/home/abui...lamchf77.f(3)

💅 NAME

/home/abuild/rpmbuild/BUILD/lapack-static-3.12.0-build/lapack-3.12.0/INSTALL/slamchf77.f

Okay so bestie, that whole entire path is the name — no cap, that's the whole vibe. This is the manual page for slamchf77.f from LAPACK, living in the Library Functions Manual, filed under /home/abui...lamchf77.f(3). It's giving "I contain six whole routines that sniff out your machine's floating-point parameters" and honestly? Iconic. ✨

🌸 SYNOPSIS — the cast list 🌸

🎀 Functions/Subroutines 🎀

real function SLAMCH (cmach) — the main character, does the actual parameter-determining 💖
subroutine SLAMC1 (beta, t, rnd, ieee1) — supporting role, finds BETA, T, RND, IEEE1
subroutine SLAMC2 (beta, t, rnd, eps, emin, rmin, emax, rmax) — finds a whole pile of parameters
real function SLAMC3 (a, b) — tiny little helper that just… adds A and B, but dramatically
subroutine SLAMC4 (emin, start, base) — service routine for SLAMC2, does the EMIN math
subroutine SLAMC5 (beta, p, emin, ieee, emax, rmax) — out here hunting for RMAX without overflow

🦋 Function/Subroutine Documentation 🦋

🧚 SLAMC1

subroutine SLAMC1 (integer beta, integer t, logical rnd, logical ieee1)

Purpose 💖

SLAMC1 figures out the machine parameters that come out as BETA, T, RND and IEEE1. That's literally the whole job. Slay.

Parameters 🎀

BETA — the base of the machine.
T — the number of ( BETA ) digits in the mantissa.
RND — tells you whether proper rounding ( RND = .TRUE. ) or chopping ( RND = .FALSE. ) happens in addition. Fr fr though, heads up: this may not be a reliable guide to the way in which the machine performs its arithmetic. So don't come for me if it's off 🙈
IEEE1 — tells you whether rounding appears to be done in the IEEE 'round to nearest' style.

Author 👑

LAPACK is a software package provided by Univ. of Tennessee, Univ. of California Berkeley, Univ. of Colorado Denver and NAG Ltd..

Further Details 🔥

Lowkey a whole heritage moment — this routine is based on the routine ENVRON by Malcolm and incorporates suggestions by Gentleman and Marovich. See:

Malcolm M. A. (1972) Algorithms to reveal properties of
   floating-point arithmetic. Comms. of the ACM, 15, 949-951.

Gentleman W. M. and Marovich S. B. (1974) More on algorithms
   that reveal properties of floating point arithmetic units.
   Comms. of the ACM, 17, 276-277.

Definition at line 206 of file slamchf77.f.

🧚 SLAMC2

subroutine SLAMC2 (integer beta, integer t, logical rnd, real eps, integer emin,
     real rmin, integer emax, real rmax)

Purpose 💖

SLAMC2 determines the machine parameters specified in its argument list. All of them. Every single one. 🫶

Author 👑

LAPACK is a software package provided by Univ. of Tennessee, Univ. of California Berkeley, Univ. of Colorado Denver and NAG Ltd..

Parameters 🎀

BETA — the base of the machine.
T — the number of ( BETA ) digits in the mantissa.
RND — specifies whether proper rounding ( RND = .TRUE. ) or chopping ( RND = .FALSE. ) occurs in addition. Again, this may not be a reliable guide to the way in which the machine performs its arithmetic. Not me repeating the warning, it's literally in the source okay 😭
EPS — the smallest positive number such that

fl( 1.0 - EPS ) .LT. 1.0,

where fl denotes the computed value.

EMIN — the minimum exponent before (gradual) underflow occurs.
RMIN — the smallest normalized number for the machine, given by BASE**( EMIN - 1 ), where BASE is the floating point value of BETA.
EMAX — the maximum exponent before overflow occurs.
RMAX — the largest positive number for the machine, given by BASE**EMAX * ( 1 - EPS ), where BASE is the floating point value of BETA.

Further Details 🔥

The computation of EPS is based on a routine PARANOIA by W. Kahan of the University of California at Berkeley. Respect the lineage 💜

Definition at line 418 of file slamchf77.f.

🧚 SLAMC3

real function SLAMC3 (real a, real b)

Purpose 💖

SLAMC3 is intended to force A and B to be stored prior to doing the addition of A and B, for use in situations where optimizers might hold one of these in a register. It's giving "I'm here to stop the optimizer from being too clever." Main character energy for a two-line addition, honestly. 💀

Author 👑

LAPACK is a software package provided by Univ. of Tennessee, Univ. of California Berkeley, Univ. of Colorado Denver and NAG Ltd..

Parameters 🎀

A — (no description, it's just vibes and an empty line 🫠)
B — the values A and B.

Purpose (yes, again) 💖

And because the docs said "encore", here it is one more time: SLAMC3 is intended to force A and B to be stored prior to doing the addition of A and B, for use in situations where optimizers might hold one of these in a register. Not me copy-pasting — they did, bestie.

Parameters (also again) 🎀

A
B — the values A and B.

Definition at line 170 of file slamch.f. (note: slamch.f, not slamchf77.f — keeping it accurate 🎀)

🧚 SLAMC4

subroutine SLAMC4 (integer emin, real start, integer base)

Purpose 💖

SLAMC4 is a service routine for SLAMC2. Just out here doing the support work with zero credit — we see you, bestie. 🌸

Parameters 🎀

EMIN — the minimum exponent before (gradual) underflow, computed by setting A = START and dividing by BASE until the previous A can not be recovered.
START — the starting point for determining EMIN.
BASE — the base of the machine.

