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
|
BUGS
====
This file describes several bugs known as of today. There do exist probably
some more, but I'm not aware of them right now. Do not hesitate to send me
<freddy@snarl.nl> your bug-reports though. I will gladly add them to this file
and maybe even solve some of them. :-)
Order of bytes broke when reading several tty devices at once
=============================================================
The display of incoming bytes can be broke if you use multiple tty devices to
read from. At the moment I do not have a solution for this problem. Since I use
select() to watch the several tty devices and after the select() I have to
read() them one by one, I can not completely 100% display which bytes came after
which on different tty devices. Take the example below:
$ jpnevulator --ascii --timing-print --tty /dev/ttyS0 --tty /dev/ttyUSB0 --read
2006-05-30 13:23:49.461075: /dev/ttyS0
00 00 05 3B 0D 00 00 05 ...;....
2006-05-30 13:23:49.461113: /dev/ttyUSB0
00 05 3B 0D 00 00 05 3B 0D ..;....;.
2006-05-30 13:23:49.473074: /dev/ttyS0
3B 0D 00 00 05 3B 0D ;....;.
2006-05-30 13:23:49.473105: /dev/ttyUSB0
00 12 05 06 39 00 12 05 06 39 1F 00 22 80 00 0E ....9....9.."...
$
And now see the order in which things really got sent on the line:
/dev/ttyS0:
00 00 05 3B 0D
/dev/ttyUSB0:
00 00 05 3B 0D
/dev/ttyS0:
00 00 05 3B 0D
/dev/ttyUSB0:
00 00 05 3B 0D
/dev/ttyS0:
00 00 05 3B 0D
/dev/ttyUSB0:
00 00 05 3B 0D 00 12 05 06 39 00 12 05 06 39 ...
As you can see /dev/ttyUSB0 receives the echo of all things sent by /dev/ttyS0.
This is exactly what happens. But since there does exist a small time between
the select() who is happy expressing something is available and the read() who
does get the available data, some extra data will be available. I have no idea
on how I can use high level system call like select() and read() and be still
able to put the bytes in the correct order. Anyone an idea?
|
