[Cfp-interest] Fwd: (SC22WG14.13013) RE: Reset for TS 18661 Part 3?

[James W Thomas]

Begin forwarded message:

> From: "Joseph S. Myers" <jsm at polyomino.org.uk>
> Subject: (SC22WG14.13013) RE: Reset for TS 18661 Part 3?
> Date: June 4, 2013 12:41:33 PM PDT
> To: "Nelson, Clark" <clark.nelson at intel.com>
> Cc: SC22 WG14 <sc22wg14 at open-std.org>
> 
> On Tue, 4 Jun 2013, Nelson, Clark wrote:
> 
>> Only after the Delft meeting ended, I realized that the FP standard is 
>> using that word in a way that's somewhat idiosyncratic, but nevertheless 
>> reasonable. They're not really talking about the ability to move binary 
>> data between implementations, with meaningful results (which is the 
>> usual sense of "interchange" in my experience). Instead I think they're 
>> talking only about the ability to move floating-point algorithms and 
>> calculations between implementations, with consistent results. That 
>> seems to be implied by the fact that the FP standard doesn't talk about 
>> byte ordering.
> 
> The relevant wording in IEEE 754-2008 is in 3.1.2, "A conforming 
> implementation of a supported interchange format shall provide means to 
> read and write that format using a specific encoding defined in this 
> clause, for that format.".  Encodings are defined as bit-strings while C 
> works in units of bytes (the floating-point standard never mentions 
> bytes); I think a C binding needs to say how the C bytes relate to the 
> bit-strings for the standard encodings, just as it needs to say what C 
> operators and functions correspond to what standard floating-point 
> operations.
> 
> That goes together with 3.1.1, "These interchange formats are identified 
> by their size (see 3.6) and can be used for the exchange of floating-point 
> data between implementations." and 3.6, "Interchange formats support the 
> exchange of floating-point data between implementations.".
> 
> -- 
> Joseph S. Myers
> joseph at codesourcery.com

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.oakapple.net/pipermail/cfp-interest/attachments/20130604/db77f1bb/attachment.html