[PATCH] Encode native floating point types in data layout string.

Charlie Turner charlesturner7c5 at gmail.com
Fri Dec 26 04:58:03 PST 2014


Hey Cameron!

On 25 December 2014 at 10:53, Cameron Esfahani <dirty at apple.com> wrote:

>>> ARM: 16-bit (if supported), single, and double (if supported) precision float.
>>
>> I wouldn't have said 16-bit was native. It'll get promoted to float
>> one way or the other for any actual operations. You just might be
>> lucky enough that there are instructions to do that conversion for
>> you.
>>
>
> I've been using Wikipedia for most of the processors that I don't know well.  In this case, I looked more carefully and went to ARM directly.  According to: http://infocenter.arm.com/help/index.jsp?topic=/com.arm.doc.dui0205j/CIHGAECI.html half-precision floating point is an extension to VFPv3.  Does hasFP16 not reflect whether this extension is present?

Half-precision FP is a storage format only. This extension to VFP
units just gives you conversion instructions in hardware (see the
VCVTB instruction for example), not half-precision arithmetic
operations.

So hasFP16 will reflect whether the conversion instructions are
present, I also don't agree that this would count for "native" 16-bit
float support.

Kind regards,
Charlie.




More information about the llvm-commits mailing list