I'm using vglrun for mate-session and startkde so that I don't have to do
it for each individual 3D application.
Here is my VirtuaGL version:
# rpm -q VirtualGL
VirtualGL-2.6.1-20190101.x86_64
VirtualGL-2.6.1-20190101.i386
Running "glxinfo --display :1" shows more Visuals an FBconfig.
My X proxy is StarNet FastX.
I tried downgrading the Nvidia driver to 410.104 and now both cards (Tesla
and Quadro) fail with glxinfo :( . I guess I'll have to go back to the
newer driver that I was using in the first email in this thread.
As far as I can tell, I can't change the GPU operating mode:
-----------------------
# nvidia-smi -q -i 0 | grep -A2 'Operation Mode'
GPU Operation Mode
Current : N/A
Pending : N/A
# nvidia-smi --gom=0 -i 0
GOM mode cannot be changed on GPU 00000000:AF:00.0.
Treating as warning and moving on.
All done.
-----------------------
Thanks,
Jason
---------------------------------------------------------------------------
Jason Edgecombe | Linux Administrator
UNC Charlotte | The William States Lee College of Engineering
9201 University City Blvd. | Charlotte, NC 28223-0001
Phone: 704-687-1943
[email protected] | http://engr.uncc.edu | Facebook
---------------------------------------------------------------------------
If you are not the intended recipient of this transmission or a person
responsible for delivering it to the intended recipient, any disclosure,
copying, distribution, or other use of any of the information in this
transmission is strictly prohibited. If you have received this transmission
in error, please notify me immediately by reply e-mail or by telephone at
704-687-1943. Thank you.
On Tue, Mar 12, 2019 at 4:24 PM DRC <[email protected]> wrote:
> Several things:
>
> - It isn't necessary to run mate-session with vglrun, since MATE is a 2D
> window manager. The only purpose of this would be to avoid having to
> type vglrun to launch other OpenGL applications in your X proxy session.
>
> - You haven't mentioned some important details, such as:
> a. which version of VirtualGL you are using
> b. which X proxy you are using
>
> - '-c proxy' is almost never necessary, since VirtualGL can
> automatically detect when you are using an X proxy.
>
> - On your Tesla card, try running glxinfo directly, without VirtualGL:
>
> glxinfo --display :1
>
> If it still doesn't show any visuals or GLXFBConfigs, then my guess is
> that the Tesla has been configured for compute mode, in which case it
> needs to be switched to graphics mode in order to use it with VirtualGL.
>
> - HFSS is known to work without VGL_DEFAULTFBCONFIG, so once you are
> able to get the Tesla to list visuals and GLXFBConfigs, then HFSS should
> work. But note that it may require setting VGL_SPOILLAST=0 (refer to
> https://cdn.rawgit.com/VirtualGL/virtualgl/2.6.1/doc/index.html#hd0015).
>
>
> On 3/12/19 2:31 PM, Jason Edgecombe wrote:
> > I was playing around with glxinfo and noticed that glxinfo doesn't list
> > any Visuals or FBConfigs on the Tesla card:
> >
> > # Tesla card:
> > % vglrun -d :1 -c proxy glxinfo 2>&1 | grep -A5 -E 'GLX
> > Visuals|GLXFBConfigs' ; echo $?
> > 1
> > # Quadro card:
> > % vglrun -d :2 -c proxy glxinfo 2>&1 | grep -A5 -E 'GLX
> > Visuals|GLXFBConfigs' ; echo $?
> > 120 GLX Visuals
> > visual x bf lv rg d st colorbuffer sr ax dp st accumbuffer ms
> cav
> > id dep cl sp sz l ci b ro r g b a F gb bf th cl r g b a ns b
> eat
> >
> ----------------------------------------------------------------------------
> > 0x021 24 tc 0 24 0 r y . 8 8 8 0 . . 4 24 8 16 16 16 16 0 0
> > None
> > 0x022 24 dc 0 24 0 r y . 8 8 8 0 . . 4 24 8 16 16 16 16 0 0
> > None
> > --
> > 263 GLXFBConfigs:
> > visual x bf lv rg d st colorbuffer sr ax dp st accumbuffer ms
> cav
> > id dep cl sp sz l ci b ro r g b a F gb bf th cl r g b a ns b
> eat
> >
> ----------------------------------------------------------------------------
> > 0x0d5 32 tc 0 24 0 r y . 8 8 8 0 . . 4 24 8 16 16 16 16 0 0
> > None
> > 0x0d6 24 dc 0 24 0 r y . 8 8 8 0 . . 4 24 8 16 16 16 16 0 0
> > None
> > 0
> >
> >
> ---------------------------------------------------------------------------
> > Jason Edgecombe | Linux Administrator
> > UNC Charlotte | The William States Lee College of Engineering
> > 9201 University City Blvd. | Charlotte, NC 28223-0001
> > Phone: 704-687-1943 <tel:704-687-1943>
> > [email protected] <mailto:[email protected]> | http://engr.uncc.edu |
> > Facebook
> >
> ---------------------------------------------------------------------------
> > If you are not the intended recipient of this transmission or a person
> > responsible for delivering it to the intended recipient, any disclosure,
> > copying, distribution, or other use of any of the information in this
> > transmission is strictly prohibited. If you have received this
> > transmission in error, please notify me immediately by reply e-mail or
> > by telephone at
> > 704-687-1943 <tel:704-687-1943>. Thank you.
