Commit | Line | Data |
760ac839 |
1 | # This file has been put together by Anno Siegel <siegel@zrz.TU-Berlin.DE>, |
2 | # Andreas Koenig <k@franz.ww.TU-Berlin.DE> and Gerd Knops <gerti@BITart.com>. |
3 | # Comments, questions, and improvements welcome! |
4aa0a1f7 |
4 | # |
5 | # These hints work for NeXT 3.2 and 3.3. 3.0 has it's own |
6 | # special hint file. |
760ac839 |
7 | # |
4aa0a1f7 |
8 | |
760ac839 |
9 | ccflags='-DUSE_NEXT_CTYPE -DUSE_PERL_SBRK -DHIDEMYMALLOC' |
75f92628 |
10 | POSIX_cflags='ccflags="-posix $ccflags"' |
85e6fe83 |
11 | ldflags='-u libsys_s' |
a0d0e21e |
12 | libswanted='dbm gdbm db' |
5d94fbed |
13 | |
61d11634 |
14 | lddlflags='-nostdlib -r' |
5d94fbed |
15 | # Give cccdlflags an empty value since Configure will detect we are |
16 | # using GNU cc and try to specify -fpic for cccdlflags. |
17 | cccdlflags=' ' |
18 | |
760ac839 |
19 | # |
20 | # Change the line below if you do not want to build 'quad-fat' |
21 | # binaries |
22 | # |
40ea3355 |
23 | archs=`/bin/lipo -info /usr/lib/libm.a | sed 's/^[^:]*:[^:]*: //'` |
24 | for d in $archs |
25 | do |
26 | mab="$mab -arch $d" |
27 | done |
28 | |
760ac839 |
29 | |
61d11634 |
30 | archname='next-fat' |
31 | ld='cc' |
32 | |
a0d0e21e |
33 | i_utime='undef' |
34 | groupstype='int' |
35 | direntrytype='struct direct' |
36 | d_strcoll='undef' |
760ac839 |
37 | |
38 | ###################################################################### |
39 | # THE MALLOC STORY |
40 | ###################################################################### |
41 | # 1994: |
a0d0e21e |
42 | # the simple program `for ($i=1;$i<38771;$i++){$t{$i}=123}' fails |
43 | # with Larry's malloc on NS 3.2 due to broken sbrk() |
760ac839 |
44 | # |
45 | # setting usemymalloc='n' was the solution back then. Later came |
46 | # reports that perl would run unstable on 3.2: |
47 | # |
48 | # From about perl5.002beta1h perl became unstable on the |
49 | # NeXT. Intermittent coredumps were frequent on 3.2 OS. There were |
50 | # reports, that the developer version of 3.3 didn't have problems, so it |
51 | # seemed pretty obvious that we had to work around an malloc bug in 3.2. |
52 | # This hints file reflects a patch to perl5.002_01 that introduces a |
53 | # home made sbrk routine (remember, NeXT's sbrk _never_ worked). This |
54 | # sbrk makes it possible to run perl with its own malloc. Thanks to |
55 | # Ilya who showed me the way to his sbrk for OS/2!! |
56 | # andreas koenig, 1996-06-16 |
57 | # |
58 | # So, this hintsfile is using perl's malloc. If you want to turn perl's |
59 | # malloc off, you need to change remove '-DUSE_PERL_SBRK' and |
60 | # '-DHIDEMYMALLOC' from the ccflags above and set usemymalloc below |
61 | # to 'n'. |
62 | # |
63 | ###################################################################### |
64 | usemymalloc='y' |
65 | |
a0d0e21e |
66 | d_uname='define' |
67 | d_setpgid='define' |
68 | d_setsid='define' |
69 | d_tcgetpgrp='define' |
70 | d_tcsetpgrp='define' |
760ac839 |
71 | |
ecfc5424 |
72 | # |
73 | # On some NeXT machines, the timestamp put by ranlib is not correct, and |
74 | # this may cause useless recompiles. Fix that by adding a sleep before |
75 | # running ranlib. The '5' is an empirical number that's "long enough." |
760ac839 |
76 | # |
ecfc5424 |
77 | ranlib='sleep 5; /bin/ranlib' |
760ac839 |
78 | |
40000a8c |
79 | # |
80 | # There where reports that the compiler on HPPA machines |
81 | # fails with the -O flag on pp.c. |
40000a8c |
82 | pp_cflags='optimize="-g"' |