Software floating point support

[GoogleCodeExporter]

Architectures with no floating point support could benefit from a software
floating point library. Also, architectures with no math library could
benefit from a math library.

Original issue reported on code.google.com by zik.sale...@gmail.com on 28 May 2009 at 4:05

[GoogleCodeExporter]

Math library added for architectures with have none.
Now I just need to get soft float running.

Original comment by zik.sale...@gmail.com on 29 May 2009 at 1:41

• Changed state: Started

[GoogleCodeExporter]

bugged on bigendian!!!

Original comment by flameman...@gmail.com on 14 Sep 2009 at 8:25

[GoogleCodeExporter]

i can test your code on

* hppa pa1..pa2
* mips-be (R3000..R12000)
* mips-le (R4000)
* m68k (68000..68060)
* ppc (ppc603,ppc405gp, ppc750...ppc7550)
* intel x86 (486,p4)

let me know any endianless fixup, i will test for you

Original comment by flameman...@gmail.com on 15 Sep 2009 at 1:28

[GoogleCodeExporter]

Have decided against introducing software floating point. This should probably 
be in
the C compiler, not PicoC. That way it can be used in the whole system rather 
than
just in PicoC.

Original comment by zik.sale...@gmail.com on 4 Oct 2009 at 2:07

[GoogleCodeExporter]

Closing this for now. Will open a bug for big-endian bug.

Original comment by zik.sale...@gmail.com on 4 Oct 2009 at 2:44

• Changed state: Fixed

[GoogleCodeExporter]

Big-endian bug now in issue #44

Original comment by zik.sale...@gmail.com on 4 Oct 2009 at 2:46