LAPACK Archives

[Lapack] incorrect RWORK size for ZGESDD

Dear Kevin,
Just to let you know we corrected the bug (bug 111)  in our repository. We 
confirmed the bug and your proposed fix.
We should be releasing your fix with the main LAPACK release - hopefully before 
the summer.
Thank you for reporting the issue
Regards,
Julie

On Aug 15, 2013, at 10:18 AM, julie langou <julie@Domain.Removed> wrote:

Thank you Kevin for reporting the problem
we added the bug to our bug list
http://netlib.org/lapack/Errata/index2.html 
<http://netlib.org/lapack/Errata/index2.html>

We should take a look at it very soon.
Regards,
Julie
On Aug 13, 2013, at 1:13 AM, Kevin Squire <kevin.squire@Domain.Removed 
<mailto:kevin.squire@Domain.Removed>> wrote:

I'm one of the developers of Julia.  We were following the recommendation 
for RWORK size calculation from the ZGESDD header here ( 
http://netlib.org/lapack/complex16/zgesdd.f 
<http://netlib.org/lapack/complex16/zgesdd.f>), and were finding that when 
INFO='N', the size was insufficient, and generally led to a segmentation 
fault.

The current definition says that "If JOBZ = 'N', LRWORK >= 5*min(M,N)", but 
we found that the minimum size needed to be at least 7*min(M,N).  This was 
through experimentation and comparison with code from Numpy and Octave, so 
you may wish to verify this through the code.

Relevant Julia links:
 
<https://github.com/JuliaLang/julia/commit/484a9f0892>https://github.com/JuliaLang/julia/issues/3966
 <https://github.com/JuliaLang/julia/issues/3966> 
fix: https://github.com/JuliaLang/julia/commit/484a9f0892 
<https://github.com/JuliaLang/julia/commit/484a9f0892>
https://github.com/JuliaLang/julia/issues/4016 
<https://github.com/JuliaLang/julia/issues/4016> 
fix: https://github.com/JuliaLang/julia/commit/9f6058804c 
<https://github.com/JuliaLang/julia/commit/9f6058804ca2b>
Thanks,

   Kevin
_______________________________________________
Lapack mailing list
Lapack@Domain.Removed <mailto:Lapack@Domain.Removed>
http://lists.eecs.utk.edu/mailman/listinfo/lapack


-------------- next part --------------
An HTML attachment was scrubbed...
URL: 
<http://lists.eecs.utk.edu/mailman/private/lapack/attachments/20150317/0f51a9e6/attachment.html>

<Prev in Thread] Current Thread [Next in Thread>


For additional information you may use the LAPACK/ScaLAPACK Forum.
Or one of the mailing lists, or