Return to trpfpe.3 CVS log

Up to [CSRG BSD Unix] / 43BSDTahoe / man / man3f

Annotation of 43BSDTahoe/man/man3f/trpfpe.3, revision 1.1.1.1

1.1       root        1: .\" Copyright (c) 1983 Regents of the University of California.
                      2: .\" All rights reserved.  The Berkeley software License Agreement
                      3: .\" specifies the terms and conditions for redistribution.
                      4: .\"
                      5: .\"    @(#)trpfpe.3f   6.1 (Berkeley) 5/15/85
                      6: .\"
                      7: .TH TRPFPE 3F "May 15, 1985"
                      8: .UC 5
                      9: .SH NAME
                     10: trpfpe, fpecnt \- trap and repair floating point faults
                     11: .SH SYNOPSIS
                     12: .B subroutine trpfpe (numesg, rtnval)
                     13: .br
                     14: .B double precision rtnval
                     15: .sp 1
                     16: .B integer function fpecnt ()
                     17: .sp 1
                     18: .B common /fpeflt/ fperr
                     19: .br
                     20: .B logical fperr
                     21: .SH DESCRIPTION
                     22: \fBNOTE: This routine applies only to Vax computers.\fR
                     23: It is a null routine on the PDP11.
                     24: .PP
                     25: .I Trpfpe
                     26: sets up a signal handler to trap arithmetic exceptions.
                     27: If the exception is due to a floating point arithmetic fault,
                     28: the result of the operation is replaced with the
                     29: .I rtnval
                     30: specified.
                     31: .I Rtnval
                     32: must be a double precision value. For example, ``0d0'' or ``dflmax()''.
                     33: .PP
                     34: The first
                     35: .I numesg
                     36: occurrences of a floating point arithmetic error will cause a message
                     37: to be written to the standard error file.
                     38: Any exception that can't be repaired will result in the default
                     39: action, typically an abort with core image.
                     40: .PP
                     41: .I Fpecnt
                     42: returns the number of faults since the last call to
                     43: .IR trpfpe .
                     44: .PP
                     45: The logical value in the common block labelled
                     46: .I fpeflt
                     47: will be set to
                     48: .B .true.
                     49: each time a fault occurs.
                     50: .SH FILES
                     51: .ie \nM /usr/ucb/lib/libF77.a
                     52: .el /usr/lib/libF77.a
                     53: .SH SEE ALSO
                     54: signal(3F), range(3F)
                     55: .SH BUGS
                     56: This routine works only for
                     57: .IR faults ,
                     58: not
                     59: .IR traps .
                     60: This is primarily due to the Vax architecture.
                     61: .PP
                     62: If the operation involves changing the stack pointer, it can't be repaired.
                     63: This seldom should be a problem with the f77 compiler, but such an operation
                     64: might be produced by the optimizer.
                     65: .PP
                     66: The POLY and EMOD opcodes are not dealt with.

This archive runs on limited infrastructure. Preserving old code on modern bandwidth. Automated agents are requested to crawl responsibly.