wcstold implementation

[Corinna Vinschen]

On Jan 23 15:57, Nicholas Clifton wrote:
> Hi Corinna,
> 
> >As an extension, for targets striving for smaller code size
> >(PREFER_SIZE_OVER_SPEED or __OPTIMIZE_SIZE__), but double != long double,
> >would it make sense to implement strtod in terms of strtold?
> 
> Hmm - are there any such targets ?
> 
> Anyway I think that in principle this might sound like a good idea, but in
> practice it probably will be bad.  The reason being that for targets where
> long double != double, and where code size matters, it is quite likely that
> double arithmetic is implemented in hardware (and hence does not take up a
> lot of room in applications) whereas long double support is almost certainly
> implemented in software and so it quite bulky.  Thus having a strtod()
> function which pulls in strtold() and the associate software long double
> support would be a bad thing, not a good thing.

Good point.


Thanks,
Corinna

-- 
Corinna Vinschen
Cygwin Maintainer
Red Hat
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 819 bytes
Desc: not available
URL: <http://sourceware.org/pipermail/newlib/attachments/20150123/88ffc462/attachment.sig>