Blackmagic Forum

Mon Jun 04, 2018 5:19 pm

I've found the Fusion tab to be very temperamental even under CentOS. Can't say I've done exactly what you describe but it sounds similar to my experience as well.

Sometimes I imagine it works better when I add nodes without having any other nodes selected so that they are not inserted into the chain automatically when they are added, and then connect them manually. Have you tried that?

Mon Jun 04, 2018 5:24 pm

Marc Gasser wrote:My question was if someone else have the same bug when doing whats described here: https://forum.blackmagicdesign.com/viewtopic.php?f=32&t=73490

on my setup i couldn't reproduce the crash.

Marc Gasser wrote:Next I will try the docker solution of Martin, hoping to get full GPU support.

just contact me, if you need any help!

Mon Jun 04, 2018 5:53 pm

@Daniel Tufvesson: Just tried what you suggested (adding nodes without anything selected) still same behavior. But good (not really good...) to hear it does not depend on the Linux distro.

@Martin: Just tested your docker-version. It seems to be the Resolve Free version, however, there is an issue with Sound Server and somehow not loading (opened an issue on gitlab).

Edit:
*******
I am using the Studio Version.
There is a slight feeling the I might test the same thing also on the Public Version to see if this crashes as well or not...

Mon Jun 04, 2018 7:03 pm

Marc Gasser wrote:@Martin:Thanks. Struggling with installing nvidia-docker2, but I might get this solved...

the installation instructions on the nvidia-docker2 webpage will not work for debian testing, because the used command to get the version number does not work in this case. but you can work around by simple typing:

Code: Select all: curl -s -L https://nvidia.github.io/nvidia-docker/gpgkey | sudo apt-key add - curl -s -L https://nvidia.github.io/nvidia-docker/debian9/nvidia-docker.list | sudo tee /etc/apt/sources.list.d/nvidia-docker.list sudo apt-get install nvidia-docker2 sudo pkill -SIGHUP dockerd

Thu Jun 07, 2018 9:35 pm

Hi there,

I've been several days struggling to run DaVinci Resolve 15 on a Linux PC.
Downloaded DaVinci Resolve 15 Public Beta 4
and run the .sh installer.

I first started with Ubuntu 16.04, then tried Ubuntu 18.04 LTS , and now I'm running on CentOS 7
After installing all libraries and dependencies, I always end getting exactly the same error:

Code: Select all: root$ /opt/resolve/bin/resolve /opt/resolve/bin/resolve: /opt/resolve/bin/../libs/libCg.so: no version information available (required by /opt/resolve/bin/resolve) /opt/resolve/bin/resolve: /opt/resolve/bin/../libs/libCgGL.so: no version information available (required by /opt/resolve/bin/resolve) ActCCMessage Already in Table: Code= 2282, Mode= 0, Level= 0, CmdKey= 8, Option= 0 ActCCMessage Already in Table: Code= c005, Mode= 13, Level= 1, CmdKey= -1, Option= 0 ActCCMessage Already in Table: Code= c006, Mode= 13, Level= 1, CmdKey= -1, Option= 0 ActCCMessage Already in Table: Code= c007, Mode= 13, Level= 1, CmdKey= -1, Option= 0 15.0b (#046) Linux/Clang Main thread starts: 612ECA80 [0x7f19612eca80] | Undefined | INFO | 2018-06-07 16:50:35,886 | -------------------------------------------------------------------------------- [0x7f19612eca80] | Undefined | INFO | 2018-06-07 16:50:35,886 | Loaded log config from /opt/resolve/configs/log-conf.xml [0x7f19612eca80] | Undefined | INFO | 2018-06-07 16:50:35,886 | -------------------------------------------------------------------------------- Segmentation fault (core dumped)

I've followed carefully the instructions at the first page of this topic (Ubuntu first , and then CentOS).
In all the cases made clean installations of the OS's , and also updated to the latest GPU drivers.

Also run as "ldd /opt/resolve/bin/resolve " but 'libCg.so' and 'libCgGL.so' are not missing.

also found many users with similar issues, but their solutions didn't help. Even read that "the libCG and libCgGL errors can be safely ignored."

Thanks in advance for any reply, because I can't figure out how to run it.
Maybe my GPU it's not supported by Resolve?

My system specs are:
Resolve version: 15.0b4
OS: Ubuntu 18.04 LTS & CentOS 7
CPU: Intel(R) Core(TM) i7-2670QM CPU @ 2.20GHz
RAM: 8GB
GPU: Radeon HD 6770M (1GB RAM)

Thu Jun 07, 2018 10:26 pm

That hardware is not adequate to run Resolve 15.

Fri Jun 08, 2018 8:56 pm

I've installed DaVinci Resolve 15.0b4 on Ubuntu 18.04 and I get this issue on launch:

ERROR LOG

Code: Select all: :/opt/resolve/bin$ sudo ./resolve ./resolve: /opt/resolve/bin/../libs/libCg.so: no version information available (required by ./resolve) ./resolve: /opt/resolve/bin/../libs/libCgGL.so: no version information available (required by ./resolve) ActCCMessage Already in Table: Code= 2282, Mode= 0, Level= 0, CmdKey= 8, Option= 0 ActCCMessage Already in Table: Code= c005, Mode= 13, Level= 1, CmdKey= -1, Option= 0 ActCCMessage Already in Table: Code= c006, Mode= 13, Level= 1, CmdKey= -1, Option= 0 ActCCMessage Already in Table: Code= c007, Mode= 13, Level= 1, CmdKey= -1, Option= 0 15.0b (#046) Linux/Clang Main thread starts: D4CAE200 [0x7f73d4cae200] | Undefined | INFO | 2018-06-08 22:28:35,994 | -------------------------------------------------------------------------------- [0x7f73d4cae200] | Undefined | INFO | 2018-06-08 22:28:35,994 | Loaded log config from /opt/resolve/configs/log-conf.xml [0x7f73d4cae200] | Undefined | INFO | 2018-06-08 22:28:35,994 | -------------------------------------------------------------------------------- Segmentation fault

