1 |
1626 |
jcastillo |
Sound driver version v3.0 (and later)
|
2 |
|
|
-------------------------------------
|
3 |
|
|
|
4 |
|
|
All features of v2.90-2 should work as earlier. There could be some
|
5 |
|
|
omissions but they are unintentional. I started this version thread
|
6 |
|
|
after v2.3 so all features implemented before it are there.
|
7 |
|
|
|
8 |
|
|
New features
|
9 |
|
|
============
|
10 |
|
|
|
11 |
|
|
There are now two new device interfaces. The /dev/midi## is a raw
|
12 |
|
|
tty like interface to MIDI ports. There is a device file for each MIDI
|
13 |
|
|
port on your system. They are named (/dev/midi00 to /dev/midiNN).
|
14 |
|
|
The second addition is the /dev/music which is higher level interface
|
15 |
|
|
than the old /dev/sequencer. It's intended for writing device independent
|
16 |
|
|
applications like sequencers.
|
17 |
|
|
|
18 |
|
|
/dev/midi##
|
19 |
|
|
-----------
|
20 |
|
|
|
21 |
|
|
This interface should be useful for applications like MIDI sysex librarians.
|
22 |
|
|
There are (currently) no timing features so making music could be impossible.
|
23 |
|
|
|
24 |
|
|
There are as many /dev/midi## devices as there are MIDI ports in the system.
|
25 |
|
|
The /dev/midi00 is connected to the first one, /dev/midi01 to the second etc.
|
26 |
|
|
|
27 |
|
|
These devices work like tty devices in raw mode. Everything written to them is
|
28 |
|
|
sent out to the MIDI port. There is currently an extra delay of at most
|
29 |
|
|
1/100th of sec but it will be removed later.
|
30 |
|
|
|
31 |
|
|
The reading algorithm is little bit more complicated. There are two different
|
32 |
|
|
cases:
|
33 |
|
|
|
34 |
|
|
1) There is at least one byte in the input buffer.
|
35 |
|
|
|
36 |
|
|
The read returns as many bytes as it can without waiting for more bytes.
|
37 |
|
|
For example when a process reads 100 bytes and there are 10 bytes in the
|
38 |
|
|
buffer, the read returns just 10 bytes.
|
39 |
|
|
|
40 |
|
|
2) The input buffer is empty when the process calls read.
|
41 |
|
|
|
42 |
|
|
The read waits for the first byte and then continues as in case 1. By
|
43 |
|
|
default it waits infinitely but there is an ioctl for setting a timeout
|
44 |
|
|
for this. The ioctl(fd, SNDCTL_MIDI_PRETIME, &time) changes the timeout.
|
45 |
|
|
The time is given in 1/10th of seconds (10 means one second).
|
46 |
|
|
|
47 |
|
|
Other ioctl calls:
|
48 |
|
|
|
49 |
|
|
ioctl(fd, SNDCTL_MIDI_MPUMODE, &mode) is available for full MPU-401
|
50 |
|
|
compatible devices such as MPU-IPC-T, MQ PC Midi Card or MQX-32.
|
51 |
|
|
It's not available for the so called MPU UART ports of some soundcards
|
52 |
|
|
(PAS16, SB16 etc). By default the MIDI port is in UART mode after open.
|
53 |
|
|
If this ioctl is called with mode=1, the interface is put to the intelligent
|
54 |
|
|
(coprocessor) mode. NOTE! The MIDI port will be reset when this ioctl is called.
|
55 |
|
|
It could have some strange effects if not called immediately after open. This
|
56 |
|
|
call returns EINVAL if the midi port doesn't support the MPU-401 intelligent
|
57 |
|
|
mode.
|
58 |
|
|
|
59 |
|
|
ioctl(fd, SNDCTL_MIDI_MPUCMD, &cmdstruct) is valid only if the MIDI port
|
60 |
|
|
is put to the coprocessor mode using ioctl(SNDCTL_MIDI_MPUMODE). It's used to
|
61 |
|
|
send commands to a MPU-401 compatible MIDI cards. Please refer to the
|
62 |
|
|
MPU-401 Technical Reference Manual (or Music Quest Technical Reference
|
63 |
|
|
Manual) for descriptions of the commands.
|
64 |
|
|
|
65 |
|
|
The argument of SNDCTL_MIDI_MPUCOMMAND is of type mpu_command_rec. It
|
66 |
|
|
has the following fields:
|
67 |
|
|
|
68 |
|
|
typedef struct {
|
69 |
|
|
unsigned char cmd;
|
70 |
|
|
|
71 |
|
|
char nr_args, nr_returns;
|
72 |
|
|
unsigned char data[30];
|
73 |
|
|
} mpu_command_rec;
|
74 |
|
|
|
75 |
|
|
where:
|
76 |
|
|
cmd Contains the command number.
