1 |
Honza Macháček schrieb: |
2 |
> Justin wrote: |
3 |
> |
4 |
>> Does anyone has rfftw.h? It should be in /usr/include/ ... |
5 |
> |
6 |
> Does any package need it? I don't have that file and nothing ever |
7 |
> complained, as far as I remember. After all, drfftw.h or srfftw.h can |
8 |
> allways be included specifically. |
9 |
> |
10 |
> That said, the symlik belongs -- or rather should belong -- to |
11 |
> fftw2, fftw3 does not structure its headers that way; I have installed |
12 |
> both sci-libs/fftw-2.1.5-r3 and sci-libs/fftw-3.1.2, bud qdepends -Q |
13 |
> says nothing depends on the 2 series version (that may be why nothing |
14 |
> complained), and only sci-mathematics/octave-3.0.0 on the 3 series one. |
15 |
> |
16 |
> From fftw-2.1.5-r3.ebuild: |
17 |
> |
18 |
>> if use float; then |
19 |
>> for f in "${D}"/usr/{include,$(get_libdir)}/*sfft*; do |
20 |
>> ln -s $(basename ${f}) ${f/sfft/fft} |
21 |
>> done |
22 |
>> else |
23 |
>> for f in "${D}"/usr/{include,$(get_libdir)}/*dfft*; do |
24 |
>> ln -s $(basename ${f}) ${f/dfft/fft} |
25 |
>> done |
26 |
>> fi |
27 |
> |
28 |
> That code does not create the rfftw.h symlink, but looks like it |
29 |
> actually was intended to do so and the omission is an ebuild error. |
30 |
> Definitely if float is in USE, srfftw.h should be linked, otherwise |
31 |
> drfftw.h. And of course it is you in the end who should decide if you |
32 |
> want single or double precision variant to be used by default. |
33 |
> |
34 |
> With best regards |
35 |
> Honza Macháček |
36 |
> |
37 |
ccp4 depends on fftw-2. And the build has an error because of not |
38 |
finding the rfftw.h. It's not fatal, but there mide be some functions |
39 |
broken. |
40 |
It should be linked with and without "float". It looks like the ebuild |
41 |
doesn' t do it. |