libCG.so and libCgGL.so installed check

Code: Select all: :/opt/resolve/libs$ ls libCg libCgGL.so libCg.so

SYSTEM INFO

Code: Select all: $ neofetch .-/+oossssoo+/-. `:+ssssssssssssssssss+:` -+ssssssssssssssssssyyssss+- .ossssssssssssssssssdMMMNysssso. /ssssssssssshdmmNNmmyNMMMMhssssss/ +ssssssssshmydMMMMMMMNddddyssssssss+ aresminos@desktop /sssssssshNMMMyhhyyyyhmNMMMNhssssssss/ ----------------- .ssssssssdMMMNhsssssssssshNMMMdssssssss. OS: Ubuntu 18.04 LTS x86_64 +sssshhhyNMMNyssssssssssssyNMMMysssssss+ Kernel: 4.15.0-22-generic ossyNMMMNyMMhsssssssssssssshmmmhssssssso Uptime: 13 mins ossyNMMMNyMMhsssssssssssssshmmmhssssssso Packages: 2424 +sssshhhyNMMNyssssssssssssyNMMMysssssss+ Shell: bash 4.4.19 .ssssssssdMMMNhsssssssssshNMMMdssssssss. Resolution: 1680x1050 /sssssssshNMMMyhhyyyyhdNMMMNhssssssss/ DE: GNOME 3.28.1 +sssssssssdmydMMMMMMMMddddyssssssss+ WM: GNOME Shell /ssssssssssshdmNNNNmyNMMMMhssssss/ WM Theme: Adwaita .ossssssssssssssssssdMMMNysssso. Theme: Ambiance [GTK2/3] -+sssssssssssssssssyyyssss+- Icons: Ubuntu-mono-dark [GTK2/3] `:+ssssssssssssssssss+:` Terminal: tilix .-/+oossssoo+/-. CPU: Intel i5-2400 (4) @ 3.400GHz GPU: AMD Radeon RX Vega Memory: 2727MiB / 3928MiB

CPU INFO:

Code: Select all: $ cat /proc/cpuinfo processor : 0 vendor_id : GenuineIntel cpu family : 6 model : 42 model name : Intel(R) Core(TM) i5-2400 CPU @ 3.10GHz stepping : 7 microcode : 0x14 cpu MHz : 1596.228 cache size : 6144 KB physical id : 0 siblings : 4 core id : 0 cpu cores : 4 apicid : 0 initial apicid : 0 fpu : yes fpu_exception : yes cpuid level : 13 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx rdtscp lm constant_tsc arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 cx16 xtpr pdcm pcid sse4_1 sse4_2 x2apic popcnt tsc_deadline_timer aes xsave avx lahf_lm epb pti tpr_shadow vnmi flexpriority ept vpid xsaveopt dtherm ida arat pln pts bugs : cpu_meltdown spectre_v1 spectre_v2 spec_store_bypass bogomips : 6185.41 clflush size : 64 cache_alignment : 64 address sizes : 36 bits physical, 48 bits virtual power management: processor : 1 vendor_id : GenuineIntel cpu family : 6 model : 42 model name : Intel(R) Core(TM) i5-2400 CPU @ 3.10GHz stepping : 7 microcode : 0x14 cpu MHz : 1596.232 cache size : 6144 KB physical id : 0 siblings : 4 core id : 1 cpu cores : 4 apicid : 2 initial apicid : 2 fpu : yes fpu_exception : yes cpuid level : 13 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx rdtscp lm constant_tsc arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 cx16 xtpr pdcm pcid sse4_1 sse4_2 x2apic popcnt tsc_deadline_timer aes xsave avx lahf_lm epb pti tpr_shadow vnmi flexpriority ept vpid xsaveopt dtherm ida arat pln pts bugs : cpu_meltdown spectre_v1 spectre_v2 spec_store_bypass bogomips : 6185.41 clflush size : 64 cache_alignment : 64 address sizes : 36 bits physical, 48 bits virtual power management: processor : 2 vendor_id : GenuineIntel cpu family : 6 model : 42 model name : Intel(R) Core(TM) i5-2400 CPU @ 3.10GHz stepping : 7 microcode : 0x14 cpu MHz : 1596.237 cache size : 6144 KB physical id : 0 siblings : 4 core id : 2 cpu cores : 4 apicid : 4 initial apicid : 4 fpu : yes fpu_exception : yes cpuid level : 13 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx rdtscp lm constant_tsc arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 cx16 xtpr pdcm pcid sse4_1 sse4_2 x2apic popcnt tsc_deadline_timer aes xsave avx lahf_lm epb pti tpr_shadow vnmi flexpriority ept vpid xsaveopt dtherm ida arat pln pts bugs : cpu_meltdown spectre_v1 spectre_v2 spec_store_bypass bogomips : 6185.41 clflush size : 64 cache_alignment : 64 address sizes : 36 bits physical, 48 bits virtual power management: processor : 3 vendor_id : GenuineIntel cpu family : 6 model : 42 model name : Intel(R) Core(TM) i5-2400 CPU @ 3.10GHz stepping : 7 microcode : 0x14 cpu MHz : 1596.245 cache size : 6144 KB physical id : 0 siblings : 4 core id : 3 cpu cores : 4 apicid : 6 initial apicid : 6 fpu : yes fpu_exception : yes cpuid level : 13 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx rdtscp lm constant_tsc arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 cx16 xtpr pdcm pcid sse4_1 sse4_2 x2apic popcnt tsc_deadline_timer aes xsave avx lahf_lm epb pti tpr_shadow vnmi flexpriority ept vpid xsaveopt dtherm ida arat pln pts bugs : cpu_meltdown spectre_v1 spectre_v2 spec_store_bypass bogomips : 6185.41 clflush size : 64 cache_alignment : 64 address sizes : 36 bits physical, 48 bits virtual power management:

GPU INFO:

Code: Select all: $ sudo lshw -C display [sudo] password for aresminos: *-display description: VGA compatible controller product: Vega [Radeon RX Vega] vendor: Advanced Micro Devices, Inc. [AMD/ATI] physical id: 0 bus info: pci@0000:03:00.0 version: c0 width: 64 bits clock: 33MHz capabilities: pm pciexpress msi vga_controller bus_master cap_list rom configuration: driver=amdgpu latency=0 resources: irq:29 memory:c0000000-cfffffff memory:d0000000-d01fffff ioport:e000(size=256) memory:fe400000-fe47ffff memory:c0000-dffff

OpenGL VERSION

Code: Select all: $ glxinfo | grep "OpenGL version" OpenGL version string: 3.0 Mesa 18.0.0-rc5

What needs to be done? I need it to work on Ubuntu, not CentOS but I think the same issue would come up on CentOS.
Do I need to provide any more information or any other logs? If I do, please list the commands for the things you want me to check.

version 15.0b2 had this same issue but would show the splash screen and some initial setup but wouldn't bring up the editor and after that setup it wouldn't eve show the splash. 15.0b.4 doesn't show anything on the screen, I just get the logs when I run it trough cli.

Any help with this issue would be greatly appreciated!
Thanks!

Fri Jun 08, 2018 11:00 pm

Those libCg.so and libCgGL.so warnings plainly refer to unavailable version information, which basically means Resolve finds them and is puzzled. This is something that has cropped up in the 15 betas, and will hopefully be fixed by RC. But I doubt they are the genuine culprit.

I'd say the crashes are GPU-related. 14 often showed exact same behaviour, by allowing the user up to the Project Manager and crashing, with the unhelpful "ActCCMessage..." in the logs. The inherent problem was that it distracted people from the real issues, as I think those libs are doing now.

On 14, the issue generally went away by updating the graphics drivers, provided the GPU was sufficient to begin with. If both are good, we take a step further. AMD or Nvidia? A startup with CUDA requires OpenCL to be installed just as well, otherwise Resolve will abort the sequence. Really hope BMD would fix this nagging issue by now.

You have a Vega, so that's curious.

HealthWyzeMedia wrote:I am running Ubuntu 16.04, and I ran into the same problem. I fixed it and was able to run Davinci 15 by installing nvidia-modprobe (apt-get install nvidia-modprobe).

While the poster unfortunately failed to provide the used GPU, it's an interesting proposal. On it's own, nvidia-modprobe will load the Nvidia kernel module and exit. Which would be similar to OpenCL keeping up appearances in a Resolve/CUDA config.

Of course, you'd need to test it and tell us if it makes a difference

Fri Jun 08, 2018 11:52 pm

Kristijan definitely faces some segmentation fault! that doesn't have much in common with misleading warnings and strange behavior. it's a much more serious issue. and it looks very unlikely, that anybody will be able to analyze the produced core dump without access to the source code or an unstripped debug version of the application.

Sat Jun 09, 2018 3:54 pm

@Kristijan

I have Resolve 15b4 studio running OK (minor problems below) with Ubuntu Studio 17.10 on two machines
One 2xCPU Xeon, one AMD Threadripper (have tried BMD Centos on both earlier with some problems, also a later 'clean' Centos with more problems...)
Both machines with Nvidia 1080 ti

I had problems with Ubuntu Studio v16 and v18, complains in Resolve about missing Cuda, but OpenGL worked as much as I tried functionwise
From advise in this thread I downloaded the Cuda developer package from Nvidia, they didn't have Ubuntu v18 support
Tried the Nvidia v17 support on Ubuntu Studio 16/18 - lots of problem
But with clean install Ubuntu Studio 17.10, Nvidia developer package for Ubuntu 17, it works!
Also added their repo as instructed, now update brings newer Nvidia things down too

I let the Ubuntu Studio do a net update after install (did not allow it to fetch 3rd party things which might have caused problems in earlier tests)

Also installed BMD Desktop video 10.10 on the AMD (which has a BMD Decklink Mini Monitor 4k pci card), it complained about libpango, ran apt-get on that and tried again, works now!

When things seemed OK I also installed (apt-get) VLC and ffmpeg
---
Days later I have allowed Ubuntu to update again - after it finished it suggests me to go up to v18 which I refuse right now

Resolve studio works fine, video/sound out through SDi on AMD, sound to a headphone on Xeon
---
Remaining issues
- The AMD included a pci 10 gbit ethernet card, doesn't work now, didn't work with Centos but it worked on Ubuntu Studio v18, no rush to fix it but performance will be better (not a BMD issue)

- None of the Linux machines work with the Micro grading panel, but it works fine on an iMac 5k with v15b4 studio - not sure how to search for the problem - tips?
Both Linux accept the dongle, the AMD was even installed via extern DVD to USB so I know USB is active on both and I've tested both USB 2/3 ports

- The AMD with Decklink Mini Monitor 4k pci card doesn't output sound with the video via HDMI, but via SDI it does
While testing with a BMD Video Assist 7" which has both SDI and HDMI in/out I can hear and see sound level from SDI
Just by intuition (or whatever

I tried the HDMI out to TV in the Video Assist when it received via SDI - and it outputs sound on HDMI too!!!
Seems to be a bug by BMD, the card obviously receives sound, why not output sound on both...