|
77 |
|
|
nr_args Number of arguments of the command.
|
78 |
|
|
MUST BE INITIALIZED BEFORE CALL
|
79 |
|
|
nr_returns Number of bytes returned by the command.
|
80 |
|
|
MUST BE INITIALIZED BEFORE CALL
|
81 |
|
|
data Buffer for the command arguments and returned
|
82 |
|
|
data.
|
83 |
|
|
|
84 |
|
|
Be extremely careful with the nr_args and nr_returns fields. They
|
85 |
|
|
must match the command. An incorrect value will put the card and
|
86 |
|
|
the driver out of sync. Refer to the MPU-401/MQX-32M documentation for further
|
87 |
|
|
details.
|
88 |
|
|
|
89 |
|
|
|
90 |
|
|
|
91 |
|
|
/dev/music (/dev/sequencer2)
|
92 |
|
|
----------------------------
|
93 |
|
|
|
94 |
|
|
This device file works much like the /dev/sequencer which has been present
|
95 |
|
|
since the beginning. The main differences are the following:
|
96 |
|
|
|
97 |
|
|
- /dev/sequencer makes the MIDI ports to look like the synth devices. In fact
|
98 |
|
|
the result is somewhere between the MIDI specification and the synth devices of
|
99 |
|
|
/dev/sequencer. Both kind of devices are accessed using the SEQ_START_NOTE()
|
100 |
|
|
like macros. The voice number parameters of the API macros have been redefined
|
101 |
|
|
to denote MIDI channels. This means that the driver allocates voices for
|
102 |
|
|
the channels automatically (this is a responsibility/right of an application
|
103 |
|
|
with /dev/sequencer). The result is that a SEQ_START_NOTE() macro has
|
104 |
|
|
similar effects for a synth channel than on a MIDI port. This kind of
|
105 |
|
|
solution provides better device independence than the /dev/sequencer. The
|
106 |
|
|
drawback is that the new interface doesn't permit so low level access to the
|
107 |
|
|
device as the /dev/sequencer does. An application developer must choose between
|
108 |
|
|
these two interfaces. I think the old /dev/sequencer is better for applications
|
109 |
|
|
like module players while the new one is better for making generic sequencer
|
110 |
|
|
programs.
|
111 |
|
|
|
112 |
|
|
- There are no separate MIDI devices with the /dev/sequencer2. The
|
113 |
|
|
ioctl(SNDCTL_SEQ_NRMIDIS) returns always zero. Instead the MIDI ports are
|
114 |
|
|
shown as synth devices. ioctl(SNDCTL_SEQ_NRSYNTHS) on /dev/sequencer2 will
|
115 |
|
|
return sum of internal synthesizers (GUS, OPL3) and MIDI ports in the systems.
|
116 |
|
|
|
117 |
|
|
- The new interface is used much like the ordinary /dev/sequencer. The
|
118 |
|
|
event format is new so you have to use the API macros defined in the
|
119 |
|
|
sys/soundcard.h. The interface is will probably change before the final 3.0
|
120 |
|
|
release but using the API macros should ensure compatibility in source level.
|
121 |
|
|
The new event format is not recognized by version 2.X so don't try to
|
122 |
|
|
distribute binaries compiled with soundcard.h of v3.X.
|
123 |
|
|
|
124 |
|
|
- The basic API usage is similar to the current one. There are some new
|
125 |
|
|
macros but the older ones should work as earlier. The most important
|
126 |
|
|
incompatibility is that the /dev/sequencer2 driver allocates voices itself.
|
127 |
|
|
The other one is that the application must send SEQ_START_TIMER() as its
|
128 |
|
|
first event. Otherwise the timer is not started and the application waits
|
129 |
|
|
infinitely.
|
130 |
|
|
|
131 |
|
|
|
132 |
|
|
There are several new features but I don't document them here. There are
|
133 |
|
|
some info in the soundcard.h (near the end). I have also included some
|
134 |
|
|
sample code in the directory v30. Full documentation will
|
135 |
|
|
appear in the Hacker's Guide later.
|
136 |
|
|
|
137 |
|
|
Don't hesitate to contact me in case you have questions or comments.
|
138 |
|
|
|
139 |
|
|
Hannu Savolainen
|
140 |
|
|
hannu@4front-tech.com
|