This is the mail archive of the cygwin mailing list for the Cygwin project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: Octave/HDF5 library/header mismatch


On 21/06/2016 06:20, Gavin King wrote:
Hello

My apologies if this is a bit verbose, or if the terminology is a bit
wrong; I know enough to get myself into trouble but not enough to
communicate problems well.

Hi Gavin,
nice to hear you.


After updating cygwin, a working script failed, causing octave to exit
(with an error code 6).

A minimal example is attached; it opens octave, creates a simple vector
of values, and tries to save the workspace to disc. The error message is
at the end.

It appears there is a mismatch between the compiled octave version and
the most recent version of the HDF5 headers.

The current version of the package octave is 4.0.1-1. Reverting it to
the previous version makes no difference.

Reverting the package libhdf5_10 from the current version 1.8.17-1 to
the previous version 1.8.16-1 means that octave 4.0.1-1 can save the
workspace.

So: workaround: revert libhdf5_10 to 1.8.16-1. Presumably, the current
version of octave was compiled at an inopportune point; I don't think
this is something that I can remedy myself, is it?

the official workaround is mentioned on the warning message

"You can, at your own risk, disable this warning by setting the
environment variable 'HDF5_DISABLE_VERSION_CHECK' to a value of '1'.
Setting it to 2 or higher will suppress the warning messages totally.
Headers are 1.8.16, library is 1.8.17"

so if you use, before running octave

  export HDF5_DISABLE_VERSION_CHECK=1

you will still see the warning but the action will be completed

  export HDF5_DISABLE_VERSION_CHECK=2

will suppress also the warning.


I think this is the correct place to report this to; it isn't a problem
per se with the individual programs, but with the way they're packed
together. I think. Do correct me if I am wrong; I'm a bit new to this
whole business.


It is the right place.
This problem is clearly a hdf5 upstream problem and a questionable
design choice.

I am following with them

https://lists.hdfgroup.org/pipermail/hdf-forum_lists.hdfgroup.org/2016-June/009564.html


yours

Gavin King

Regards
Marco


--
Problem reports:       http://cygwin.com/problems.html
FAQ:                   http://cygwin.com/faq/
Documentation:         http://cygwin.com/docs.html
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]