> >
> >
> > On Tue, Mar 12, 2019 at 3:05 PM Jason Edgecombe <[email protected]
> > <mailto:[email protected]>> wrote:
> >
> > Hi DRC,
> >
> > Things work fine with the Nvidia Quardo M2000, but not the Tesla
> > V100. My test system has both the Tesla V100 and the Quadro M2000
> > cards installed and online at the same time.
> > Setting VGL_DEFAULTFBCONFIG does not have any effect on
> > mate-session, HFSS, or glxinfo when I tested them.
> >
> > So far, I'm having the "Could not obtain RGB visual on the server
> > suitable for off-screen rendering" error with the following programs:
> >
> > * glxinfo
> > * mate-session (RHEL 7.6 with latest
> > EPEL, mate-session-manager-1.16.1-3.el7.x86_64)
> > * Ansys Electromagnetics Desktop 2019R1 (HFSS 19.3)
> >
> > Applications that work without the RGB visual error include:
> >
> > * Matlab 2017b
> > * startkde
> >
> > I'm using the Nvidia and CUDA driver packages
> > from
> http://developer.download.nvidia.com/compute/cuda/repos/rhel7/x86_64
> >
> > Driver details:
> > % cat /proc/driver/nvidia/version
> > NVRM version: NVIDIA UNIX x86_64 Kernel Module 418.39 Sat Feb 9
> > 19:19:37 CST 2019
> > GCC version: gcc version 4.8.5 20150623 (Red Hat 4.8.5-36) (GCC)
> > % rpm -q nvidia-driver
> > nvidia-driver-418.39-4.el7.x86_64
> >
> > Strace confirms that the error is related to glXGetConfig():
> > -------------------
> > % strace -s2560 vglrun -d :1 -c proxy mate-session 2>&1 | grep -C1
> > 'RGB visual'
> > futex(0x7fe81b61a190, FUTEX_WAKE_PRIVATE, 2147483647) = 0
> > write(2, "[VGL] ERROR: in glXGetConfig--\n[VGL] 1124: Could not
> > obtain RGB visual on the server suitable for off-screen
> > rendering\n", 122[VGL] ERROR: in glXGetConfig--
> > [VGL] 1124: Could not obtain RGB visual on the server suitable
> > for off-screen rendering
> > ) = 122
> > -------------------
> > % strace -s2560 vglrun -d :1 -c proxy glxinfo 2>&1 | grep -C1 'RGB
> > visual'
> > futex(0x7f253e384190, FUTEX_WAKE_PRIVATE, 2147483647) = 0
> > write(2, "[VGL] ERROR: in glXGetConfig--\n[VGL] 1124: Could not
> > obtain RGB visual on the server suitable for off-screen
> > rendering\n", 122[VGL] ERROR: in glXGetConfig--
> > [VGL] 1124: Could not obtain RGB visual on the server suitable
> > for off-screen rendering
> > ) = 122
> > -------------------
> >
> > Thanks,
> > Jason
> >
> >
> ---------------------------------------------------------------------------
> > Jason Edgecombe | Linux Administrator
> > UNC Charlotte | The William States Lee College of Engineering
> > 9201 University City Blvd. | Charlotte, NC 28223-0001
> > Phone: 704-687-1943 <tel:704-687-1943>
> > [email protected] <mailto:[email protected]> | http://engr.uncc.edu
> > | Facebook
> >
> ---------------------------------------------------------------------------
> > If you are not the intended recipient of this transmission or a
> > person responsible for delivering it to the intended recipient, any
> > disclosure, copying, distribution, or other use of any of the
> > information in this transmission is strictly prohibited. If you have
> > received this transmission in error, please notify me immediately by
> > reply e-mail or by telephone at
> > 704-687-1943 <tel:704-687-1943>. Thank you.
