mirror of
https://github.com/codex-storage/constantine.git
synced 2025-01-13 12:34:31 +00:00
e5612f5705
* unoptimized msm * MSM: reorder loops * add a signed windowed recoding technique * improve wNAF table access * use batchAffine * revamp EC tests * MSM signed digit support * refactor MSM: recode signed ahead of time * missing test vector * refactor allocs and Alloca sideeffect * add an endomorphism threshold * Add Jacobian extended coordinates * refactor recodings, prepare for parallelizable on-the-fly signed recoding * recoding changes, introduce proper NAF for pairings * more pairings refactoring, introduce miller accumulator for EVM * some optim to the addchain miller loop * start optimizing multi-pairing * finish multi-miller loop refactoring * minor tuning * MSM: signed encoding suitable for parallelism (no precompute) * cleanup signed window encoding * add prefetching * add metering * properly init result to infinity * comment on prefetching * introduce vartime inversion for batch additions * fix JacExt infinity conversion * add batchAffine for MSM, though slower than JacExtended at the moment * add a batch affine scheduler for MSM * Add Multi-Scalar-Multiplication endomorphism acceleration * some tuning * signed integer fixes + 32-bit + tuning * Some more tuning * common msm bench + don't use affine for c < 9 * nit
63 lines
3.4 KiB
Nim
63 lines
3.4 KiB
Nim
# Constantine
|
|
# Copyright (c) 2018-2019 Status Research & Development GmbH
|
|
# Copyright (c) 2020-Present Mamy André-Ratsimbazafy
|
|
# Licensed and distributed under either of
|
|
# * MIT license (license terms in the root directory or at http://opensource.org/licenses/MIT).
|
|
# * Apache v2 license (license terms in the root directory or at http://www.apache.org/licenses/LICENSE-2.0).
|
|
# at your option. This file may not be copied, modified, or distributed except according to those terms.
|
|
|
|
import
|
|
std/[strformat, strutils],
|
|
../benchmarks/platforms,
|
|
tracer
|
|
|
|
# Reporting benchmark result
|
|
# -------------------------------------------------------
|
|
|
|
proc reportCli*(metrics: seq[Metadata], flags: string) =
|
|
|
|
let name = when SupportsCPUName: cpuName() else: "(name auto-detection not implemented for this CPU family)"
|
|
echo "\nCPU: ", name
|
|
|
|
when SupportsGetTicks:
|
|
# https://blog.trailofbits.com/2019/10/03/tsc-frequency-for-all-better-profiling-and-benchmarking/
|
|
# https://www.agner.org/optimize/blog/read.php?i=838
|
|
echo "The CPU Cycle Count is indicative only. It cannot be used to compare across systems, works at your CPU nominal frequency and is sensitive to overclocking, throttling and frequency scaling (powersaving and Turbo Boost)."
|
|
|
|
const lineSep = &"""|{'-'.repeat(150)}|{'-'.repeat(14)}|{'-'.repeat(20)}|{'-'.repeat(15)}|{'-'.repeat(17)}|{'-'.repeat(26)}|{'-'.repeat(26)}|"""
|
|
echo "\n"
|
|
echo lineSep
|
|
echo &"""|{"Procedures":^150}|{"# of Calls":^14}|{"Throughput (ops/s)":^20}|{"Time (µs)":^15}|{"Avg Time (µs)":^17}|{"CPU cycles (in billions)":^26}|{"Avg cycles (in billions)":^26}|"""
|
|
echo &"""|{flags:^150}|{' '.repeat(14)}|{' '.repeat(20)}|{' '.repeat(15)}|{' '.repeat(17)}|{"indicative only":^26}|{"indicative only":^26}|"""
|
|
echo lineSep
|
|
for m in metrics:
|
|
if m.numCalls == 0:
|
|
continue
|
|
# TODO: running variance / standard deviation but the Welford method is quite costly.
|
|
# https://nim-lang.org/docs/stats.html / https://en.wikipedia.org/wiki/Algorithms_for_calculating_variance#Welford's_online_algorithm
|
|
let cumulTimeUs = m.cumulatedTimeNs.float64 * 1e-3
|
|
let avgTimeUs = cumulTimeUs / m.numCalls.float64
|
|
let throughput = 1e6 / avgTimeUs
|
|
let cumulCyclesBillions = m.cumulatedCycles.float64 * 1e-9
|
|
let avgCyclesBillions = cumulCyclesBillions / m.numCalls.float64
|
|
echo &"""|{m.procName:<150}|{m.numCalls:>14}|{throughput:>20.3f}|{cumulTimeUs:>15.3f}|{avgTimeUs:>17.3f}|"""
|
|
echo lineSep
|
|
|
|
else:
|
|
const lineSep = &"""|{'-'.repeat(50)}|{'-'.repeat(14)}|{'-'.repeat(20)}|{'-'.repeat(15)}|{'-'.repeat(17)}|"""
|
|
echo "\n"
|
|
echo lineSep
|
|
echo &"""|{"Procedures":^150}|{"# of Calls":^14}|{"Throughput (ops/s)":^20}|{"Time (µs)":^15}|{"Avg Time (µs)":^17}|"""
|
|
echo &"""|{flags:^150}|{' '.repeat(14)}|{' '.repeat(20)}|{' '.repeat(15)}|{' '.repeat(17)}|"""
|
|
echo lineSep
|
|
for m in metrics:
|
|
if m.numCalls == 0:
|
|
continue
|
|
# TODO: running variance / standard deviation but the Welford method is quite costly.
|
|
# https://nim-lang.org/docs/stats.html / https://en.wikipedia.org/wiki/Algorithms_for_calculating_variance#Welford's_online_algorithm
|
|
let cumulTimeUs = m.cumulatedTimeNs.float64 * 1e-3
|
|
let avgTimeUs = cumulTimeUs / m.numCalls.float64
|
|
let throughput = 1e6 / avgTimeUs
|
|
echo &"""|{m.procName:<150}|{m.numCalls:>14}|{throughput:>20.3f}|{cumulTimeUs:>15.3f}|{avgTimeUs:>17.3f}|"""
|
|
echo lineSep
|