|
Boost : |
Subject: Re: [boost] [Multiprecision] Benchmarking
From: Vicente Botet (vicente.botet_at_[hidden])
Date: 2012-06-15 07:24:22
John Maddock-3 wrote
>
>>I agree, the library should be able to decide whether expression templates
>>should be an improvement or not.
>>
>>A minor change in the library, defaulting the ExpressionTemplates
parameter
>>to a trait depending on the backend could help to that.
>>The library could default to something reasonable, while the backend
>>developer could always provide a specialization.
>
> That would certainly be easy to do, but:
>
> * That wouldn't help for types like cpp_int where the best strategy may
> depend on the runtime size of the number.
>
>
>>
>> The backend developer could define the trait depending on the
>> compile-time size. For run-time size the library set use a default, and
>> it will be up to the user to force or not the use of expression
>> templates.
>>
>>
> * How would this differ in practice from the use of the typedefs we
> already
> provide where we know that expression templates don't help (mp_intXXX_t
> etc)?
>
Well, you will not have specific typedefs for all the sizes, so a way to
associate the default value will be convenient.
Vicente
-- View this message in context: http://boost.2283326.n4.nabble.com/Multiprecision-Benchmarking-tp4631293p4631330.html Sent from the Boost - Dev mailing list archive at Nabble.com.
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk