I am using 390.xx on RHEL 6.9 and 7.5 with no issues, so unfortunately I
can't reproduce the problem.

On 7/26/18 4:18 AM, Jan-Albert van Ree wrote:
> Good morning,
> 
> To answer the first question : without VirtualGL everything was working
> fine, glxgears, glxinfo etc.
> 
> After some serious testing I've finally found a working combination.
> 
> RedHat/Scientific Linux glibc 2.17 build 222 appears to give compile
> problems with any nVidia driver older dan release 384 (I settled on 384.130)
> VirtualGL won't work with any nVidia driver newer than release 384 (at
> least on these systems...)
> So it appears the nVidia driver 390 introduced something which breaks
> the VirtualGL interface.
> 
> If the VirtualGL developers want more info or to have more scenario's
> tested to further debug the issue contact me, I'd be happy to help as
> far as I can.
> 
> Again thanks for the help!
> --
> Jan-Albert
> 
> Op woensdag 25 juli 2018 18:00:26 UTC+2 schreef DRC:
> 
>     Does the segfault happen without VirtualGL?  Can you run glxinfo on the
>     3D X server successfully without vglrun?
> 
>     Can you verify whether downgrading the nVidia driver fixes the problem?
> 
>     On 7/25/18 1:56 AM, Jan-Albert van Ree wrote:
>     > Hello,
>     >
>     > We've been running VirtualGL 2.5.2 on several HPC login nodes for a
>     > year, without any issues. Machines run Scientific Linux 7.3 , kernel
>     > 3.10.0-327.10.1
>     > Inside the systems is an nVidia Quadro K2200 with the 367.48
>     nVidia driver.
>     >
>     > Now we are deploying a new cluster with newer hardware thus we
>     have to
>     > use a newer kernel and newer nVidia drivers, and that's were all
>     falls
>     > apart...
>     > Kernel is now 3.10.0.862.6.3 , nVidia Quadro P4000 cards with the
>     nVidia
>     > 390.46 driver
>     >
>     > No matter what program we try to start with vglrun, we get a segfault
>     > such as
>     > [78457.806124] glxinfo[163647]: segfault at 2aaaafb0e000 ip
>     > 00002aaaab8b8521 sp 00007fffffffcc88 error 6 in
>     > libc-2.17.so <http://libc-2.17.so>[2aaaab75d000+1c3000]
>     >
>     > Different runs with different programs all produce a segfault,
>     were the
>     > ip address location is always the same, only the first address and
>     the
>     > sp location change, as well as the address in brackets at the end
>     of the
>     > line
>     >
>     > We tried upgrading to the latest beta build of VirtualGL but that did
>     > not solve the issue , it just slightly changes all the addresses
>     > [77279.757070] glxinfo[155176]: segfault at 2aaaafb23000 ip
>     > 00002aaaab8cd521 sp 00007fffffffc298 error 6 in
>     > libc-2.17.so <http://libc-2.17.so>[2aaaab772000+1c3000]
>     >
>     > Can anybody suggest a solution?
>     > Regards,
>     > --
>     > Jan-Albert van Ree
> 
> -- 
> 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/849991df-ce8b-4f68-9699-ea9898aab38f%40googlegroups.com
> <https://groups.google.com/d/msgid/virtualgl-users/849991df-ce8b-4f68-9699-ea9898aab38f%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/c43ba2a4-8af9-bfb9-e329-3961b7688219%40virtualgl.org.
For more options, visit https://groups.google.com/d/optout.

Reply via email to