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
|
$Id: README.beos,v 1.2 1998/05/02 01:55:04 dogcow Exp dogcow $
Notes on building perl under BeOS:
GENERAL ISSUES
--------------
perl will almost compile straight out of the box with ./Configure -d, but
there are a few gotchas:
Currently, you have to edit config.sh and remove SDBM_File from the
dynamic_ext= and extensions= lines. SDBM_File does not build properly
at this time. You need to run ./Configure -S after editing config.sh.
In addition, with mwcc, after doing `make depend`, you need to edit
makefile and x2p/makefile and remove the lines that mention 'Bletch:'.
This is not necessary if you're using gnu cpp.
in short:
./Configure -d
remove SDBM_File from config.sh
./Configure -S
make depend
remove Bletch: from makefile and x2p/makefile
make
Other than that, perl should build without problems. There are some
technical comments in hints/beos.sh.
OS RELEASE-SPECIFIC NOTES
-------------------------
PR1/PPC:
See R3/X86. Same bug, different form.
PR2/PPC:
Signals are somewhat unreliable, but they can work. Use caution.
The POSIX module is still somewhat buggy.
R3/X86:
Under R3 x86, there are some serious problems with the math routines
such that numbers are incorrectly printed. This causes problems with
modules that encode their version numbers - in particular, IO.pm will
probably not work properly. This should be fixed under R3.1.
The problem has manifested itself if you see something similar to the
following during the compile:
cc -c -I/usr/local/include -O -DVERSION=\"1.1504\" -DXS_VERSION=\"1.1499999999\" -fpic -I../.. IO.c
(lots of 9's are the indication of the problem.)
In the meantime, you can use the following workaround:
make perl
cd ext/IO
cc -c -I/usr/local/include -O -DVERSION=\"1.1504\" -DXS_VERSION=\"1.15\" -fpic -I../.. IO.c
cd ..
make
(Substitute the correct numbers if IO has been updated.)
R3/PPC-
There's math problems, but of a different kind. In particular,
perl -e 'print (240000 - (3e4<<3))' gives a non-zero answer.
I'm looking into this. There is no workaround as yet. Hopefully,
this will be fixed in R3.1.
CONTACT INFORMATION
-------------------
If you have comments, problem reports, or even patches or bugfixes (gasp!)
please email me.
1 May 1998
Tom Spindler
dogcow@merit.edu
|
