From 7cc67fc4ced068201c89238e9ef0589db6c54cc6 Mon Sep 17 00:00:00 2001 From: lkcl Date: Thu, 5 May 2022 14:40:30 +0100 Subject: [PATCH] --- openpower/sv/SimpleV_rationale.mdwn | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/openpower/sv/SimpleV_rationale.mdwn b/openpower/sv/SimpleV_rationale.mdwn index 04e6a411e..fa4ccd2d9 100644 --- a/openpower/sv/SimpleV_rationale.mdwn +++ b/openpower/sv/SimpleV_rationale.mdwn @@ -13,3 +13,14 @@ Khronos Group, with multi-man-century development committment. Therefore it begs the question, why on earth would anyone consider this task? + +Hints as to the answer emerge from an article +"[SIMD considered harmful](https://www.sigarch.org/simd-instructions-considered-harmful/)" +which illustrates a catastrophic rabbit-hole taken by Industry Giants +since the 90s (over 3 decades) whereby SIMD width, an Order(N^6) opcode +proliferation nightmare, with its mantra "make it easy for hardware engineers, +let software sort out the mess" literally overwhelming programmers. +Worse than that, specialists in charging clients Optimisation Services +are finding that AVX-512, to take an example, is anything but optimal: +overall performance actually *decreases* even as power consumption goes +up. -- 2.30.2