bug-gsl
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Bug-gsl] Re: more info on vector check failure on solaris


From: Britton Kerin
Subject: [Bug-gsl] Re: more info on vector check failure on solaris
Date: Mon, 12 Jan 2004 11:30:49 -0900 (AKST)

That was dumb of me.  Here is the fixed code, as you advised (the make 
check log is attached):

       if (fabs(r - z) > 2 * GSL_FLT_EPSILON * fabs(z)) {
         status = 1;
         printf ("\nr: " OUT_FORMAT ", x: " OUT_FORMAT ", y: " OUT_FORMAT
                 ", z: " OUT_FORMAT "\nfabs(r - z): %.18e\n"
                 "2 * GSL_FLT_EPSILON * fabs(z): %.18e\n", r, x, y, z,
                 fabs(r - z), 2 * GSL_FLT_EPSILON * fabs(z));
       }

Britton

On Sat, 10 Jan 2004, Brian Gough wrote:

> Britton Kerin writes:
>  >                ", z: " OUT_FORMAT "\nfabs(r - z): %d\n"
>  >                "2 * GSL_FLT_EPSILON * fabs(z): %d\n", r, x, y, z,
>  >                fabs(r - z), 2 * GSL_FLT_EPSILON * fabs(z));
> 
> Hello,
> 
> %d is an integer format, you need to use a floating point format in
> the output, e.g. %.18e
> 
>  > r: 1, x: 3, y: 3, z: 1
>  > fabs(r - z): 1072693248              <<<< should use floating point format
>  > 2 * GSL_FLT_EPSILON * fabs(z): 0     <<<< should use floating point format
> 
> Thanks,
> 
> 

Attachment: make_check_log_fixed.txt
Description: Text document


reply via email to

[Prev in Thread] Current Thread [Next in Thread]