1 ==================== 1 ====================
2 How FunctionFS works 2 How FunctionFS works
3 ==================== 3 ====================
4 4
5 Overview <<
6 ======== <<
7 <<
8 From kernel point of view it is just a composi 5 From kernel point of view it is just a composite function with some
9 unique behaviour. It may be added to an USB c 6 unique behaviour. It may be added to an USB configuration only after
10 the user space driver has registered by writin 7 the user space driver has registered by writing descriptors and
11 strings (the user space program has to provide 8 strings (the user space program has to provide the same information
12 that kernel level composite functions provide 9 that kernel level composite functions provide when they are added to
13 the configuration). 10 the configuration).
14 11
15 This in particular means that the composite in 12 This in particular means that the composite initialisation functions
16 may not be in init section (ie. may not use th 13 may not be in init section (ie. may not use the __init tag).
17 14
18 From user space point of view it is a file sys 15 From user space point of view it is a file system which when
19 mounted provides an "ep0" file. User space dr 16 mounted provides an "ep0" file. User space driver need to
20 write descriptors and strings to that file. I 17 write descriptors and strings to that file. It does not need
21 to worry about endpoints, interfaces or string 18 to worry about endpoints, interfaces or strings numbers but
22 simply provide descriptors such as if the func 19 simply provide descriptors such as if the function was the
23 only one (endpoints and strings numbers starti 20 only one (endpoints and strings numbers starting from one and
24 interface numbers starting from zero). The Fu 21 interface numbers starting from zero). The FunctionFS changes
25 them as needed also handling situation when nu 22 them as needed also handling situation when numbers differ in
26 different configurations. 23 different configurations.
27 24
28 For more information about FunctionFS descript <<
29 <<
30 When descriptors and strings are written "ep#" 25 When descriptors and strings are written "ep#" files appear
31 (one for each declared endpoint) which handle 26 (one for each declared endpoint) which handle communication on
32 a single endpoint. Again, FunctionFS takes ca 27 a single endpoint. Again, FunctionFS takes care of the real
33 numbers and changing of the configuration (whi 28 numbers and changing of the configuration (which means that
34 "ep1" file may be really mapped to (say) endpo 29 "ep1" file may be really mapped to (say) endpoint 3 (and when
35 configuration changes to (say) endpoint 2)). 30 configuration changes to (say) endpoint 2)). "ep0" is used
36 for receiving events and handling setup reques 31 for receiving events and handling setup requests.
37 32
38 When all files are closed the function disable 33 When all files are closed the function disables itself.
39 34
40 What I also want to mention is that the Functi 35 What I also want to mention is that the FunctionFS is designed in such
41 a way that it is possible to mount it several 36 a way that it is possible to mount it several times so in the end
42 a gadget could use several FunctionFS function 37 a gadget could use several FunctionFS functions. The idea is that
43 each FunctionFS instance is identified by the 38 each FunctionFS instance is identified by the device name used
44 when mounting. 39 when mounting.
45 40
46 One can imagine a gadget that has an Ethernet, 41 One can imagine a gadget that has an Ethernet, MTP and HID interfaces
47 where the last two are implemented via Functio 42 where the last two are implemented via FunctionFS. On user space
48 level it would look like this:: 43 level it would look like this::
49 44
50 $ insmod g_ffs.ko idVendor=<ID> iSerialNumbe 45 $ insmod g_ffs.ko idVendor=<ID> iSerialNumber=<string> functions=mtp,hid
51 $ mkdir /dev/ffs-mtp && mount -t functionfs 46 $ mkdir /dev/ffs-mtp && mount -t functionfs mtp /dev/ffs-mtp
52 $ ( cd /dev/ffs-mtp && mtp-daemon ) & 47 $ ( cd /dev/ffs-mtp && mtp-daemon ) &
53 $ mkdir /dev/ffs-hid && mount -t functionfs 48 $ mkdir /dev/ffs-hid && mount -t functionfs hid /dev/ffs-hid
54 $ ( cd /dev/ffs-hid && hid-daemon ) & 49 $ ( cd /dev/ffs-hid && hid-daemon ) &
55 50
56 On kernel level the gadget checks ffs_data->de 51 On kernel level the gadget checks ffs_data->dev_name to identify
57 whether its FunctionFS is designed for MTP ("m 52 whether its FunctionFS is designed for MTP ("mtp") or HID ("hid").
58 53
59 If no "functions" module parameters is supplie 54 If no "functions" module parameters is supplied, the driver accepts
60 just one function with any name. 55 just one function with any name.
61 56
62 When "functions" module parameter is supplied, 57 When "functions" module parameter is supplied, only functions
63 with listed names are accepted. In particular, 58 with listed names are accepted. In particular, if the "functions"
64 parameter's value is just a one-element list, 59 parameter's value is just a one-element list, then the behaviour
65 is similar to when there is no "functions" at 60 is similar to when there is no "functions" at all; however,
66 only a function with the specified name is acc 61 only a function with the specified name is accepted.
67 62
68 The gadget is registered only after all the de 63 The gadget is registered only after all the declared function
69 filesystems have been mounted and USB descript 64 filesystems have been mounted and USB descriptors of all functions
70 have been written to their ep0's. 65 have been written to their ep0's.
71 66
72 Conversely, the gadget is unregistered after t 67 Conversely, the gadget is unregistered after the first USB function
73 closes its endpoints. 68 closes its endpoints.
74 <<
75 DMABUF interface <<
76 ================ <<
77 <<
78 FunctionFS additionally supports a DMABUF base <<
79 userspace can attach DMABUF objects (externall <<
80 and subsequently use them for data transfers. <<
81 <<
82 A userspace application can then use this inte <<
83 objects between several interfaces, allowing i <<
84 zero-copy fashion, for instance between IIO an <<
85 <<
86 As part of this interface, three new IOCTLs ha <<
87 IOCTLs have to be performed on a data endpoint <<
88 <<
89 ``FUNCTIONFS_DMABUF_ATTACH(int)`` <<
90 Attach the DMABUF object, identified by it <<
91 data endpoint. Returns zero on success, an <<
92 on error. <<
93 <<
94 ``FUNCTIONFS_DMABUF_DETACH(int)`` <<
95 Detach the given DMABUF object, identified <<
96 from the data endpoint. Returns zero on su <<
97 errno value on error. Note that closing th <<
98 descriptor will automatically detach all a <<
99 <<
100 ``FUNCTIONFS_DMABUF_TRANSFER(struct usb_ffs_ <<
101 Enqueue the previously attached DMABUF to <<
102 The argument is a structure that packs the <<
103 the size in bytes to transfer (which shoul <<
104 the size of the DMABUF), and a 'flags' fie <<
105 for now. Returns zero on success, and a ne <<
106 error. <<
Linux® is a registered trademark of Linus Torvalds in the United States and other countries.
TOMOYO® is a registered trademark of NTT DATA CORPORATION.