Re: Throughput of PXOR

"James Van Buskirk"  wrote in message
> While I was timing the popcnt code posted earlier I noticed that my
> CPU (Core 2 Duo E6700) was showing results consistent with issuing
> 2 SSE2 logical operations per clock cycle.  This seemed odd to me
> because Intel's manuals and Agner Fog both say I should be getting
> a throughput of 3 logical operations (PXOR, XORPS, XORPD) per cycle.
>
> Accordingly I made up a less complicated test:

In your loop the 4th xorps is dependent on the 1st xorps etc.
In throughput they should probably be all independent,
but I don't know how to do that.
Probably it's somehwere in Agner Fog's do***ents.