[Mono-dev] [PATCH] Optimize Mono.Simd non-accelerated bitwise-ops
cedricv at neonux.com
Mon Nov 10 00:50:07 EST 2008
While playing with Mono.Simd on a (for now ;( ) non-accelerated x86-64
setup , I've
noticed that performance of bitwise-ops on Vector16b and others could
This patch implements these ops as two 64-bit bitwise ops  instead
each vector element separately. Giving up to 4X perf increase for
Vector16(s)b and 2X for others.
1: any plans about enabling acceleration on x86-64 soon?
2: using two ulong obviously gives the best perf on x86-64, I have no
32-bit setup handy
to test if it would give significantly better perf using four uint
instead on them (at the expense of 64-bit platforms though if it
-------------- next part --------------
A non-text attachment was scrubbed...
Size: 10824 bytes
Desc: not available
Url : http://lists.ximian.com/pipermail/mono-devel-list/attachments/20081110/d63b4987/attachment-0001.bin
More information about the Mono-devel-list