No subject


Wed May 9 12:08:04 EDT 2012


the gui, use mesa, and set to not use x. With the current build it looks
like I can't do this anymore in this order.

Andy

On Thu, Jun 7, 2012 at 3:50 PM, Marcus D. Hanwell <
marcus.hanwell at kitware.com> wrote:

> On Wed, Jun 6, 2012 at 2:47 PM, Fabian, Nathan <ndfabia at sandia.gov> wrote:
> > In a git pull from yesterday, it appears that VTK_USE_X is being marked
> > internal somehow, so that I can't set it from ccmake.  I am trying to
> > setup OSMESA and normally I think when checking OPENGL_HAS_OSMESA, it
> > changes VTK_USE_X from internal to visible, but I'm not really sure where
> > to look for that.
> >
> > Any help?
> >
> Why do you think VTK_USE_X is set to internal? I refactored the OS
> Mesa support in VTK (and hence ParaView). I just pulled ParaView
> master, made a clean build directory with no arguments (cmake
> ~/src/ParaView), and VTK_USE_X is there (in advanced) and turned ON.
>
> You can see a selection of dashboard submissions including
> http://open.cdash.org/buildSummary.php?buildid=2339963 on VTK that
> test OS Mesa each night, and they are passing. I preserved the same
> variables (searching MESA in the cache shows VTK_OPENGL_HAS_OSMESA
> etc), so that the wiki instructions continued to work.
>
> I can look into this further, but I am not aware of any CMake code
> making these variables internal (and in my local tests they are
> advanced as they have always been). I would appreciate feedback if any
> regressions were introduced I can get them fixed up.
>
> Thanks,
>
> Marcus
> _______________________________________________
> Paraview-developers mailing list
> Paraview-developers at paraview.org
> http://public.kitware.com/mailman/listinfo/paraview-developers
>

--e89a8f22c6f51b6ca004c1e75bfa
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

If you turn off PARAVIEW_BUILD_QT_QUI then VTK_USE_X disappears as an optio=
n. Checking CMakeCache.txt it is still set to ON. If I turn back PARAVIEW_B=
UILD_QT_GUI though the VTK_USE_X disappears.<br><br>From past builds withou=
t the GUI and with mesa I would choose to not build the gui, use mesa, and =
set to not use x. With the current build it looks like I can&#39;t do this =
anymore in this order. <br>
<br>Andy<br><br><div class=3D"gmail_quote">On Thu, Jun 7, 2012 at 3:50 PM, =
Marcus D. Hanwell <span dir=3D"ltr">&lt;<a href=3D"mailto:marcus.hanwell at ki=
tware.com" target=3D"_blank">marcus.hanwell at kitware.com</a>&gt;</span> wrot=
e:<br>
<blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=
x #ccc solid;padding-left:1ex"><div class=3D"im">On Wed, Jun 6, 2012 at 2:4=
7 PM, Fabian, Nathan &lt;<a href=3D"mailto:ndfabia at sandia.gov">ndfabia at sand=
ia.gov</a>&gt; wrote:<br>

&gt; In a git pull from yesterday, it appears that VTK_USE_X is being marke=
d<br>
&gt; internal somehow, so that I can&#39;t set it from ccmake. =A0I am tryi=
ng to<br>
&gt; setup OSMESA and normally I think when checking OPENGL_HAS_OSMESA, it<=
br>
&gt; changes VTK_USE_X from internal to visible, but I&#39;m not really sur=
e where<br>
&gt; to look for that.<br>
&gt;<br>
&gt; Any help?<br>
&gt;<br>
</div>Why do you think VTK_USE_X is set to internal? I refactored the OS<br=
>
Mesa support in VTK (and hence ParaView). I just pulled ParaView<br>
master, made a clean build directory with no arguments (cmake<br>
~/src/ParaView), and VTK_USE_X is there (in advanced) and turned ON.<br>
<br>
You can see a selection of dashboard submissions including<br>
<a href=3D"http://open.cdash.org/buildSummary.php?buildid=3D2339963" target=
=3D"_blank">http://open.cdash.org/buildSummary.php?buildid=3D2339963</a> on=
 VTK that<br>
test OS Mesa each night, and they are passing. I preserved the same<br>
variables (searching MESA in the cache shows VTK_OPENGL_HAS_OSMESA<br>
etc), so that the wiki instructions continued to work.<br>
<br>
I can look into this further, but I am not aware of any CMake code<br>
making these variables internal (and in my local tests they are<br>
advanced as they have always been). I would appreciate feedback if any<br>
regressions were introduced I can get them fixed up.<br>
<br>
Thanks,<br>
<br>
Marcus<br>
<div class=3D"HOEnZb"><div class=3D"h5">___________________________________=
____________<br>
Paraview-developers mailing list<br>
<a href=3D"mailto:Paraview-developers at paraview.org">Paraview-developers at par=
aview.org</a><br>
<a href=3D"http://public.kitware.com/mailman/listinfo/paraview-developers" =
target=3D"_blank">http://public.kitware.com/mailman/listinfo/paraview-devel=
opers</a><br>
</div></div></blockquote></div><br>

--e89a8f22c6f51b6ca004c1e75bfa--


More information about the Paraview-developers mailing list