News for hackers. Stuff that matters.
Re:Ara: A 1 GHz+ Scalable and Energy-Efficient RISC-V Vector... by vardump on Friday June 07, 2019 @08:00PM (#20129651)
> The whole point of Risc is to avoid instruction complexity for rarely used instructions

Yeah, that was true 20-30 years ago.

> Popcount and family are most likely microcoded to expand to a loop anyway, so why add this complexity to silicon?

No, you can just make a network of small adders. You could easily build a CPU that has throughput of 1 popcount per cycle (or even better, if SIMD and/or multiple execution ports supported this ALU operation). Silicon area is not a problem anyways anymore. Power and heat on the other hand...

Re:Ara: A 1 GHz+ Scalable and Energy-Efficient RISC-V Vector... by hnuser123456 on Friday June 07, 2019 @10:32PM (#20130267)
>Silicon area is not a problem anyways anymore. Power and heat on the other hand...

Yes, so then why are ARM chips in mobile devices approaching x86 performance with an order of magnitude less power? C̶o̶u̶l̶d̶ ̶i̶t̶ ̶b̶e̶ ̶t̶h̶a̶t̶ ̶h̶a̶v̶i̶n̶g̶ ̶a̶ ̶l̶a̶r̶g̶e̶ ̶c̶h̶u̶n̶k̶ ̶o̶f̶ ̶r̶a̶r̶e̶l̶y̶-̶u̶s̶e̶d̶ ̶s̶i̶l̶i̶c̶o̶n̶ ̶e̶n̶d̶s̶ ̶u̶p̶ ̶c̶o̶n̶s̶u̶m̶i̶n̶g̶ ̶s̶o̶m̶e̶ ̶s̶i̶g̶n̶i̶f̶i̶c̶a̶n̶t̶ ̶a̶m̶o̶u̶n̶t̶ ̶o̶f̶ ̶p̶o̶w̶e̶r̶,̶ ̶b̶e̶c̶o̶m̶i̶n̶g̶ ̶d̶o̶m̶i̶n̶a̶n̶t̶ ̶a̶t̶ ̶e̶x̶c̶e̶s̶s̶i̶v̶e̶l̶y̶ ̶l̶a̶r̶g̶e̶ ̶s̶c̶a̶l̶e̶s̶?̶

Re:Ara: A 1 GHz+ Scalable and Energy-Efficient RISC-V Vector... by rayiner on Friday June 07, 2019 @11:07PM (#20130383)
That’s not the reason, especially not in this day and age of clock gating. It’s because those ARM designs are optimized for low power, at the expense of not having headroom to take advantage of higher power envelopes if they are available.
Re:Ara: A 1 GHz+ Scalable and Energy-Efficient RISC-V Vector... by ryacko on Friday June 07, 2019 @11:18PM (#20130411)
Die area does impact latency, and by including popcount, you would be increasing gate count by a fraction of a full multiplier.
Re:Ara: A 1 GHz+ Scalable and Energy-Efficient RISC-V Vector... by atq2119 on Saturday June 08, 2019 @04:15AM (#20131257)
The number of gates required for popcount scales linearly with the input size. It's much, much smaller than a multiplier.
Re:Ara: A 1 GHz+ Scalable and Energy-Efficient RISC-V Vector... by ryacko on Saturday June 08, 2019 @12:32PM (#20133293)
No, it would scale logarithmically if you want to minimize latency.
Re:Ara: A 1 GHz+ Scalable and Energy-Efficient RISC-V Vector... by atq2119 on Saturday June 08, 2019 @02:40PM (#20134201)
It cannot scale logarithmically because you need at least a linear number of gates to represent the inputs.

It doesn't have to scale superlinearly either, because the natural circuit using a linear number of adders is already of a logarithmic depth, which is best possible asymptotically.

Re:Ara: A 1 GHz+ Scalable and Energy-Efficient RISC-V Vector... by ncmncm on Friday June 21, 2019 @01:14AM (#20240043)
The gate depth scales logarithmically, and that determines how many cycles it takes, or (worse) how fast your cycles can be, if you want the answer in one cycle.
Re:Ara: A 1 GHz+ Scalable and Energy-Efficient RISC-V Vector... by ryacko on Saturday June 08, 2019 @07:10PM (#20135810)
A lot of complexity could be eschewed if processors weren't faster than 100 Mhz.