[Bug libc/11422] sin(x) (actually probably all trig) is inaccurate for large x

["simon dot fenney at imgtec dot com" <sourceware-bugzilla at sourceware dot org>]

------- Additional Comments From simon dot fenney at imgtec dot com  2010-03-24 10:46 -------
With all due respect, it can't just be changes in the CPU. I just ran code
compiled with  
  gcc version 3.4.6 20060404 (Red Hat 3.4.6-11)
with libc version 2.3.4
i.e.
   GNU C Library stable release version 2.3.4, by Roland McGrath et al.
   Copyright (C) 2005 Free Software Foundation, Inc.
   This is free software; see the source for copying conditions.
 
   Compiled by GNU CC version 3.4.6 20060404 (Red Hat 3.4.6-11).
   Compiled on a Linux 2.4.20 system on 2009-06-01.

On a 4 month old processor:
       vendor_id       : GenuineIntel
       cpu family      : 6
       model           : 26
       model name      : Intel(R) Xeon(R) CPU           L5520  @ 2.27GHz
       stepping        : 5
       cpu MHz         : 2268.000

and *THAT* works. (I have also tried this build with older and non-Intel CPUs)


The bug, OTOH, occurs with (amongst others) 
    gcc version 4.4.1 (Ubuntu 4.4.1-4ubuntu9)

    GNU C Library (EGLIBC) stable release version 2.10.1, 
    Compiled by GNU CC version 4.4.1.
    Compiled on a Linux >>2.6.24-23-server<< system on 2010-01-03.

with both new CPUs (e.g. Atom) and those that are several years old, eg
   "Intel(R) Pentium(R) 4 CPU 2.80GHz"

Something else *must* have changed, either in the source code for the maths
library OR in the way gcc compiles it. The gcc guys are claiming it's not their
responsibility (http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43490#c7) but I
don't think it's fair to lay the blame on the CPU manufacturers.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |REOPENED
         Resolution|INVALID                     |


http://sourceware.org/bugzilla/show_bug.cgi?id=11422

------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.