'\" te .\" Copyright (c) 1996, Sun Microsystems, Inc. .\" The contents of this file are subject to the terms of the Common Development and Distribution License (the "License"). You may not use this file except in compliance with the License. .\" You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE or http://www.opensolaris.org/os/licensing. See the License for the specific language governing permissions and limitations under the License. .\" When distributing Covered Code, include this CDDL HEADER in each file and include the License file at usr/src/OPENSOLARIS.LICENSE. If applicable, add the following below this CDDL HEADER, with the fields enclosed by brackets "[]" replaced with your own identifying information: Portions Copyright [yyyy] [name of copyright owner] .TH SIGFPE 3C "May 4, 2004" .SH NAME sigfpe \- signal handling for specific SIGFPE codes .SH SYNOPSIS .LP .nf #include #include \fBsigfpe_handler_type\fR \fBsigfpe\fR(\fBsigfpe_code_type\fR \fIcode\fR, \fBsigfpe_handler_type\fR \fIhdl\fR); .fi .SH DESCRIPTION .LP The \fBsigfpe()\fR function allows signal handling to be specified for particular \fBSIGFPE\fR codes. A call to \fBsigfpe()\fR defines a new handler \fIhdl\fR for a particular \fBSIGFPE\fR \fIcode\fR and returns the old handler as the value of the function \fBsigfpe()\fR. Normally handlers are specified as pointers to functions; the special cases \fBSIGFPE_IGNORE\fR, \fBSIGFPE_ABORT\fR, and \fBSIGFPE_DEFAULT\fR allow ignoring, dumping core using \fBabort\fR(3C), or default handling respectively. Default handling is to dump core using \fBabort\fR(3C). .sp .LP The \fIcode\fR argument is usually one of the five \fBIEEE\|754-related\fR \fBSIGFPE\fR codes: .sp .in +2 .nf FPE_FLTRES fp_inexact \(mi floating-point inexact result FPE_FLTDIV fp_division \(mi floating-point division by zero FPE_FLTUND fp_underflow \(mi floating-point underflow FPE_FLTOVF fp_overflow \(mi floating-point overflow FPE_FLTINV fp_invalid \(mi floating-point invalid operation .fi .in -2 .LP And additionally on the x86 architecture: .sp .in +2 .nf FPE_FLTDEN fp_denormalized \(mi floating-point denormalized result .fi .in -2 .sp .LP Three steps are required to intercept an \fBIEEE\|754-related\fR \fBSIGFPE\fR code with \fBsigfpe()\fR: .RS +4 .TP 1. Set up a handler with \fBsigfpe()\fR. .RE .RS +4 .TP 2. Enable the relevant \fBIEEE\|754\fR trapping capability in the hardware, perhaps by using assembly-language instructions. .RE .RS +4 .TP 3. Perform a floating-point operation that generates the intended \fBIEEE\|754\fR exception. .RE .sp .LP The \fBsigfpe()\fR function never changes floating-point hardware mode bits affecting \fBIEEE\|754\fR trapping. No \fBIEEE\|754-related\fR \fBSIGFPE\fR signals will be generated unless those hardware mode bits are enabled. .sp .LP \fBSIGFPE\fR signals can be handled using \fBsigfpe()\fR, \fBsigaction\fR(2) or \fBsignal\fR(3C). In a particular program, to avoid confusion, use only one of these interfaces to handle \fBSIGFPE\fR signals. .SH EXAMPLES .LP \fBExample 1 \fRExample Of A User-Specified Signal Handler .sp .LP A user-specified signal handler might look like this: .sp .in +2 .nf #include #include #include /* * The sample_handler prints out a message then commits suicide. */ void sample_handler(int sig, siginfo_t *sip, ucontext_t *uap) { char *label; switch (sip\(mi>si_code) { case FPE_FLTINV: label = "invalid operand"; break; case FPE_FLTRES: label = "inexact"; break; case FPE_FLTDIV: label = "division-by-zero"; break; case FPE_FLTUND: label = "underflow"; break; case FPE_FLTOVF: label = "overflow"; break; default: label = "???"; break; } fprintf(stderr, "FP exception %s (0x%x) occurred at address %p.\en", label, sip\(mi>si_code, (void *) sip\(mi>si_addr); abort(); } .fi .in -2 .sp .LP and it might be set up like this: .sp .in +2 .nf #include #include #include extern void sample_handler(int, siginfo_t *, ucontext_t *); main(void) { sigfpe_handler_type hdl, old_handler1, old_handler2; /* * save current fp_overflow and fp_invalid handlers; set the new * fp_overflow handler to sample_handler(\|) and set the new * fp_invalid handler to SIGFPE_ABORT (abort on invalid) */ hdl = (sigfpe_handler_type) sample_handler; old_handler1 = sigfpe(FPE_FLTOVF, hdl); old_handler2 = sigfpe(FPE_FLTINV, SIGFPE_ABORT); .\|.\|. /* * restore old fp_overflow and fp_invalid handlers */ sigfpe(FPE_FLTOVF, old_handler1); sigfpe(FPE_FLTINV, old_handler2); } .fi .in -2 .SH FILES .ne 2 .na \fB\fB/usr/include/floatingpoint.h\fR\fR .ad .sp .6 .RS 4n .RE .sp .ne 2 .na \fB\fB/usr/include/siginfo.h\fR\fR .ad .sp .6 .RS 4n .RE .SH ATTRIBUTES .LP See \fBattributes\fR(5) for descriptions of the following attributes: .sp .sp .TS box; c | c l | l . ATTRIBUTE TYPE ATTRIBUTE VALUE _ MT-Level Safe .TE .SH SEE ALSO .LP \fBsigaction\fR(2), \fBabort\fR(3C), \fBsignal\fR(3C), \fBattributes\fR(5), \fBfloatingpoint.h\fR(3HEAD) .SH DIAGNOSTICS .LP The \fBsigfpe()\fR function returns (void(*)())-1 if \fIcode\fR is not zero or a defined \fBSIGFPE\fR code.