g++ and floating point calculations - bug?

From: Ross A Macintyre <raz_at_cee.hw.ac.uk>
Date: Mon, 13 Nov 1995 14:48:03 +0000 (GMT)

Are there known problems with g++ and floating point precision on the Alphas?
We're running gcc 2.6.0 and libg++-2.6.

I have had 2 students complain recently that
1. setprecision(2) did not work properly,
and
2. The bug is with iomanip and floating point values... in that I say
   "1 decimal place, please" and get about 5,

This is only on the alphas and I am told the code works fine for the same
versions of gcc/libg++ on other architectures.

If this is a known bug and is fixed in a certain release, could you tell me
what in what release it is fixed?
I don't want to change all the versions of gcc in mid term and would like to
know if this could easily be fixed by say a quick upgrade of libg++.

Thanks in advance,

Ross
Received on Mon Nov 13 1995 - 22:14:56 NZDT

This archive was generated by hypermail 2.4.0 : Wed Nov 08 2023 - 11:53:46 NZDT