---
Positive things:
- Stabilisation flies on 4k from drone and Sony RX10, same things on the iMac is so slow...
- Playing things graded on both Linux just plays, iMac quick to stutter with just few nodes
All three machines have SSD, right now resolve just use the system SSD on all three
---
Kristijan

I have a spare disk ready, easy for me to repeat/confirm a clean installation
Please make a try as above, tell me if you get stuck/how/where and I can do a repeat install on one of the machines and take more detailed notes

Good luck
Björn

Sat Jun 09, 2018 4:55 pm

Sulo, Martin, Bjorn I don't think installing any Nvidia specific software will help since I run AMD Radeon RX Vega 64 Liquid 8GB gpu. It might get Resolve to show up in the gui but won't work correctly since the workaround would just be a hack to fool Resolve into thinking it has some deps which won't actually do anything?

I forgot to mention explicitly that I am running the open source amdgpu driver from the kernel.
The support for Ubuntu 18.04 is coming in the amdgpu-pro 18.20 but that driver is still in early preview.
I don't thing it supports OpenCL but it will come with ROCM which is a thing worth exploring for running Resolve with Vega.

I will try again when amdgpu-pro 18.20 is out.

Also I was using a script I found at:

Code: Select all: http://grsk.co/resolveubuntu

to make a Resolve Debian package before installing it.

I don't know who the maintainer of that script is but I was thinking if it would be possible to create a DaVinci Resolve Snap package. That way it would basically be a sandboxed application with all dependencies and paths included in the snap package so it wouldn't conflict or complain on any distro.

The thing is, I'm not really that technical and that's why I'm not doing it already. I basically know how to package a node snap and that's it.
For anyone else willing to explore this route here is the link to snap package and it's tool snapcraft's page

Code: Select all: https://snapcraft.io/

Some good tutorials on snap and snapcraft can be found at:

Code: Select all: https://tutorials.ubuntu.com/

and that's how far I've got atm on exploring this solution but I think it would be very promising.

Sorry for the urls in /code, I'm not allowed to post urls idk why...

Sat Jun 09, 2018 10:32 pm

bjornzitting wrote:None of the Linux machines work with the Micro grading panel, but it works fine on an iMac 5k with v15b4 studio - not sure how to search for the problem - tips?

The panel expects a certain folder, which is alright with CentOS, but breaks the multiarch on distros like Ubuntu.

See here.

That's another fun thing BMD should seriously fix.

Sat Jun 09, 2018 11:27 pm

Sulo, Martin, Bjorn I don't think installing any Nvidia specific software will help since I run AMD Radeon RX Vega 64 Liquid 8GB gpu. It might get Resolve to show up in the gui but won't work correctly since the workaround would just be a hack to fool Resolve into thinking it has some deps which won't actually do anything?

I forgot to mention explicitly that I am running the open source amdgpu driver from the kernel.
The support for Ubuntu 18.04 is coming in the amdgpu-pro 18.20 but that driver is still in early preview.
I don't thing it supports OpenCL but it will come with ROCM which is a thing worth exploring for running Resolve with Vega.

I will try again when amdgpu-pro 18.20 is out.

Also I was using a script I found at:

Code: Select all: http://grsk.co/resolveubuntu

to make a Resolve Debian package before installing it.

I don't know who the maintainer of that script is but I was thinking if it would be possible to create a DaVinci Resolve Snap package. That way it would basically be a sandboxed application with all dependencies and paths included in the snap package so it wouldn't conflict or complain on any distro.

The thing is, I'm not really that technical and that's why I'm not doing it already. I basically know how to package a node snap and that's it.
For anyone else willing to explore this route here is the link to snap package and it's tool snapcraft's page

Code: Select all: https://snapcraft.io/

Some good tutorials on snap and snapcraft can be found at:

Code: Select all: https://tutorials.ubuntu.com/

and that's how far I've got atm on exploring this solution but I think it would be very promising.

Mon Jun 11, 2018 7:43 am

KristijanZic, FYI your post was moderated like every new user.

Mon Jun 11, 2018 8:36 am

KristijanZic, I'm afraid you still need to have CUDA installed for Resolve to run properly even if you are running OpenCL and vice versa.

Also, that script you linked to looks like a modified version of my makeresolvedeb. Please use the latest makeresolvedeb from this thread (links to related posts are on the first post) or from my website http://www.danieltufvesson.com/makeresolvedeb

Mon Jun 11, 2018 3:25 pm

Peter, I hope that's not the case with everyone. I have projects to complete, what good is the community support if I can't get it immediately while I'm working on the issue. 3 days later is unacceptable.
This delay greatly impacts the UX of this forum in the negative way. I would hope for BM to at least open a Discourse forum for Linux users or even better for all users. Discourse has great anti spam capabilities like "Built in Akismet spam protection and heuristics including new user sandboxing, user flag blocking, and standard nofollow." so for me it would be a no-brainer decision. So you could focus on Linux compatibility rather than forum maintenance.

Please take a look at:

Code: Select all: https://www.discourse.org/

The absolute best thing would be to accept bug reports on an empty GitLab repository or something like that. This one forum thread is not good for the purpose it's being used for.

Daniel, thank you so much for the script. Could you please open a GitLab repository and host it there? That way you could accept patches and pull requests from other people who may be willing to help. Plus every paid GitLab feature is free for open source repositories unlike GitHub.

Code: Select all: https://about.gitlab.com/

Also, a good idea would be if you could just replace all script text version numbers and DaVinci Resolve text version numbers with two javascript variables, that way instead of changing the string on the download button on your website, you could just change those two variables and every version number on that page would be updated.