> >
> >
> > On Tue, Mar 12, 2019 at 12:36 PM DRC <[email protected]
> > <mailto:[email protected]>> wrote:
> >
> > What are "some 3D apps"? Please be specific. Are these
> commercial
> > applications? Do they all use a particular framework/toolkit
> > like GTK,
> > Qt, JOGL, wxWidgets, etc.? Most likely, the applications in
> > question
> > are "visual hunting", which means they are iterating through
> > available
> > visuals and calling glXGetConfig() on each in an attempt to find
> a
> > visual with particular OpenGL rendering attributes. When an
> > application
> > does that rather than call glXChooseVisual() or
> > glXGetVisualFromFBConfig() to obtain an OpenGL-compatible visual,
> > VirtualGL has no idea which OpenGL rendering attributes the
> > application
> > wants. All VirtualGL can do in this case is map the unknown
> > visuals to
> > a default GLXFBConfig on the 3D X server. Referring to
> > https://github.com/VirtualGL/virtualgl/issues/96, applications
> > that do
> > this are most commonly looking for either an alpha channel or a
> > stencil
> > buffer, so try setting (in the environment)
> >
> > VGL_DEFAULTFBCONFIG=GLX_STENCIL_SIZE,8
> > or
> > VGL_DEFAULTFBCONFIG=GLX_ALPHA_SIZE,8
> > or
> > VGL_DEFAULTFBCONFIG=GLX_ALPHA_SIZE,8,GLX_STENCIL_SIZE,8
> >
> > I'm looking into better ways of addressing this problem in the
> long
> > term, since it seems to be popping up more frequently lately.
> > That's
> > why I'm asking you about toolkits. I'm wondering if some OSS
> widget
> > toolkit or other graphics framework has started visual hunting
> > for some
> > reason. Minimally, I need to document the scope of the problem
> > and a
> > workaround, so I need to know which applications (or, more
> > generally,
> > which toolkits) are affected.
> >
> > On 3/12/19 8:09 AM, Jason Edgecombe wrote:
> > > Hi everyone,
> > >
> > > I'm having trouble using VirtualGL with an Nvidia Tesla card.
> When
> > > trying to run some 3D apps, I get the error "Could not obtain
> > RGB visual
> > > on the server suitable for off-screen rendering"
> > >
> > > Here is an example:
> > > ----------------------------------------------------
> > > % vglrun -d :1 glxinfo | grep OpenGL
> > > OpenGL vendor string: NVIDIA Corporation
> > > OpenGL renderer string: Tesla V100-PCIE-32GB/PCIe/SSE2
> > > OpenGL core profile version string: 4.6.0 NVIDIA 418.39
> > > OpenGL core profile shading language version string: 4.60
> NVIDIA
> > > OpenGL core profile context flags: (none)
> > > OpenGL core profile profile mask: core profile
> > > OpenGL core profile extensions:
> > > OpenGL version string: 4.6.0 NVIDIA 418.39
> > > OpenGL shading language version string: 4.60 NVIDIA
> > > OpenGL context flags: (none)
> > > OpenGL profile mask: (none)
> > > OpenGL extensions:
> > > [VGL] ERROR: in glXGetConfig--
> > > [VGL] 1124: Could not obtain RGB visual on the server
> > suitable for
> > > off-screen rendering
> > > ----------------------------------------------------
> > >
> > > glxspheres runs as expected without errors.
> > >
> > > Does anyone have any insights into this problem?
> > >
> > > Thanks,
> > > Jason
> > >
> >
> ---------------------------------------------------------------------------
> > > Jason Edgecombe | Linux Administrator
> > > UNC Charlotte | The William States Lee College of Engineering
> > > 9201 University City Blvd. | Charlotte, NC 28223-0001
> > > Phone: 704-687-1943 <tel:704-687-1943>
> > > [email protected] <mailto:[email protected]>
> > <mailto:[email protected] <mailto:[email protected]>> |
> > http://engr.uncc.edu |
> > > Facebook
> > >
> >
> ---------------------------------------------------------------------------
> >
> > --
> > 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:virtualgl-users%[email protected]>.
> > To view this discussion on the web visit
> >
> https://groups.google.com/d/msgid/virtualgl-users/8a0be31c-b377-4131-8735-955c7132e9ab%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]
> > <mailto:[email protected]>.
> > To view this discussion on the web visit
> >
> https://groups.google.com/d/msgid/virtualgl-users/CAAR6MGChi90PUvjYpaC4aU4JEzZSoWQeG6O-rTuhPPECg%3DoABA%40mail.gmail.com
> > <
> https://groups.google.com/d/msgid/virtualgl-users/CAAR6MGChi90PUvjYpaC4aU4JEzZSoWQeG6O-rTuhPPECg%3DoABA%40mail.gmail.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/83113256-1ded-d7ca-2b56-77b7c9c5fa0c%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/CAAR6MGBVJP9wVFY0pu8ME4X6EALZBe-djCMnF2mUpCC6zfOpfw%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.
