Hello,
Apparently there is an environmental variable that runs before Fluent.
It's named CORTEX_PRE
If you set this as an environmental variable, this will force vglrun to
redirect Fluent when called from Workbench. Now the VGL_LOGO appears when
opening Fluent from workbench, and it also shows as being executed on the
GPU using "nvidia-smi".
setenv CORTEX_PRE "/bin/vglrun"
You may want to put the full path to vglrun just in case.
I am using modules, so added it to my ansys/192 module file
#%Module1.0
proc ModulesHelp { } {
global version prefix
puts stderr "\tANSYS 19.2"
}
module-whatis "ANSYS 19.2"
setenv FLUENT_WB_OPTIONAL_ARGS "-driver opengl"
setenv CORTEX_PRE "vglrun"
prepend-path PATH /ansys_inc/v192/bin
prepend-path PATH /ansys_inc/v192/CFX/bin
prepend-path PATH /ansys_inc/v192/fluent/bin
prepend-path PATH /ansys_inc/v192/icemcfd/linux64_amd/bin
prepend-path PATH /ansys_inc/v192/Framework/bin/Linux64
prepend-path PATH /ansys_inc/v192/TurboGrid/bin
prepend-path PATH /ansys_inc/shared_files/licensing/lic_admin
prepend-path AWP_ROOT192 /ansys_inc/v192
prepend-path ANSYS192_DIR /ansys_inc/v192/ansys
prepend-path ANSYSLIC_DIR /ansys_inc/shared_files/licensing
prepend-path LD_LIBRARY_PATH /usr/lib64/nvidia
prepend-path LD_LIBRARY_PATH /ansys_inc/v192/Framework/Linux64/bin
Thanks.
Josh
On Mon, Oct 29, 2018 at 2:02 PM DRC <[email protected]> wrote:
> If you see the VGL logo, that means that hardware acceleration is
> working. The logo is drawn by VirtualGL after the OpenGL-rendered
> pixels are read back from the GPU and before they are drawn into the X
> window, so it is guaranteed that, if you see the logo, the underlying
> pixels were rendered by the GPU.
>
> This has been a long-standing problem, and I'm not aware that anyone has
> discovered a workaround. I'll repeat what I said in the message on the
> old SF Mailing list: if ANSYS is claiming that they "support" VirtualGL
> as a platform, then it's their responsibility to fix this or suggest how
> I can work around it in VGL. Given that other ANSYS applications work
> properly, I strongly suspect that this isn't my problem. At this point,
> VirtualGL is 14 years old, and I have bent over backwards to accommodate
> enterprise ISVs and their wonky software, but it would be nice if they
> made even one step in my direction for a change. It would be even nicer
> if companies like ANSYS stepped forward to fund my labor in supporting
> their applications, given that they're claiming that they "support"
> VirtualGL. Otherwise, I'm flat broke from giving away too much of my
> work to the open source community for free. Can't do it anymore.
>
> DRC
>
> On 10/28/18 12:14 PM, Joshua wrote:
> > If the VGL_LOGO appears using TigerVNC, does that mean hardware
> > acceleration is working, or would the VGL_LOGO appear even if it was
> > running using the GLX extensions, unaccelerated?
> >
> >
> > On Sunday, October 28, 2018 at 1:05:51 PM UTC-4, Joshua wrote:
> >
> > Hello,
> >
> > I am having the same issues as were reported here for Fluent 16.2
> >
> > https://sourceforge.net/p/virtualgl/mailman/message/34859603/
> > <https://sourceforge.net/p/virtualgl/mailman/message/34859603/>
> >
> > I can launch the standalone version of Fluent 19.2 and get the
> > VGL_LOGO to appear with VGL_LOGO, but when launching from within
> > workbench, I am not able to get it to work. In general you get
> > "Warning: Cannot find accelerated drivers. Performance might be
> > severely impacted." warning, but then you can set "setenv
> > FLUENT_WB_OPTIONAL_ARGS "-driver opengl". However, this appears to
> > not use hardware rendering and the VGL_LOGO does not appear.
> >
> > All the other applications (Mechanical, Meshing, Design Modeler)
> > launch from workbench with the VGL_LOGO, and VGL_VERBOSE shows
> > VirtualGL being called, so these seem to be working correctly.
> >
> > I am using TigerVNC + VirtualGL.
> >
> > All the glxspheres64 and glxgears, etc. tests work fine and the
> > VGL_LOGO appear in those and VGL_VERBOSE shows that VirtualGL is
> > being called.
> >
> > Has anyone had any luck with Fluent from Workbench since v16.2?
> >
> > Thanks.
> >
> > Josh
> >
> >
> >
> >
> > --
> > You received this message because you are subscribed to the Google
> > Groups "VirtualGL User Discussion/Support" group.
> > To unsubscribe from this group and stop receiving emails from it, send
> > an email to [email protected]
> > <mailto:[email protected]>.
> > To view this discussion on the web visit
> >
> https://groups.google.com/d/msgid/virtualgl-users/6b00c485-4386-41e8-b8f5-e0b68f336378%40googlegroups.com
> > <
> https://groups.google.com/d/msgid/virtualgl-users/6b00c485-4386-41e8-b8f5-e0b68f336378%40googlegroups.com?utm_medium=email&utm_source=footer
> >.
> > For more options, visit https://groups.google.com/d/optout.
>
> --
> You received this message because you are subscribed to the Google Groups
> "VirtualGL User Discussion/Support" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to [email protected].
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/virtualgl-users/5abac8de-9cbe-b96c-f420-004e7326a901%40virtualgl.org
> .
> For more options, visit https://groups.google.com/d/optout.
>
--
You received this message because you are subscribed to the Google Groups
"VirtualGL User Discussion/Support" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To view this discussion on the web visit
https://groups.google.com/d/msgid/virtualgl-users/CAO2PkyxpHT2iwkhdKgzPyxaGnt%2BP-5WawkASy4-1XAnByddC-g%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.
