Linux ALPHA glibc 2.1.2-??

Linux ALPHA glibc 2.1.2-??

Post by U.V. Ravindr » Fri, 16 Jun 2000 04:00:00



How does one find out the differences between libc-2.1.2-28
and libc-2.1.2-32?

Advance thanks for all your replies.

-uvr

Sent via Deja.com http://www.deja.com/
Before you buy.

 
 
 

1. small doubles, g++, glibc, and linux/alpha problems

g++ 2.7 and  egcs 1.0.1 running on  a linux-alpha-glibc
system
is not taking kindly to small double values - those funky
near zero values that you know should be zero, but have
stupid values
like 2.38723e-314.

The problem does not show itself on my linux-intel-glibc
system.

#include <iostream.h>
#include <cmath>

int main()
{
        cerr << "hello" << endl;

        double u = 2.38723e-314;

        cerr  << "u = " << u << " M_PI = " << M_PI << endl;
        cerr << "u * M_PI = " << u * M_PI << endl;

        return 0;

The results:

on linux - alpha - glibc
+++++++++++++++++++++++++++++++++++++


2.7.1

hello
u = Floating point exception (core dumped)

egcs 1.0.1

hello
u = 2.38723e-314 M_PI = 3.14159
u * M_PI = Floating point exception (core dumped)

slightly different results, but both hosed.

on linux - intel - glibc
+++++++++++++++++++++++++



hello
u = 2.38723e-314 M_PI = 3.14159
u * M_PI = 7.4997e-314

No problem ( with either compiler ) .

Can anybody explain this? Is it a known problem?  Can it be
fixed. Must I
resort to float's?

Thanks for any help.

cliff

2. compile option -Wall

3. glibc on R.H 4.1 Alpha Linux Replaceble ?

4. Interrupted system call and sockets --> Help

5. Linux for Alphas (was Re: OSes for Alphas)

6. Problems with Xconfig for old Gateway 2000

7. Troubling Alpha: SMP support of Compaq Fort under Linux/Alpha

8. How to install compiled GNU tools?

9. Upgrading glibc-2.1.1 to glibc-2.1.2

10. glibc 2.1 and glibc 2.2

11. glibc-2 -> glibc-2.1.x

12. glibc 586 vs glibc 386

13. undefined reference /lib/lib.so.6 (glibc-2.2.3 -> glibc-2.2-9)