[Maxima-bugs] [ maxima-Bugs-2913614 ] LAPACK: dgesvd is broken

[SourceForge.net <no...@so...>]

Bugs item #2913614, was opened at 2009-12-13 05:40
Message generated for change (Comment added) made by rtoy
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=104933&aid=2913614&group_id=4933

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: Share Libraries
Group: None
Status: Open
Resolution: None
Priority: 5
Private: No
Submitted By: Andrej Vodopivec (andrejv)
Assigned to: Nobody/Anonymous (nobody)
Summary: LAPACK: dgesvd is broken

Initial Comment:
(%i4) A:genmatrix(lambda([i,j], random(2.0)), 4, 4)$
(%i5) dgesvd(A);
Maxima encountered a Lisp error:
 Error during processing of --eval option "(cl-user::run)":
  The value "G" is not of type (SIMPLE-ARRAY DOUBLE-FLOAT (*)).
Automatically continuing.
To enable the Lisp debugger set *debugger-hook* to nil.

(%i6) build_info();
Maxima version: 5.20post
Maxima build date: 8:37 12/12/2009
Host type: i386-apple-darwin9.8.0
Lisp implementation type: SBCL
Lisp implementation version: 1.0.19

----------------------------------------------------------------------

>Comment By: Raymond Toy (rtoy)
Date: 2009-12-14 08:57

Message:
I cannot reproduce this either with cmucl on mac os x.  Running this with
maxima -g and looking at the Lisp error will help a lot in figuring this
problem out.

The boolean results are  documented; it means the left and right singular
vectors were not requested.  Perhaps that should be changed, but that's a
different issue.

----------------------------------------------------------------------

Comment By: Dieter Kaiser (crategus)
Date: 2009-12-13 12:39

Message:
On my system I get:

(%i2) build_info();

Maxima version: 5.20post
Maxima build date: 17:36 12/13/2009
Host type: i686-pc-linux-gnu
Lisp implementation type: CLISP
Lisp implementation version: 2.44.1 (2008-02-23) (built 3436700604)
(memory 3469710995)

(%o2) 
(%i3) A:genmatrix(lambda([i,j], random(2.0)), 4, 4)$
(%i4) dgesvd(A);
(%o4) [[3.755933922081446, 1.830035687063915, .7142611635301948, 
                                              0.127875116930834], false,
false]

I do not get a Lisp error. But the result contains boolean values.

Dieter Kaiser

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=104933&aid=2913614&group_id=4933