I mean here:

Code: Select all: http://www.danieltufvesson.com/makeresolvedeb

Mon Jun 11, 2018 6:57 pm

KristijanZic wrote:Peter, I hope that's not the case with everyone. I have projects to complete, what good is the community support if I can't get it immediately while I'm working on the issue. 3 days later is unacceptable.

Every new user’s first post is moderated. This is pretty standard across phpBB forums. If you are looking for responsive feedback, keep in mind your post was on a Friday and BMD is an Australian company. I don’t know your time zone but your post might have ended up on the weekend for them.

Cheers,
Mike

Tue Jun 12, 2018 10:27 am

Hi everyone,

I'm running into issues here on my Manjaro install (based on Arch Linux).

I made the install using a package from the Arch User Repository (script can be found here: https://aur.archlinux.org/cgit/aur.git/ ... solve-beta ).
I've also installed opencl-mesa for my graphic card to be recognised.

Alas, Resolve still doesn't start. Well, to be precise, it starts, but no GUI appears (but I can see the process running in background).

The displayed output in console is as follows:

Code: Select all: bin/resolve: /opt/resolve/bin/../libs/libCg.so: no version information available (required by bin/resolve) bin/resolve: /opt/resolve/bin/../libs/libCgGL.so: no version information available (required by bin/resolve) ActCCMessage Already in Table: Code= 2282, Mode= 0, Level= 0, CmdKey= 8, Option= 0 ActCCMessage Already in Table: Code= c005, Mode= 13, Level= 1, CmdKey= -1, Option= 0 ActCCMessage Already in Table: Code= c006, Mode= 13, Level= 1, CmdKey= -1, Option= 0 ActCCMessage Already in Table: Code= c007, Mode= 13, Level= 1, CmdKey= -1, Option= 0 15.0b (#046) Linux/Clang Main thread starts: D038CA40 log4cxx: No appender could be found for logger (Undefined). log4cxx: Please initialize the log4cxx system properly.

CPU use is null, as if resolve is waiting for some external event.

The rolling log captures this information:

Code: Select all: [0x7f62d038ca40] | Main | INFO | 2018-06-12 12:10:55,106 | Updating display GPU information...

I did a

Code: Select all: ldd /opt/resolve/bin/resolve

but nothing was missing. I also did an strace, I can join the output here if needed, but I saw nothing special in it.

Any ideas on where to look?

I joined the ouput of various information sources on my machine.

neofetch:

Code: Select all: ██████████████████ ████████ rbazile@pc-ub7-2016 ██████████████████ ████████ ------------------- ██████████████████ ████████ OS: Manjaro Linux x86_64 ██████████████████ ████████ Host: Precision 7510 ████████ ████████ Kernel: 4.16.14-2-MANJARO ████████ ████████ ████████ Uptime: 2 hours, 17 mins ████████ ████████ ████████ Packages: 1482 ████████ ████████ ████████ Shell: bash 4.4.19 ████████ ████████ ████████ Resolution: 1920x1080, 1920x1200 ████████ ████████ ████████ DE: KDE ████████ ████████ ████████ WM: KWin ████████ ████████ ████████ Theme: Breeze Dark [KDE], Adwaita [GTK2], Breeze-Dark [GTK3] ████████ ████████ ████████ Icons: Breeze-dark [KDE], Adwaita [GTK2], Breeze-dark [GTK3] ████████ ████████ ████████ Terminal: yakuake CPU: Intel i7-6820HQ (8) @ 3.600GHz GPU: AMD Radeon HD 8890M GPU: Intel HD Graphics 530 Memory: 8274MiB / 15925MiB

glxinfo -B:

Code: Select all: name of display: :0 display: :0 screen: 0 direct rendering: Yes Extended renderer info (GLX_MESA_query_renderer): Vendor: Intel Open Source Technology Center (0x8086) Device: Mesa DRI Intel(R) HD Graphics 530 (Skylake GT2) (0x191b) Version: 18.1.1 Accelerated: yes Video memory: 3072MB Unified memory: yes Preferred profile: core (0x1) Max core profile version: 4.5 Max compat profile version: 3.0 Max GLES1 profile version: 1.1 Max GLES[23] profile version: 3.2 OpenGL vendor string: Intel Open Source Technology Center OpenGL renderer string: Mesa DRI Intel(R) HD Graphics 530 (Skylake GT2) OpenGL core profile version string: 4.5 (Core Profile) Mesa 18.1.1 OpenGL core profile shading language version string: 4.50 OpenGL core profile context flags: (none) OpenGL core profile profile mask: core profile OpenGL version string: 3.0 Mesa 18.1.1 OpenGL shading language version string: 1.30 OpenGL context flags: (none) OpenGL ES profile version string: OpenGL ES 3.2 Mesa 18.1.1 OpenGL ES profile shading language version string: OpenGL ES GLSL ES 3.20

cat /proc/cpuinfo :

Code: Select all: processor : 0 vendor_id : GenuineIntel cpu family : 6 model : 94 model name : Intel(R) Core(TM) i7-6820HQ CPU @ 2.70GHz stepping : 3 microcode : 0xc2 cpu MHz : 900.024 cache size : 8192 KB physical id : 0 siblings : 8 core id : 0 cpu cores : 4 apicid : 0 initial apicid : 0 fpu : yes fpu_exception : yes cpuid level : 22 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc art arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf tsc_known_freq pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm 3dnowprefetch cpuid_fault epb invpcid_single pti ibrs ibpb stibp tpr_shadow vnmi flexpriority ept vpid fsgsbase tsc_adjust bmi1 hle avx2 smep bmi2 erms invpcid rtm mpx rdseed adx smap clflushopt intel_pt xsaveopt xsavec xgetbv1 xsaves dtherm ida arat pln pts hwp hwp_notify hwp_act_window hwp_epp bugs : cpu_meltdown spectre_v1 spectre_v2 spec_store_bypass bogomips : 5426.00 clflush size : 64 cache_alignment : 64 address sizes : 39 bits physical, 48 bits virtual power management: [...] x 8

clinfo:

Code: Select all: Number of platforms 1 Platform Name Clover Platform Vendor Mesa Platform Version OpenCL 1.1 Mesa 18.1.1 Platform Profile FULL_PROFILE Platform Extensions cl_khr_icd Platform Extensions function suffix MESA Platform Name Clover Number of devices 1 Device Name AMD CAPE VERDE (DRM 2.50.0, 4.16.14-2-MANJARO, LLVM 6.0.0) Device Vendor AMD Device Vendor ID 0x1002 Device Version OpenCL 1.1 Mesa 18.1.1 Driver Version 18.1.1 Device OpenCL C Version OpenCL C 1.1 Device Type GPU Device Profile FULL_PROFILE Device Available Yes Compiler Available Yes Max compute units 10 Max clock frequency 925MHz Max work item dimensions 3 Max work item sizes 256x256x256 Max work group size 256 Preferred work group size multiple 64 Preferred / native vector sizes char 16 / 16 short 8 / 8 int 4 / 4 long 2 / 2 half 8 / 8 (cl_khr_fp16) float 4 / 4 double 2 / 2 (cl_khr_fp64) Half-precision Floating-point support (cl_khr_fp16) Denormals No Infinity and NANs Yes Round to nearest Yes Round to zero No Round to infinity No IEEE754-2008 fused multiply-add No Support is emulated in software No Single-precision Floating-point support (core) Denormals No Infinity and NANs Yes Round to nearest Yes Round to zero No Round to infinity No IEEE754-2008 fused multiply-add No Support is emulated in software No Correctly-rounded divide and sqrt operations No Double-precision Floating-point support (cl_khr_fp64) Denormals Yes Infinity and NANs Yes Round to nearest Yes Round to zero Yes Round to infinity Yes IEEE754-2008 fused multiply-add Yes Support is emulated in software No Address bits 64, Little-Endian Global memory size 2147483648 (2GiB) Error Correction support No Max memory allocation 1503238553 (1.4GiB) Unified memory for Host and Device No Minimum alignment for any data type 128 bytes Alignment of base address 32768 bits (4096 bytes) Global Memory cache type None Image support No Local memory type Local Local memory size 32768 (32KiB) Max number of constant args 16 Max constant buffer size 1503238553 (1.4GiB) Max size of kernel argument 1024 Queue properties Out-of-order execution No Profiling Yes Profiling timer resolution 0ns Execution capabilities Run OpenCL kernels Yes Run native kernels No Device Extensions cl_khr_byte_addressable_store cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_local_int32_base_atomics cl_khr_local_int32_extended_atomics cl_khr_int64_base_atomics cl_khr_int64_extended_atomics cl_khr_fp64 cl_khr_fp16 NULL platform behavior clGetPlatformInfo(NULL, CL_PLATFORM_NAME, ...) Clover clGetDeviceIDs(NULL, CL_DEVICE_TYPE_ALL, ...) Success [MESA] clCreateContext(NULL, ...) [default] Success [MESA] clCreateContextFromType(NULL, CL_DEVICE_TYPE_DEFAULT) Success (1) Platform Name Clover Device Name AMD CAPE VERDE (DRM 2.50.0, 4.16.14-2-MANJARO, LLVM 6.0.0) clCreateContextFromType(NULL, CL_DEVICE_TYPE_CPU) No devices found in platform clCreateContextFromType(NULL, CL_DEVICE_TYPE_GPU) Success (1) Platform Name Clover Device Name AMD CAPE VERDE (DRM 2.50.0, 4.16.14-2-MANJARO, LLVM 6.0.0) clCreateContextFromType(NULL, CL_DEVICE_TYPE_ACCELERATOR) No devices found in platform clCreateContextFromType(NULL, CL_DEVICE_TYPE_CUSTOM) No devices found in platform clCreateContextFromType(NULL, CL_DEVICE_TYPE_ALL) Success (1) Platform Name Clover Device Name AMD CAPE VERDE (DRM 2.50.0, 4.16.14-2-MANJARO, LLVM 6.0.0) ICD loader properties ICD loader Name OpenCL ICD Loader ICD loader Vendor OCL Icd free software ICD loader Version 2.2.12 ICD loader Profile OpenCL 2.2

Tue Jun 12, 2018 8:26 pm

Daniel,
I installed Resolve Studio 15.0b4 on Ubuntu 18.04 using your deb package. No errors on installation and Resolve loads fine and plays mp4 and Avchd mts video with no problems. But no audio output at all. Transcoded clips play sound fine.

Any suggestions for me?

Wed Jun 13, 2018 1:50 am

What codec are those original formats using for audio? I recommend double checking your material against the Codec guide to see if you’re using unsupported material.

Cheers,
Mike

Wed Jun 13, 2018 9:03 am

A bit more information.

I realised that when I removed the opencl-mesa package from my system (and keep compute-runtime, Intel opencl driver for its gpu) I reach the splashscreen, but then I have an error message that says no supported graphic card was found.