Definition at line 688 of file slamchf77.f.

🧚 SLAMC5

subroutine SLAMC5 (integer beta, integer p, integer emin, logical ieee,
     integer emax, real rmax)

Purpose 💖

SLAMC5 attempts to compute RMAX, the largest machine floating-point number, without overflow. It assumes that EMAX + abs(EMIN) sum approximately to a power of 2.

⚠️ Heads up bestie, it will fail on machines where this assumption does not hold — for example, the Cyber 205 ( EMIN = -28625, EMAX = 28718 ). It will also fail if the value supplied for EMIN is too large (i.e. too close to zero), probably with overflow. So like… know your machine, okay? 😭

Parameters 🎀

BETA — the base of floating-point arithmetic.
P — the number of base BETA digits in the mantissa of a floating-point value.
EMIN — the minimum exponent before (gradual) underflow.
IEEE — a logical flag specifying whether or not the arithmetic system is thought to comply with the IEEE standard.
EMAX — the largest exponent before overflow.
RMAX — the largest machine floating-point number.

Definition at line 796 of file slamchf77.f.

👑 SLAMCH — the actual headliner 👑

real function SLAMCH (character cmach)

SLAMCH SLAMCHF77 deprecated 🚨 — yes she's deprecated, we still love her, but note it fr fr.

Purpose 💖

SLAMCH determines single precision machine parameters. Simple as that, and yet so much power.

Parameters 🎀

CMACH is CHARACTER*1. Here's the whole menu, verbatim, because accuracy is the vibe:

Specifies the value to be returned by SLAMCH:
= 'E' or 'e',   SLAMCH := eps
= 'S' or 's ,   SLAMCH := sfmin
= 'B' or 'b',   SLAMCH := base
= 'P' or 'p',   SLAMCH := eps*base
= 'N' or 'n',   SLAMCH := t
= 'R' or 'r',   SLAMCH := rnd
= 'M' or 'm',   SLAMCH := emin
= 'U' or 'u',   SLAMCH := rmin
= 'L' or 'l',   SLAMCH := emax
= 'O' or 'o',   SLAMCH := rmax
where
eps   = relative machine precision
sfmin = safe minimum, such that 1/sfmin does not overflow
base  = base of the machine
prec  = eps*base
t     = number of (base) digits in the mantissa
rnd   = 1.0 when rounding occurs in addition, 0.0 otherwise
emin  = minimum exponent before (gradual) underflow
rmin  = underflow threshold - base**(emin-1)
emax  = largest exponent before overflow
rmax  = overflow threshold  - (base**emax)*(1-eps)

(Yes, 's , is written like that in the original. Leaving it exactly as found, don't @ me 💅)

Author 👑

Univ. of Tennessee
Univ. of California Berkeley
Univ. of Colorado Denver
NAG Ltd.

Purpose (the sequel) 💖

SLAMCH determines single precision machine parameters. Same sentence, second time. We love consistency. 🌈

Parameters (the sequel) 🎀

CMACH — specifies the value to be returned by SLAMCH:

= 'E' or 'e',   SLAMCH := eps
= 'S' or 's ,   SLAMCH := sfmin
= 'B' or 'b',   SLAMCH := base
= 'P' or 'p',   SLAMCH := eps*base
= 'N' or 'n',   SLAMCH := t
= 'R' or 'r',   SLAMCH := rnd
= 'M' or 'm',   SLAMCH := emin
= 'U' or 'u',   SLAMCH := rmin
= 'L' or 'l',   SLAMCH := emax
= 'O' or 'o',   SLAMCH := rmax
where
eps   = relative machine precision
sfmin = safe minimum, such that 1/sfmin does not overflow
base  = base of the machine
prec  = eps*base
t     = number of (base) digits in the mantissa
rnd   = 1.0 when rounding occurs in addition, 0.0 otherwise
emin  = minimum exponent before (gradual) underflow
rmin  = underflow threshold - base**(emin-1)
emax  = largest exponent before overflow
rmax  = overflow threshold  - (base**emax)*(1-eps)

Author (the sequel) 👑

Univ. of Tennessee
Univ. of California Berkeley
Univ. of Colorado Denver
NAG Ltd.

Definition at line 67 of file slamch.f.

🌟 Author 🌟

Generated automatically by Doxygen for LAPACK from the source code. (Not written by a human hand, just Doxygen doing Doxygen things 🫶)

And that's the whole page, bestie! 💜 LAPACK, Version 3.12.0, brought to you by /home/abui...lamchf77.f(3) — go forth and know your machine's floating-point parameters ✨🎀🦋

📦 where this page comes from

opensuse-leap 16.0: lapack-man 3.12.0-160000.2.2 · BSD-3-Clause
opensuse-leap 16.0: lapack-man 3.12.0-160000.2.2 · BSD-3-Clause
opensuse-leap 16.0: lapack-man 3.12.0-160000.2.2 · BSD-3-Clause
opensuse-leap 16.0: lapack-man 3.12.0-160000.2.2 · BSD-3-Clause
💖 girlified by deepseek-ai/DeepSeek-V4.1-Flash-1017 on 2026-11-04: a language model's rewrite of the original, under its terms