[Mono-dev] Incoming changes to Mono.Simd
stapostol at gmail.com
Thu Dec 4 05:23:11 EST 2008
On Thu, Dec 4, 2008 at 3:37 AM, Jonathan Pryor <jonpryor at vt.edu> wrote:
> On Thu, 2008-12-04 at 02:26 +0200, StApostol wrote:
> > Extension methods are great for simplifying the API, but Mono.Simd
> > should still be usable by C# 2.0 programs.
> Silly question, but why? Since Mono.Simd will only be accelerated under
> Mono, and Mono supports C# 3, I don't see much use for the restriction.
> Furthermore, even if you did want to use Mono.Simd under .NET, and you
> wanted to use VS.NET 2005 (.NET 2.0), that's still no reason to avoid
> using extension methods -- you just need to bundle Mono's
> System.Core.dll with your app (which will allow .NET to resolve the
> ExtensionAttribute type), and invoke the extension method as a static
> method (extension methods can still be called as static methods, they
> don't need to be called as instance methods).
> So I really don't see the point in avoiding extension methods.
> - Jon
> Our library is designed to support VS2005+ and Mono 1.2.6+ with a single
dll. While the System.Core.dll approach works, it's unsupported and not
really suitable for production.
That said, I can certainly see your point and I agree that Mono.Simd should
use the capabilities of the current Mono release (C# 3 in this case).
One last question: if one builds a custom version of Mono.Simd (or even
integrates it in another assembly), will the runtime still be able to
resolve and accelerate the operations?
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Mono-devel-list