LAPACK-3.2 errata file
updated on Tue Feb 10 2009
maintained by J. Langou, U. Colorado Denver

Released
========

(*) bug :: LAPACK-3.2.0 does not have manpages

    o reported on Fr. Dec 05 2008 by Tobias Burnus
    o manpages generated and released by Julie Langou on Jan 2nd 2009

Improvements in SVN to be released
==================================

(*) change xLARRD (MRRR routines) to deal with some matrices from Godunov
    svn :: r630

    o improvements provided by Christof Voemel on Tue Jan 27 2009
    o commited by Julien Langou (SVN revision 630) on Wed Jan 28 2009
    o see: svn log from Julien

Corrected in SVN to be released
===============================

(*) bug :: in xLAQR2 and xLAQR3, access nonallocated part of workspace array WORK during workspace query (LWORK=-1)
    svn :: r603

    o reported by Edouard Canot (IRISA) on Tue Dec 02 2008 by mail to lapack@cs.utk.edu
    o confirmed and corrected by Julien Langou and Edward Smyth
    o commited by Edward Smyth (SVN revision 603, on Wed Dec 03 2008)
    o see: emails

(*) bug :: illegal memory access in ILACLC,ILACLR, ILADLC,ILADLR,ILASLC,ILASLR,ILAZLC and ILAZLR
    svn :: r610, r614

    o reported on Fr. Dec 19 2008 on forum topic 854
      and on http://groups.google.com/group/comp.lang.fortran/browse_thread/thread/635192e11beadb93
      and by email from Tobias Burnus.
    o commited by Julien Langou (SVN revision 610) for ILASLR on Dec 19 2008
      commited by Jason Riedy (SVN revision 614) for ILACLC,ILACLR, ILADLC,ILADLR,ILASLC,ILASLR,ILAZLC and ILAZLR on Dec 27 2008
    o see: svn log from Jason
      see: email from Tobias

(*) bug :: illegal memory access in xTFSM
    svn :: r622

    o reported by Julie and Jason
    o commited by Julien Langou (SVN revision 622) on Sun Jan 11 2009
    o see: svn log from Julien

(*) bug :: out-of-bound reference in BLAS routines DROTM and SROTM
    svn :: r624

    o reported/fixed/commited by Julie Langou on Mon Jan 12 2009
    o see: svn log from Julie

(*) bug :: wrong leading dimension used in xGHEQZ at line 780
    svn :: r632

    o reported by Vasile Sima on Thu Jan 22 2009
    o commited by Julien Langou (SVN revision 632) on Sun Feb 01 2009
    o see: email from Vasile

(*) bug :: mismatch argument types between calling routine and called routine
    svn :: r633, r635, r636, r637

    o reported by faber80 on Sat Feb 7 2009
    o confirmed by Julien Langou on Mon Feb 9 2009
    o ZCPOSV corrected by Julien (r633) on Mon Feb 9 2009
    o CGBRFSX, CGERFSX, CHERFSX, CHESVXX, CPORFSX, CPOSVXX, CSYRFSX, CSYSVXX, ZGBRFSX,
      ZGERFSX, ZHERFSX, ZHESVXX, ZPORFSX, ZPOSVXX, ZSYRFSX, and ZSYSVXX corrected by
      Deaglan (r635, r636, r637)
    o see: emails

Confirmed problems -- bug needs to be fixed!
============================================

(*) bug :: 'make install' generates a "*** Error: Time for operations was zero" too easily
(*) bug :: negative flop counts can be generated by 'make install'

    o reported by Sergei Steshenko on Thu Nov 20 2008
    o see: emails

(*) bug :: in DGESDD, workspace query gives a value smaller than the minimal value given in the header to run the routine

    o reported by  Guy Bencteux on Sat Dec 06 2008
    o see forum topic 846

(*) bug :: workspace query bug with DGEJSV (when only left or right singular vectors desired)

    o reported by Allin Cottrell on Tue Jan 15 2009
    o confirmed by Julien Langou on Fri Jan 30 2009
    o see forum topic 866
    o all information sent to Zlatko Drmac

Bug reports -- bug needs to be confirmed!
=========================================

(*) bug :: serious problem in DGESDD: DGESDD returns negative singular values with INFO = 0

    o reported by Guy Bencteux on Sat Dec 06 2008
    o see forum topic 847

(*) bug :: problem when playing with DGEEV and gfortran optimization flags

    o reported by Pauli Virtanen (scipy team) on Sat Dec 06 2008 (email to lapack@cs.utk.edu)
    o see: email from Pauli Virtanen
      see: http://scipy.org/scipy/scipy/ticket/709

(*) bug :: misleading comment in the header of DGHEQZ, the 2x2 block of the real Schur form are not as claimed

    o reported by Vasile Sima on Tue Jan 27 2009
    o see: email from Vasile
    o more information:
      more specifically Vasile is reporting that the comments:

*          [...] H contains the upper quasi-triangular
*          matrix S from the generalized Schur factorization;
*          2-by-2 diagonal blocks (corresponding to complex conjugate
*          pairs of eigenvalues) are returned in standard form, with
*          H(i,i) = H(i+1,i+1) and H(i+1,i)*H(i,i+1) < 0.

     are not correct

(*) bug :: failure of convergence of DHGEQZ for some specific matrix pencils

    o reported by Vasile Sima on Thu Jan 22 2009
    o see: email from Vasile
    o more information at: lapack known issues
     
Deprecated routines
===================

(*) dead code :: xLAHRD is now replaced by xLAHR2, xLAHRD is a dead code since LAPACK 3.1

    o xLAHRD will be removed at some point in the future from the distribution

Previous version of this file
=============================

   o Mon Feb 01 2009