1 |
On 2:58:58 pm 11/11/04 Ned Ludd <solar@g.o> wrote: |
2 |
> On Thu, 2004-11-11 at 09:23, Peter S. Mazinger wrote: |
3 |
> > On Thu, 11 Nov 2004, Ned Ludd wrote: |
4 |
> > |
5 |
> > > > X (tested older XFree86 and xorg-x11-6.8.1) can't be built |
6 |
> > > > against uclibc for now (missing math functions) |
7 |
> > > > |
8 |
> > > > solar,vapier: |
9 |
> > > > I could add to the masked uclibc |
10 |
> > > > use X && (apply X needed math funcs) |
11 |
> > > > and/or |
12 |
> > > > use java && (apply sun-jdk math funcs) |
13 |
> > > > the java condition is not quite true, they are needed to build |
14 |
> > > sun-j2sdk |
15 |
> > > Well here is what I'm thinking in this respect. mjn3 is working |
16 |
> > > on a new libm but it's going to be a while before he is done. |
17 |
> > > Reading the uClibc/TODO states that libm is targeted for 1.0 |
18 |
> > > which we know is going to be a ways out. So if we have |
19 |
> > > replacement functions then lets post them here for review and |
20 |
> > > consider patching out own -lm till such time as mjn3 has it |
21 |
> > > ready and it works for us. |
22 |
> > > http://uclibc.org/cgi-bin/cvsweb/uClibc/TODO?&view=auto |
23 |
> > > TODO list for the uClibc 1.0.0 release: |
24 |
> > > *) Implement the float and long double versions of math funcs |
25 |
> > > using wrappers on top of the double versions (size / |
26 |
> > > precision trade off where size clearly wins). |
27 |
> > |
28 |
> > look in the current uclibc patches tarball subdir math, those are |
29 |
> > that I needed until now, they are edited versions of net/openbsd |
30 |
> > (as almost the whole original libm stuff in uclibc) |
31 |
> > sinf/cosf/tanf/atan2f/sqrtf/powf/fabsf are implemented |
32 |
> |
33 |
> |
34 |
> > I'll add IUSE="X" and use X && epatch patch/math to the new uclibc |
35 |
> ebuild |
36 |
> USE=X is not logical and should not be the controlling factor which |
37 |
> decide if libm is built. |
38 |
> |
39 |
> Please consider. |
40 |
> |
41 |
> if use build || use bootstrap; then |
42 |
> # It's ok to skip it here. |
43 |
> else |
44 |
> # It really should be enabled here. |
45 |
> fi |
46 |
|
47 |
I like this method more than using the X use flag. I've actually run into a |
48 |
few other packages that tried to use those math funcs. (although I can't |
49 |
remember which ones off the top of my head). |
50 |
|
51 |
--Iggy |
52 |
|
53 |
> |
54 |
> |
55 |
> -- |
56 |
> Ned Ludd <solar@g.o> |
57 |
> Gentoo (hardened,security,infrastructure,embedded) Developer |
58 |
> |
59 |
|
60 |
|
61 |
-- |
62 |
gentoo-embedded@g.o mailing list |