I've got this error message in ResolveDebug.txt:

Code: Select all: [0x7fbe377fe700] | GPUManager | INFO | 2018-06-13 10:58:55,770 | Let There Be OpenCL Light! [0x7fbe377fe700] | GPUManager | INFO | 2018-06-13 10:58:55,770 | Start detecting OpenCL boards [0x7fbe377fe700] | GPUManager | INFO | 2018-06-13 10:58:55,770 | Skipping unqualified OpenCL platform: Intel(R) OpenCL HD Graphics [0x7fbe377fe700] | GPUManager | INFO | 2018-06-13 10:58:55,770 | 0 OpenCL boards detected Urgent message: CRITICAL_PREF: DaVinci Resolve could not find any OpenCL capable GPUs. Would you like to review and update your DaVinci Resolve Hardware Configuration? ||Please ensure that appropriate GPU drivers are installed and your DaVinci Resolve hardware is configured appropriately. [0x7fbe377fe700] | GPUManager | ERROR | 2018-06-13 10:58:55,770 | CRITICAL_PREF: DaVinci Resolve could not find any OpenCL capable GPUs. Would you like to review and update your DaVinci Resolve Hardware Configuration? ||Please ensure that appropriate GPU drivers are installed and your DaVinci Resolve hardware is configured appropriately. [0x7fbe377fe700] | IP | ERROR | 2018-06-13 10:58:55,770 | Failed to initialize OpenCL boards

Looks like my Intel GPU is not supported.
What I don't understand is why when I have the right opencl driver for my second GPU (Radeon) it stops at the "Updating display GPU information..." step.

I have PRIME installed and running properly on my machine (I can start programs to use the main GPU instead of Intel's), and this problem happens also when I force Resolve to start on the main GPU (with DRI_PRIME=1 environment variable).

I'm going to try more combination (like removing compute-engine and keeping only opencl-mesa) but I'm a bit at loss I have to admit.

Any pointers?

Wed Jun 13, 2018 1:32 pm

MikeRochefort wrote:What codec are those original formats using for audio? I recommend double checking your material against the Codec guide to see if you’re using unsupported material.

Thanks Mike, just shows how much I need to learn.

My first look at the info Resolve gives about those two clips is:
Avchd mts is AC3
mp4 is ACC

I see the list of supported codecs on BM support page does not include AC3.
ACC is listed but not for Linux.

I take it from this then that its not a problem with my installation. These codecs are just not supported and the only solution is to transcode. is this likely to change in future 15 beta releases?

Wed Jun 13, 2018 7:57 pm

Neil Preston wrote:I take it from this then that its not a problem with my installation. These codecs are just not supported and the only solution is to transcode. is this likely to change in future 15 beta releases?

Glad I could point you in the right direction. I and others do hope that it’s something that does change in the future, it is really annoying to have to do far more transcoding than our Win/macOS counterparts. But if it’s in the cards BMD employees are keeping their mouths sealed in a nuclear bunker several miles under the ice caps.

Cheers,
Mike

Sat Jun 16, 2018 8:31 am

While I realize that Resolve was intended for high level use and not for consumer use, now that it does support mp4 and mts video it seems obvious that audio codecs must follow for Linux systems.

I have thousands of mp4 and mts clips filmed over long periods that I would like to edit, but it’s seems unnecessary to spend the time transcoding them if the capability will soon be available in Linux. To ask for some indication of when this might happen is understandaable. I know I can do this in Windows, but I have waited for years for the right Linux based software. I have bought an unlock key for Resolve Studio and I now want to get away from Windows.

The recent price reduction for Resolve Studio has made Resolve available to a much wide range of users including amateurs and hobbyists like myself who want to raise their level of work.

Sat Jun 16, 2018 9:50 am

The latest version of MakeResolveDeb:

makeresolvedeb_15.0b5-1.sh.tar.gz: (2.32 KiB) Downloaded 1474 times

Or download from here for the latest updates and more info:
http://www.danieltufvesson.com/makeresolvedeb

I agree that MakeResolveDeb could be made more universal and, at least in theory, be made to support both current and future versions of Resolve.

The tricky thing with that approach is test and verification. BMD may at any point change things in Resolve or the installer that can cause the conversion to fail in one way of the other.

What I offer to the community is that I stand behind my product and I maintain it for free for everyone to use. Whenever there is a new release of Resolve I manually go through new installer, track the eventual changes and test *.deb creation and installation of Resolve and Resolve Studio on several Debian based distributions. All my versions of MakeResolveDeb are verified working on my test setups before they are released. It may take a little bit of extra time for me to do this however.

I get questions daily about Resolve and Debian, and it's taking up more and more of my spare time. By limiting every release to a specific Resolve version life gets a little bit easier for me and it ends up being more reliable for the end user as well.

My focus has been to provide something that just works for now with the hopes that BMD will address this situation in the future. Hope you understand.

Sat Jun 16, 2018 8:40 pm

Thanks Daniel for your 15.ob5 deb.

It installed Resolve 15.ob5 beta with no problems and it is looking good.

I hope that my last post did not give the impression that I was implying that you should fix the AAC audio problems in your deb packages. What I was trying to say was that we would like BMD to give us an estimate of when they expect the audio problems to be rectified, i.e, a sort of roadmap.

I understand that you cannot alter their software, but that your debs are taking care of the technicalities of ensuring that each new beta of Resolve installs correctly on Debian and Ubuntu based distributions, for those of us who don’t have the necessary knowledge to do it.

Sat Jun 16, 2018 9:20 pm

Daniel Tufvesson wrote:The tricky thing with that approach is test and verification. BMD may at any point change things in Resolve or the installer that can cause the conversion to fail in one way of the other.

yes -- that's indeed a very important point!

i was also puzzling over this issue, how i could test my resolve containers automatically by typical CI/CD means?
until now i couldn't figure out a satisfaying solution. it's very hard to provide the required hardware preconditions to successfully start resolve in this kind of typical DevOps environments somewhere in the cloud and examine at least a few elementary test automatically. but even if we could figure out a minimalist setup, which would work, it hardly could be taken as a representative inspection, because most of the real world troubles are at least partially caused by the interplay with various graphic cards and their proprietary drivers and can not be realistically tested in this kind of synthetic envrionments...

