Your message dated Tue, 29 Jan 2008 10:27:29 -0500
with message-id <[EMAIL PROTECTED]>
and subject line Sorry but realize the error is a floating point aproximation
problem
has caused the attached Bug report to be marked as done.
This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
Bug report if necessary, and/or fix the problem forthwith.
(NB: If you are a system administrator and have no idea what I am
talking about this indicates a serious mail system misconfiguration
somewhere. Please contact me immediately.)
Debian bug tracking system administrator
(administrator, Debian Bugs database)
--- Begin Message ---
Package: php5-common
Version: 5.2.0-8+etch10
Severity: important
I am not sophisticated enough to know which module in php is doing this.
It is generated by the round function. We found it because we
discovered rounding errors in processes we were doing using mysql to do
calculations on a large number of records. I was able to isolate the
problem to the round function which does the same thing whether it is
written round(17.775 * 100)/100 or round(17.775,2). It returns 17.77
as a result. I know that there are many other operations of this
function that work correctly because of other calculations. This
problem occurs when running under linux-image-2.6.18-5-amd64.
Under linux-image-2.6.18-5 it does fail to round not even when the
processor is an amd athlon 64 X2. We have tested it on several intel
processor machines without finding the problem.
-- System Information:
Debian Release: 4.0
APT prefers stable
APT policy: (500, 'stable')
Architecture: i386 (i686)
Shell: /bin/sh linked to /bin/bash
Kernel: Linux 2.6.18-5-686
Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8)
Versions of packages php5-common depends on:
ii sed 4.1.5-1 The GNU sed stream editor
php5-common recommends no packages.
-- no debconf information
--- End Message ---
--- Begin Message ---
I understand that the floating point numbers approximate the decimal
values and when that value is slightly less than a 1/2 cent the round
function does not round up. There is apparently a difference between
how they are handled in the AM64 processor. Perhaps a different level
of precision.
Tony
--- End Message ---
