DOCUMENT:Q167750 13-FEB-2002 [visualc] TITLE :BUG: The pow() Function Doesn't Set Underflow Condition PRODUCT :Microsoft C Compiler PROD/VER::5.0,6.0 OPER/SYS: KEYWORDS:kbtool kbVC500bug kbVC600bugkbbuglist ====================================================================== ------------------------------------------------------------------------------- The information in this article applies to: - The C Run-Time (CRT), used with: - Microsoft Visual C++, 32-bit Enterprise Edition, versions 5.0, 6.0 - Microsoft Visual C++, 32-bit Professional Edition, versions 5.0, 6.0 - Microsoft Visual C++, 32-bit Learning Edition, version 6.0 - Microsoft Visual C++.NET (2002) ------------------------------------------------------------------------------- SYMPTOMS ======== The pow() function does not set the errno variable to indicate that an underflow condition has occurred. RESOLUTION ========== A potential solution is to check the base of the number that is being raised to a power and check the result of the pow() function. By customizing the _matherr() function, the underflow condition can be trapped and you can display an error message. The program below demonstrates this workaround. #include #include #include #include extern int errno; /* The _matherr routine can be customized to handle the specific underflow problem that is encountered. See online help for more information on _matherr(). */ int _matherr( struct _exception *except ) { /* Handle errors caused by pow() function.*/ if(strcmp(except->name,"pow")==0) { switch(except->type) { case _DOMAIN: { printf("Domain Error: Argument not in domain.\n"); return 1; break; } case _SING: { printf("Singularity Error: Argument singularity " "error.\n"); return 1; break; } case _OVERFLOW: { printf("Overflow Error: Overflow range error.\n"); return 1; break; } case _PLOSS: { printf("Precision Error: Partial loss of " "significance.\n"); return 1; break; } case _TLOSS: { printf("Precision Error: Total loss of " "significance.\n"); return 1; break; } case _UNDERFLOW: { printf("Underflow Error: The result is too small to " "be represented.\n"); return 1; break; } default: { printf("Unknown Error Occurred While Performing Math " "Operation.\n"); return 1; break; } } } else return 0; /* non-error condition */ } void main(void) { double x = -1.0e+307; double y = -1.0e+307; double z; z = pow(x,y); /* Check for error conditions. _matherr() is automatically called should there be a math related error. */ /*Note that errno is not set properly & _matherr() is called.*/ if (errno != 0) printf("ERROR! Errno: %i\n",errno); else printf("pow( %e, %e ) = %f\n", x, y, z); } STATUS ====== Microsoft has confirmed this to be a bug in the Microsoft products listed at the beginning of this article. MORE INFORMATION ================ The help topic for pow() states: No error message is printed on overflow or underflow. However, an overflow does set errno to the value of 34 indicating "Result too large". When an underflow occurs, the returned value is 0. Additional query words: ====================================================================== Keywords : kbtool kbVC500bug kbVC600bug kbbuglist Technology : kbVCsearch kbAudDeveloper kbCRT Version : :5.0,6.0 Issue type : kbbug ============================================================================= THE INFORMATION PROVIDED IN THE MICROSOFT KNOWLEDGE BASE IS PROVIDED "AS IS" WITHOUT WARRANTY OF ANY KIND. MICROSOFT DISCLAIMS ALL WARRANTIES, EITHER EXPRESS OR IMPLIED, INCLUDING THE WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. IN NO EVENT SHALL MICROSOFT CORPORATION OR ITS SUPPLIERS BE LIABLE FOR ANY DAMAGES WHATSOEVER INCLUDING DIRECT, INDIRECT, INCIDENTAL, CONSEQUENTIAL, LOSS OF BUSINESS PROFITS OR SPECIAL DAMAGES, EVEN IF MICROSOFT CORPORATION OR ITS SUPPLIERS HAVE BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES. SOME STATES DO NOT ALLOW THE EXCLUSION OR LIMITATION OF LIABILITY FOR CONSEQUENTIAL OR INCIDENTAL DAMAGES SO THE FOREGOING LIMITATION MAY NOT APPLY. Copyright Microsoft Corporation 2002.