[uClibc]Re: Bonehead floating-point question

Erik Andersen andersen at codepoet.org
Thu Mar 13 20:01:27 UTC 2003


On Thu Mar 13, 2003 at 11:19:49AM -0800, David Wuertele wrote:
> >> For example, uClibc Configuration asks at one point:
> >> 
> >> Target CPU has a floating point unit (FPU) (HAS_FPU) [Y/n/?] y
> >> 
> >> Why should it care?  As a matter of fact my target CPU does *not* have
> >> an FPU, but since I've got emulation it shouldn't matter.  What do I
> >> gain/loose by saying "n" here?
> 
> Erik> uClibc cares since if you don't have an FPU I need to make
> Erik> certain that uClibc is compiled with -msoft-float and that
> Erik> all uClibc asm code properly elides fpu opcodes.
> 
> But if I have an emulator, it doesn't matter, right?  I could answer
> 'y' here and the kernel will take care of things?

Yes.  This is described in the HAS_FPU help text which I
wrote to make sure people knew the answer to this question.

HAS_FPU:

  If your target CPU does not have a Floating Point Unit (FPU) or
  a kernel FPU emulator, but you still wish to support floating
  point functions, then uClibc will need to be compiled with soft
  floating point support (-msoft-float).  If your target CPU does
  not have an FPU or an FPU emulator within the Linux kernel,
  then you should answer N.

  Most people will answer Y.

> >> The default buildroot builds gcc "--without-float" ... Why is that
> >> needed?
> 
> Erik> I do not believe that is correct.
> 
> What's not correct, my statement about the default buildroot acts this
> way, or the fact that buildroot *should* build gcc "--without-float"?

The default buildroot does not build gcc with the
"--without-float" configure option.  If that is set, it is
because you enabled it in the top level Makefile.  It is
not the default.

 -Erik

--
Erik B. Andersen             http://codepoet-consulting.com/
--This message was written using 73% post-consumer electrons--



More information about the uClibc mailing list