-rw-r--r-- 1790 lib25519-20240928/doc/man/lib25519-speed.1 raw
.\" Automatically generated by Pandoc 2.17.1.1
.\"
.\" Define V font for inline verbatim, using C font in formats
.\" that render this, and otherwise B font.
.ie "\f[CB]x\f[]"x" \{\
. ftr V B
. ftr VI BI
. ftr VB B
. ftr VBI BI
.\}
.el \{\
. ftr V CR
. ftr VI CI
. ftr VB CB
. ftr VBI CBI
.\}
.TH "lib25519-speed" "1" "" "" ""
.hy
.SS NAME
.PP
lib25519-speed - run some lib25519 speed tests
.SS SYNOPSIS
.IP
.nf
\f[C]
lib25519-speed
\f[R]
.fi
.SS DESCRIPTION
.PP
\f[V]lib25519-speed\f[R] runs some tests of the speed of the lib25519
implementation of X25519 and Ed25519.
.PP
The \f[V]lib25519-speed\f[R] output format is subject to change.
Examples of output lines with the current format, from a
\f[V]lib25519-speed\f[R] run on an Intel Haswell CPU core, include the
following:
.IP
.nf
\f[C]
nG_merged25519 selected implementation amd64-mxaa-opt compiler clang -Wall -fPIC -fwrapv -Qunused-arguments -O2 -mmmx -msse -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -mavx -mbmi -mbmi2 -mavx2 -mtune=haswell; Ubuntu clang version 14.0.0-1ubuntu1.1; Target: x86_64-pc-linux-gnu; Thread model: posix; InstalledDir: /usr/bin
nG_merged25519 selected 32 43276 +17556+272+38-60-200-13-109-24+139+118+3-24+39+37-86-221
\f[R]
.fi
.PP
These lines say that the automatically selected implementation for the
\f[V]nG_merged25519\f[R] subroutine was the \f[V]amd64-mxaa-opt\f[R]
implementation compiled with the \f[V]clang\f[R] options shown, taking
an interquartile mean (IQM) of 43276 CPU cycles.
The numbers \f[V]43276 +17556+272+...\f[R] show the timings for a series
of consecutive operations: the IQM was 43276 cycles, the first run took
17556 cycles more than the IQM, the second run took 272 cycles more than
the IQM, etc.
.SS SEE ALSO
.PP
\f[B]x25519\f[R](1), \f[B]ed25519\f[R](1), \f[B]lib25519\f[R](3)