From: lkcl Date: Sun, 16 Apr 2023 10:25:51 +0000 (+0100) Subject: (no commit message) X-Git-Tag: opf_rfc_ls009_v1~31 X-Git-Url: https://git.libre-soc.org/?a=commitdiff_plain;h=545739f0d9ab7b716d5a8f687c4c1c96561cb9dc;p=libreriscv.git --- diff --git a/openpower/sv/remap/appendix.mdwn b/openpower/sv/remap/appendix.mdwn index 2aac9482b..aa16bb2cd 100644 --- a/openpower/sv/remap/appendix.mdwn +++ b/openpower/sv/remap/appendix.mdwn @@ -78,26 +78,25 @@ pipeline overlaps. Out-of-order / Superscalar micro-architectures with register-renaming will have an easier time dealing with this than DSP-style SIMD micro-architectures. -## REMAP FFT pseudocode +## REMAP FFT, DFT, NTT -The algorithm below shows how FFT REMAP works, and may be -executed as a python program: - -``` -[[!inline pages="openpower/sv/remap_fft_yield.py" quick="yes" raw="yes" ]] -``` - -The executable code above is designed to show how a hardware +The algorithm from a later section of this Appendix shows how FFT REMAP works, +and it may be executed as a standalone python3 program. +The executable code is designed to illustrate how a hardware implementation may generate Indices which are completely independent of the Execution of element-level operations, even for something as complex as a Triple-loop Tukey-Cooley Schedule. A comprehensive demo and test suite may be found [here](https://git.libre-soc.org/?p=openpower-isa.git;a=blob;f=src/openpower/decoder/isa/test_caller_svp64_fft.py;hb=HEAD) +including Complex Number FFT which deploys Vertical-First Mode +on top of the REMAP Schedules. -Other uses include more than DFT and NTT: the Schedules are not -restricted in any way and if the programmer can find any algorithm +Other uses include more than DFT and NTT: as abstracted RISC-paradigm +the Schedules are not +restricted in any way or tied to any particular instructtion. +If the programmer can find any algorithm which has identical triple nesting then the FFT Schedule may be -used. +used even there. # 4x4 Matrix to vec4 Multiply Example