Gentoo Websites Logo
Go to: Gentoo Home Documentation Forums Lists Bugs Planet Store Wiki Get Gentoo!
Bug 122925 - please add ~amd64 to sci-visualization/gfsview-0.4
Summary: please add ~amd64 to sci-visualization/gfsview-0.4
Status: RESOLVED FIXED
Alias: None
Product: Gentoo Linux
Classification: Unclassified
Component: New packages (show other bugs)
Hardware: AMD64 Linux
: High enhancement
Assignee: AMD64 Project
URL:
Whiteboard:
Keywords:
Depends on: 122922 122924
Blocks:
  Show dependency tree
 
Reported: 2006-02-15 07:11 UTC by Markus Dittrich (RETIRED)
Modified: 2006-03-16 21:10 UTC (History)
1 user (show)

See Also:
Package list:
Runtime testing required: ---


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description Markus Dittrich (RETIRED) gentoo-dev 2006-02-15 07:11:08 UTC
According to the user who submitted the ebuild originally, sci-visualization/gfsview-0.4
should compile fine on amd64.

To test that gfsview works properly, please download

http://dev.gentoo.org/~markusle/sci-overlay/gfsview-test/boussinesq.gfs
http://dev.gentoo.org/~markusle/sci-overlay/gfsview-test/boussinesq.gfv
http://dev.gentoo.org/~markusle/sci-overlay/gfsview-test/cylinder.gts

and do a

gerris2D boussinesq.gfs | gfsview2D > /dev/null

This should fire up gfsview and provide you with a real time view 
of a fluid dynamics simulation.

Thanks,
Markus
Comment 1 Tim Cera 2006-02-16 16:01:42 UTC
I have discovered a problem with ImageMagick on my amd64.  This becomes a problem with boussinesq.gfs because there are three "EventScript..." commands that tells gfsview to use ImageMagick commands to make ppm and eps files.  The only way that I have been able to make the simulation work is to comment out those three "EventScript..." lines in boussinesq.gfs.

I don't know whether this ImageMagick problem is on my machine, or a problem with gerris/gfsview.

There are other examples that install with gerris.  See /usr/share/doc/gerris-0.8/examples.  The rt example also needs editing, but in that case I had to comment out the "OutputPPM..." blocks (made up of multiple lines!).

I forgot about the dependency on ImageMagick when I submitted the first ebuild.  I will put together a new one with ImageMagick as a dependency (maybe with the use keywords?).

Thanks for evaluating gerris and gfsview.
Comment 2 Markus Dittrich (RETIRED) gentoo-dev 2006-02-17 03:30:50 UTC
(In reply to comment #1)
> I have discovered a problem with ImageMagick on my amd64.  This becomes a
> problem with boussinesq.gfs because there are three "EventScript..." commands
> that tells gfsview to use ImageMagick commands to make ppm and eps files.  The
> only way that I have been able to make the simulation work is to comment out
> those three "EventScript..." lines in boussinesq.gfs.
> 

The last EventScript line is a call to "convert" and any problem with it indicates
and issue with Imagemagick rather than gfsview. As such, Imagemagick
is not a dependency of gfsview but rather of the example I have chosen.

However, glancing at the gfsview's source code it looks like the ppm generating
routine is part of gfsview itself and needs fixing in case it shouldn't work. Works
just fine here on my P4 box. 

What exactly is the problem?

best,
Markus
Comment 3 Tim Cera 2006-02-17 15:32:05 UTC
The boussinesq simulation...

% gerris2D boussinesq.gfs | gfsview2D > /dev/null
step:       0 t:      0.00000000 dt:  1.000000e+00 cpu:      0.00000000
Balance summary: 1 PE
  domain size:
      min:       647 avg:       647         |       0 max:       647
Approximate projection
    niter:    0
    residual.bias:    0.000e+00  0.000e+00
    residual.first:   0.000e+00  0.000e+00      0
    residual.second:  0.000e+00  0.000e+00      0
    residual.infty:   0.000e+00  0.000e+00      0
Segmentation fault
Exit 139

What led me to think it was ImageMagick was the result of trying the rt example.

% shapes rayleigh-taylor | transform --sy 2. | transform --ty 0.5 > rt.gts && gerris2D rt.gfs | gfsview2D rt.gfv
step:       0 t:      0.00000000 dt:  1.000000e+00 cpu:      0.00000000
Balance summary: 1 PE
  domain size:
      min:     87380 avg:     87380         |       0 max:     87380
Approximate projection
    niter:    0
    residual.bias:    0.000e+00  0.000e+00
    residual.first:   0.000e+00  0.000e+00      0
    residual.second:  0.000e+00  0.000e+00      0
    residual.infty:   0.000e+00  0.000e+00      0
U diffusion
    niter:    0
    residual.bias:    0.000e+00  0.000e+00
    residual.first:   0.000e+00  0.000e+00      0
    residual.second:  0.000e+00  0.000e+00      0
    residual.infty:   0.000e+00  0.000e+00      0
**ERROR: [mpeg2enc] Could not read YUV4MPEG2 header: stream requires unsupported features!
**ERROR: [mpeg2enc] Could not read YUV4MPEG2 header: stream requires unsupported features!
gfsview: <stdin>: Broken pipe

In both cases gfsview flashes for a moment on the screen and then is gone.  Comment out the attempts to make output files with ImageMagick and gfsview works fine.
Comment 4 Markus Dittrich (RETIRED) gentoo-dev 2006-02-27 07:42:20 UTC
Dear amd64 team,

Tim Cera was so kind to create five test sets for gfsview to allow
for proper testing of the package. You can grab the files at the
following locations; each test can then simply be run by the respective
test_*.sh shell script
http://dev.gentoo.org/~markusle/sci-overlay/gfsview-test/boussinesq/
http://dev.gentoo.org/~markusle/sci-overlay/gfsview-test/cylinder/
http://dev.gentoo.org/~markusle/sci-overlay/gfsview-test/rt/
http://dev.gentoo.org/~markusle/sci-overlay/gfsview-test/shear/
http://dev.gentoo.org/~markusle/sci-overlay/gfsview-test/cylinder/heated/

Thanks,
Markus
Comment 5 Scott Stoddard (RETIRED) gentoo-dev 2006-03-16 21:10:41 UTC
Added ~amd64.  Thanks both for the test scripts.