[Mono-dev] [PATCH] Optimize Mono.Simd non-accelerated bitwise-ops

[Paolo Molaro]

On 11/10/08 Cedric Vivier wrote:
> This patch implements these ops as two 64-bit bitwise ops [2] instead
> of processing

Sadly this is not safe as the alignment we currently can guarantee is
only to 4 bytes and a few architectures don't support loading
a 64 bit value unless it's 8-byte aligned. Could you implement it using
uints instead?

lupus

-- 
-----------------------------------------------------------------
lupus at debian.org                                     debian/rules
lupus at ximian.com                             Monkeys do it better