[MPlayer-cvslog] r19207 - trunk/libmpdemux/freesdp/parser.c

[Rich Felker]

On Thu, Jul 27, 2006 at 11:02:44PM +0200, diego wrote:
> Author: diego
> Date: Thu Jul 27 23:02:44 2006
> New Revision: 19207
> 
> Modified:
>    trunk/libmpdemux/freesdp/parser.c
> 
> Log:
> The C99 function strtof is unavailable in some BSD variants.
> patch by Jan Knutar, jknutar $@$ nic fi

IMO we should work out some sort of policy about this stuff. I can't
imagine strtof being implemented anyway but as a call to strtold
casting the result down to float.. Are there really platforms we
support where float arithmetic is faster than double (or long double)
arithmetic?

I heard at one time i387 arithmetic was faster in single precision
mode but that the current ABI leaves the fpu in extended precision
mode all the time regardless of operand size. Anyone know if this is
true (or on which cpus)? If so it might be worthwhile to throw in some
inline asm to switch the fpu mode for codecs and stuff that certainly
don't need more than single precision.

Rich