home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.os.linux
- Path: sparky!uunet!enterpoop.mit.edu!bloom-picayune.mit.edu!athena.mit.edu!alsaggaf
- From: alsaggaf@athena.mit.edu (M. Saggaf)
- Subject: Re: Kernel FP Math bug?
- In-Reply-To: jliddle@rs6000.cmp.ilstu.edu's message of Sat, 09 Jan 1993 22:48:46 GMT
- Message-ID: <ALSAGGAF.93Jan9183512@e40-008-4.mit.edu>
- Sender: news@athena.mit.edu (News system)
- Nntp-Posting-Host: e40-008-4.mit.edu
- Organization: Massachusetts Institute of Technology
- References: <1993Jan9.144510.9590@walter.cray.com>
- <1993Jan09.224846.12403@rs6000.cmp.ilstu.edu>
- Date: Sat, 9 Jan 1993 23:35:18 GMT
- Lines: 30
-
- >>>>> On Sat, 09 Jan 1993 22:48:46 GMT, jliddle@rs6000.cmp.ilstu.edu (Jean Liddle) said:
-
- JL> In article <1993Jan9.144510.9590@walter.cray.com> rsocia@alberti.cray.com (Rick Socia) writes:
- >I compiled the following program:
- >
- >main(int argc, char *argv) { double f;
- >
- > f = atof(argv[1]);
- > printf("The answere is %f\n", f);
- >}
- >
- >The core dumps. (passing an argument of "0.50").
- >
-
- JL> I am running gcc 2.3.3 with all the new libs (gcc 486 binaries)
- JL> and the new binutils, on an i486/33. I compiled the same program
- JL> and also experienced the "core dump" problem.
-
- JL> Also, my kernel is compiled without math emulation, if that helps
- JL> in narrowing down the possibilities ...
-
- Declare argv as *argv[] as it should be declared and the 'problem'
- disappears. argv is not a pointer to a character, it's an array of
- such pointers. The above program would bomb even on a Cray.
-
- Cheers,
-
- /Muhammad M. Saggaf | Stop the genocide
- alsaggaf@athena.mit.edu | Save Bosnia
-
-