High
Complexity: C2
- Unify the GATT server and client D-Bus implementations into a single module.
While these don't share a lot of code, keeping them all in src/gatt-dbus seems
to make more sense from an organizational perspective.
Priority: Low
Complexity: C1
- Isolate all GATT code inside the daemon into its own module and perform
interaction with other modules (e.g. src/device.c) via callbacks. This
includes client/server management, tracking incoming/outgoing connections for
ATT, and callbacks to perform profile probing.
Priority: Low
Complexity: C4
- Support included services in the GATT D-Bus client API.
Priority: Medium
Complexity: C1
Management Interface
====================
Mesh
====
- Add unit tests
Priority: High
Complexity: C1
- Implement unified feature management mechanism for mesh nodes that are hosted
on the same device
Priority: Medium
Complexity: C2
- Implement mandatory Health Server model. Most likely, this will involve
additions to mesh D-Bus API
Priority: Medium
Complexity: C2
- Add support for GATT proxy server
Priority: Low
Complexity: C4
- Merge common functionality from tools/mesh. Ideally, source code from the
tools/mesh directory should completely dissapear.
Priority: Low
Complexity: C2
- Support for Low Power Node (LPN)
Priority: Low
Complexity: C4