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
|
This file describes how to install the Berkeley MPEG encoder. Installation
should be a quick and painless task. Just do the following in order:
1) The encoder uses jpeg code. You don't have to use this if you don't
want. The jpeg code is only useful if you have images stored in JPEG
format. And even if you do, you don't have to use it (you can use
the INPUT_CONVERT option); but it's faster and better if you do use it.
Here's what you do: cd into the jpeg/ directory and type 'make' there.
That should create libjpeg.a. That's it. (Well, if you dont have
gcc, you'll need to edit the Makefile there). If you have problems
making it, or don't want to bother, you'll have to go into the Makefile
in THIS directory and change all the JPEG stuff. Look for the comment
"JPEG STUFF" and make everything there = nothing, then do this:
MP_JPEG_OBJS = nojpeg.o
MP_JPEG_SRCS = nojpeg.c
Or, if you already have libjpeg.a (v4!), change
JPEG_LIB = -ljpeg
JPEG_LIB_DIR = -Ljpeg-lib-path
(and you can delete the jpeg subdir to save space)
If you have a lot of trouble compiling the jpeg directory,
run ./configure there.
2) If your machine does not have UNIX sockets, then you cannot use
the parallel version of the encoder. The default is to not
have a parallel version. If you wish to run on several machines,
Edit the Makefile, changing MP_PARALLEL_OBJS and MP_PARALLEL_SRCS.
Instead of noparallel.c and noparallel.o, they should be parallel.c
and parallel.o.
3) We assume that right shift corresponds to signed division by 2 with
rounding towards minus infinity. This is correct for typical "arithmetic
shift" instructions that shift in copies of the sign bit. But some
C compilers implement >> with an unsigned shift. For these machines you
must define RIGHT_SHIFT_IS_UNSIGNED. To do so, add
-DRIGHT_SHIFT_IS_UNSIGNED to your CFLAGS line.
4) Edit the Makefile. Set CC accordingly. Set CFLAGS
accordingly. Some sample CFLAGS lines for several different
compilers/machines are given in the Makefile. If long's on your
machine are 32 bits but int's are NOT 32 bits, then you must include
the -DLONG_32 option in your CFLAGS line. If your system does not
have the netinet/in.h header, then you must add either -DFORCE_BIG_ENDIAN
or -DFORCE_LITTLE_ENDIAN to the CFLAGS line (which you use depends on
your machine).
5) Set the LIBS lines appropriately. For most machines you don't need
to change this. However, for some machines you'll need to include
additional libraries to use sockets.
6) If you are using a NON-ANSI compiler, then your CFLAGS should
include -DNON_ANSI_COMPILER
7) Type 'make clean; make all'
If all is well, you should get the executable file 'mpeg_encode.' To
test it, try 'make test.' At the end of the test, it should say SUCCESS
or FAILURE. If it says FAILURE, please let us know!
Note: if you get warnings about certain C library procedures not being
defined before being called, that's not our fault. Prototypes
*should* exist in the appropriate system header files in /usr/include,
but on some systems they don't. None of this should affect the
correctness of the program, but if you get tired of the warnings, you
can probably stop them by setting PROTOFLAG = -DMISSING_PROTOTYPES in
the Makefile.
|
