freebsd-dev/share/man/man3/Q_SIGNED.3
Edward Tomasz Napierala 5a38af1fb8 Introduce <sys/qmath.h>, a fixed-point math library from Netflix.
This makes it possible to perform mathematical operations
on
fractional values without using floating point. It operates on Q
numbers, which are integer-sized, opaque structures initialized
to hold a chosen number of integer and fractional
bits.


For a general description of the Q number system, see the "Fixed Point
Representation & Fractional Math" whitepaper[1]; for the actual
API see the qmath(3) man page.

This is one of dependencies for the upcoming stats(3) framework[2]
that will be applied to the TCP stack in a later commit.

1. https://www.superkits.net/whitepapers/Fixed%20Point%20Representation%20&%20Fractional%20Math.pdf
2. https://reviews.freebsd.org/D20477

Reviewed by:	bcr (man pages, earlier version), sef (earlier version)
Discussed with:	cem, dteske, imp, lstewart
Sponsored By:	Klara Inc, Netflix
Obtained from:	Netflix
Differential Revision:	https://reviews.freebsd.org/D20116
2019-08-27 11:46:22 +00:00

211 lines
4.3 KiB
Groff

.\"
.\" Copyright (c) 2018 Netflix, Inc.
.\" All rights reserved.
.\"
.\" Redistribution and use in source and binary forms, with or without
.\" modification, are permitted provided that the following conditions
.\" are met:
.\" 1. Redistributions of source code must retain the above copyright
.\" notice, this list of conditions, and the following disclaimer,
.\" without modification, immediately at the beginning of the file.
.\" 2. The name of the author may not be used to endorse or promote products
.\" derived from this software without specific prior written permission.
.\"
.\" THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
.\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
.\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
.\" ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE FOR
.\" ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
.\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
.\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
.\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
.\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
.\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
.\" SUCH DAMAGE.
.\"
.\" $FreeBSD$
.\"
.Dd July 8, 2018
.Dt Q_SIGNED 3
.Os
.Sh NAME
.Nm Q_SIGNED ,
.Nm Q_LTZ ,
.Nm Q_PRECEQ ,
.Nm Q_QLTQ ,
.Nm Q_QLEQ ,
.Nm Q_QGTQ ,
.Nm Q_QGEQ ,
.Nm Q_QEQ ,
.Nm Q_QNEQ ,
.Nm Q_OFLOW ,
.Nm Q_RELPREC
.Nd fixed-point math comparison and logic functions
.Sh SYNOPSIS
.In sys/qmath.h
.Ft bool
.Fn Q_SIGNED "NTYPE n"
.Ft bool
.Fn Q_LTZ "NTYPE n"
.Ft bool
.Fn Q_PRECEQ "QTYPE a" "QTYPE b"
.Ft bool
.Fn Q_QLTQ "QTYPE a" "QTYPE b"
.Ft bool
.Fn Q_QLEQ "QTYPE a" "QTYPE b"
.Ft bool
.Fn Q_QGTQ "QTYPE a" "QTYPE b"
.Ft bool
.Fn Q_QGEQ "QTYPE a" "QTYPE b"
.Ft bool
.Fn Q_QEQ "QTYPE a" "QTYPE b"
.Ft bool
.Fn Q_QNEQ "QTYPE a" "QTYPE b"
.Ft bool
.Fn Q_OFLOW "QTYPE q" "ITYPE iv"
.Ft int
.Fn Q_RELPREC "QTYPE a" "QTYPE b"
.Sh DESCRIPTION
.Fn Q_SIGNED
returns
.Ft true
if the numeric data type passed in as
.Fa n
is signed, or
.Ft false
otherwise.
.Pp
.Fn Q_LTZ
returns
.Ft true
if the numeric value
passed in as
.Fa n
is negative
.Pq requires types which use the MSB as the sign bit ,
or
.Ft false
otherwise.
.Pp
.Fn Q_PRECEQ
returns
.Ft true
if the number of
.Fa a
and
.Fa b
fractional bits is the same,
.Ft false
otherwise.
.Pp
The
.Fn Q_QLTQ ,
.Fn Q_QLEQ ,
.Fn Q_QGTQ ,
.Fn Q_QGEQ ,
.Fn Q_QEQ
and
.Fn Q_QNEQ
functions compare two Q numbers, returning
.Ft true
if
.Fa a
is less than, less than or equal to, greater than, greater than or equal to,
equal to, or not equal to
.Fa b
respectively, or
.Ft false
otherwise.
The integral and fractional values are used to perform the comparison, without
explicit concern for the underlying number of integer versus fractional bits.
.Pp
.Fn Q_OFLOW
returns
.Ft true
if integer value
.Fa iv
cannot be stored in
.Fa q
without truncation, or false otherwise.
.Pp
.Fn Q_RELPREC
returns the relative precision of
.Fa a
versus
.Fa b .
In terms of
.Em Qm.n
notation, this function returns the difference between the
.Em n
values of
.Fa a
and
.Fa b .
For example, a return value of +4 means that
.Fa a
has an additional 4 bits of fractional precision compared to
.Fa b .
.Pp
All of those functions operate on
the following data types:
.Vt s8q_t ,
.Vt u8q_t ,
.Vt s16q_t ,
.Vt u16q_t ,
.Vt s32q_t ,
.Vt u32q_t ,
.Vt s64q_t ,
and
.Vt u64q_t ,
which are referred to generically as
.Fa QTYPE .
The
.Fa ITYPE
refers to the
.Xr stdint 7
integer types.
.Fa NTYPE
is used to refer to any numeric type and is therefore a superset of
.Fa QTYPE
and
.Fa ITYPE .
.Pp
For more details, see
.Xr qmath 3 .
.Sh RETURN VALUES
The
.Fn Q_SIGNED ,
.Fn Q_LTZ ,
.Fn Q_PRECEQ ,
.Fn Q_QLTQ ,
.Fn Q_QLEQ ,
.Fn Q_QGTQ ,
.Fn Q_QGEQ ,
.Fn Q_QEQ ,
.Fn Q_QNEQ
and
.Fn Q_OFLOW
functions return expressions that evaluate to boolean
.Vt true
or
.Vt false .
.Pp
.Fn Q_RELPREC
returns the relative precision difference as a signed integer.
.Sh SEE ALSO
.Xr errno 2 ,
.Xr qmath 3 ,
.Xr stdint 7
.Sh HISTORY
The
.Xr qmath 3
functions first appeared in
.Fx 13.0 .
.Sh AUTHORS
.An -nosplit
The
.Xr qmath 3
functions and this manual page were written by
.An Lawrence Stewart Aq Mt lstewart@FreeBSD.org
and sponsored by Netflix, Inc.