> LEADZ was part of the HPF spec. If your compiler supports (or used to
> support) HPF, it may already have a routine in the supplied intrinsics
> library. You could probably call such a routine directly with the right
> interface. Check for symbols in the library containing the string 'leadz'.
For your interest, and since I can't be bothered reading all the
documentation or trying it...
Sun Studio 11
tromba-5.10:lib% nm *.so | grep -i leadz
[3512] | 94676| 136|FUNC |GLOB |0 |11 |__f_leadz_g
[3516] | 94828| 180|FUNC |GLOB |0 |11 |__f_leadz_h
[3518] | 95024| 184|FUNC |GLOB |0 |11 |__f_leadz_i
[3521] | 95224| 364|FUNC |GLOB |0 |11 |__f_leadz_l
[466] | 0| 0|FILE |LOCL |0 |ABS |leadz.c
PGI 6.1-6
s0001:lib% nm *.a | grep -i leadz
0000000000003746 T pghpf_leadz
0000000000003888 T pghpf_leadz_i8
000000000000373b T pghpf_leadz_
0000000000003872 T pghpf_leadz_i8_
Pathscale 2.5
nm: f77wrappers.o: no symbols
U _leadz
U _leadz
U _leadz
000000000000016f T _leadz
U __leadz8
U _leadz
U __leadz8
000000000000010a T __leadz1
000000000000014e T __leadz2
0000000000000192 T __leadz4
00000000000001d5 T __leadz8
nm: f77wrappers.pg.o: no symbols
U _leadz
U _leadz
U _leadz
0000000000000239 T _leadz
U __leadz8
U _leadz
U __leadz8
0000000000000164 T __leadz1
00000000000001b7 T __leadz2
0000000000000209 T __leadz4
000000000000025a T __leadz8
Intel 8.1
Some for_xxleadz symbols where xxx varies
Absoft 9?
s0001:lib% nm *.a | grep -i leadz
0000000000000320 t _leadz4
0000000000000460 t _leadz8
00000000000002c0 t _leadz4
0000000000000400 t _leadz8
0000000000000320 t _leadz4
0000000000000460 t _leadz8
leadz.o:
0000000000000100 T _LEADZ
00000000000000e0 T _LEADZ_4
0000000000000040 T _LEADZ_I4
Harvey
|