in gfortran, buffered direct I/O reads wrong record

[Miguel Martinez]

Binary package hint: gcc-4.4

In ubuntu, gfortran 4.3 and 4.4 (in both Karmic and Lucid) have issues with ph.x, a binary of the quantum-Espresso suite:
http://www.quantum-espresso.org/download.php

In order to compile it, untar and after./configure try "make pw ph". This should generate the pw.x and ph.x binaries together with a few related tools. Phonon calculations as seen in example02, example06 and example07 produce different results from the reference ones.

This is better explained in upstream bug 43551:
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43551

Is it possible to have this fixed for Lucid? Would it qualify for an SRU towards Karmic?

ProblemType: Bug
DistroRelease: Ubuntu 10.04
Package: gfortran-4.4 4.4.3-4ubuntu5
ProcVersionSignature: Ubuntu 2.6.32-20.30-generic 2.6.32.11+drm33.2
Uname: Linux 2.6.32-20-generic x86_64
Architecture: amd64
Date: Tue Apr 13 22:23:18 2010
InstallationMedia: Ubuntu 9.10 "Karmic Koala" - Release amd64 (20091027)
ProcEnviron:
 PATH=(custom, user)
 LANG=es_ES.utf8
 SHELL=/bin/bash
SourcePackage: gcc-4.4

[Matthias Klose (doko)]

yes, candidate for a SRU, unlikely for the release

[Rolf Leggewie (r0lf)]

lucid has seen the end of its life and is no longer receiving any updates. Marking the lucid task for this ticket as "Won't Fix".