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
76
77
78
79
80
81
|
Subject: dynamic loading on NeXTSTEP
From: horst@il.us.swissbank.com (Ted Horst)
To: python-list@cwi.nl
Date: Fri, 6 Sep 1996 15:02:26 GMT
X-Newsgroups: comp.lang.python
X-Organization: Swiss Bank Corporation CM&T Division
These are the steps that I took to get dynamic loading working under NeXTSTEP
(3.3 m68k, i486) for Python1.4beta3 (including the Numeric stuff).
Unfortunately, I don't know anything about the configure scripts, so I don't
know how to fix these things properly, but this works for me. I would be
interested in how other people have done this as well.
/configure --prefix=/home/horst/Unix --with-readline=/home/horst/Unix/lib
--with-thread --with-next-archs='m68k i486' --without-libm
edit Python/importdl.c
under:
#ifdef USE_RLD
add:
#define SHORT_EXT ".a"
#define LONG_EXT "module.a"
edit Modules/Setup
uncomment *shared*
(other editing as desired)
make (just to create the Makefiles)
edit Makefiles
change all AR= ar to AR = libtool
change all $(AR) cr ... to $(AR) -o ...
in Modules/Makefile change:
libm =
LDSHARED= cc -r -nostdlib $(CFLAGS)
CCSHARED= -r -nostdlib
LINKFORSHARED= -u libsys_s
make "OPT=-O3 -fschedule-insns2 -ObjC -arch m68k -arch i486"
Ted Horst
(not speaking for Swissbank)
Subject: Re: dynamic loading on NeXTSTEP
From: horst@il.us.swissbank.com (Ted Horst)
To: python-list@cwi.nl
Date: Fri, 6 Sep 1996 16:27:32 GMT
X-Newsgroups: comp.lang.python
X-Organization: Swiss Bank Corporation CM&T Division
Ted Horst writes
>
> These are the steps that I took to get dynamic loading working under NeXTSTEP
> (3.3 m68k, i486) for Python1.4beta3 (including the Numeric stuff).
<SNIP>
> make "OPT=-O3 -fschedule-insns2 -ObjC -arch m68k -arch i486"
>
> Ted Horst
> (not speaking for Swissbank)
Wow, I guess I was overzealous with the optimizations. Upon further
testing, I discovered that the Numeric stuff did not work properly on
Intel machines, but did work properly on m68k. I dropped the
optimization to -O2 (leaving out the -fschedule-insns2), and it seems
to behave properly on both architectures now.
So,
make "OPT=-O2 -ObjC -arch m68k -arch i486"
looks like a better choice.
Sorry for the misguidance,
Ted Horst
(still not speaking for SwissBank)
|
