[cfe-dev] similar clang options for : -mno-80387 -mno-fp-ret-in-387

[Reid Kleckner]

No, we don't support these options, and I don't think we'll want to
implement -mno-fp-ret-in-387.  To my knowledge, nobody actually needs to
return floating point values in GPRs.  What the Linux kernel probably wants
is to completely disable floating point.  In an ideal world, both gcc and
clang would grow an -f[no-]floating-point flag, which would reject floating
point code in the frontend.


On Wed, Apr 23, 2014 at 2:37 AM, Jan-Simon Möller <dl9pf at gmx.de> wrote:

>
> Hi !
>
> Are there options similar to:
>
>         -mno-80387:
>
>            On machines where a function returns floating-point results in
> the
>            80387
>            register stack, some floating-point opcodes may be emitted even
> if
>            -msoft-float is used.
>
>        -mno-fp-ret-in-387
>            Do not use the FPU registers for return values of functions.
>
>            The usual calling convention has functions return values of
> types
>            "float"
>            and "double" in an FPU register, even if there is no FPU.  The
> idea
>            is that
>            the operating system should emulate an FPU.
>
>            The option -mno-fp-ret-in-387 causes such values to be returned
> in
>            ordinary
>            CPU registers instead.
>
>
> Or what combination of flags makes sure we won't hit these quirks in the
> first
> place ?
>
> Thanks.
>
> --
>
> Dipl.-Ing.
> Jan-Simon Möller
>
> jansimon.moeller at gmx.de
>
> _______________________________________________
> cfe-dev mailing list
> cfe-dev at cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/cfe-dev
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/cfe-dev/attachments/20140424/16a4ff1b/attachment.html>