Sun Jun 17, 2018 10:07 pm

Do not use Ubuntu 18.04 or derivates, since there are no proprietary Nvidia drivers available (which are essential).
Use 17.10 or 16.04 LTS.

Sun Jun 17, 2018 10:15 pm

Does installing the driver yourself not work? I never use repository modules, always do it by hand. It’s a bit easier to manage in CentOS land as the kernel doesn’t change often and DKMS saves things from breaking completely. I realize not everyone will agree with this strategy, however.

Cheers,
Mike

Mon Jun 18, 2018 2:32 pm

18.04 works for me, but I installed the nvidia drivers manually.

Mon Jul 02, 2018 1:52 pm

With DaVinci Resolve 15.0b5-1 on Ubuntu 18.04 I still get this error:

Code: Select all: /opt/resolve/bin$ ./resolve ./resolve: /opt/resolve/bin/../libs/libCg.so: no version information available (required by ./resolve) ./resolve: /opt/resolve/bin/../libs/libCgGL.so: no version information available (required by ./resolve) ActCCMessage Already in Table: Code= 2282, Mode= 0, Level= 0, CmdKey= 8, Option= 0 ActCCMessage Already in Table: Code= c005, Mode= 13, Level= 1, CmdKey= -1, Option= 0 ActCCMessage Already in Table: Code= c006, Mode= 13, Level= 1, CmdKey= -1, Option= 0 ActCCMessage Already in Table: Code= c007, Mode= 13, Level= 1, CmdKey= -1, Option= 0 15.0b (#057) Linux/Clang Main thread starts: 30E1F500 [0x7f6430e1f500] | Undefined | INFO | 2018-07-02 15:47:59,337 | -------------------------------------------------------------------------------- [0x7f6430e1f500] | Undefined | INFO | 2018-07-02 15:47:59,338 | Loaded log config from /opt/resolve/configs/log-conf.xml [0x7f6430e1f500] | Undefined | INFO | 2018-07-02 15:47:59,338 | -------------------------------------------------------------------------------- Segmentation fault (core dumped)

Wed Jul 04, 2018 12:42 pm

Those error messages are red herrings, at best.

What other logs you got?

Fri Jul 06, 2018 9:29 am

CONFIRMING DaVinci Resolve Studio v15b6 (15.0.0B.065 - 1GPU) installs easily and works on Linux Mint 19 "Tara" (released 29Jun2018).

Linux Mint 19 core is similar enough to Ubuntu 18.x except no/or little data collection

SYSTEM:
10 yr old Asus P7P55D MB, 4 core i5 @3.8GHz, 6GB RAM (yeap - not great), AMD RX550 4GB (LexaPro) GPU card running AMD-GPU-PRO-18.20 Radeon Software for Linux, BMD PCIe Intensity Pro (HDMI to playback monitor with audio c/ HDMI). BMD DR dongle.

OPERATION:
DaVinci starts up nicely, shows 1xGPU as AMD RX550. I can edit UHD, grade, playback..shows Fusion tab but didn't want to push it too much till I add some more memory. Actually rocks along much better than expected

Even the AMD RX 550 OpenCL seems very fast and so far quite stable. A surprise was that DR spans both DVI & HDMI GUI monitors perfectly without any setup or effort.

ISSUES:
* playback audio sometimes dropping out over Intensity HDMI audio output.
* h.265/HEVC UHD source footage is "very slow" to playback compared with h.264 (realtime) (probably since h.265/HEVC decoding is not GPU accelerated compared to my iMac where the i7/MacOS does super fast hardware h.265 (macOS->Intel iGPU) decode/encode. I expected DR would do the HEVC hardware decode via the system GPU(s). Checking with Linux VLC, they use the VDPAU (AMD/nVidia/Intel) API seems available for HEVC hardware decoding (and appears encoding) but Linux DR15b6 does not appear to use it - BMB DR dev team - possible Linux version enhancement

* DaVinci crashed with a log about Fusion server disconnected when I set a non-typical or >SD Playback resolution and frame rate for Intensity. I did install the v10.11 DEB package latest BMD Desktop Video suite. I'm guessing system/GPU memory shortage to do so much ? I had to remove the Intensity card, restart Resolve and reduce & save the playback resolution playback setting to NTSC, than re-installl the Intensity card and we seemed good.
* DR pops up a Report Problem box every time I quit DR app normally, but nothing failed ?
* Both DR v15b5 or b6 can not send out the email for Report Problem popup box. Firefox on the same box works. The onscreen error message says: Unable to connect. Please check your network connectivity and try again. Hmmm ?

INSTALLATION:
Install was easy - need the libssl dependancy & sym linking, note: target dir: is /usr/lib on Linux Mint. Also installed the xserver-xorg-hwe as discussed in many places. Months ago I did tried to install DaVinci Resolve v14_free on Linux Mint 18 and gave up - too many OS/GPU bits not working, even the RX550 was a mess, and finally no Open CL. Besides even working, I expect DR on Linux on a modern system would really rock for little $. Great job BMD team !

AMD GPU Caution Note:
Sadly I learnt the hard way that my RX550 is not supported by macOS High Sierra due to the possibility that the card can have one of either chipset that is not easily recognised from the retail packaging - one RX550 uses the Lexa chip (PCI DeviceID: 0x699F) (which isn't supported in macOS, but is in Windows and Linux, and the other RX550 uses the Baffin chip (PCI DeviceID: 0x67FF) which is supported in macOS.

Hope this helps and makes someone's day a bit brighter

Jeff

Fri Jul 06, 2018 2:37 pm

Just tested Resolve Beta 6.

When the box under

Code: Select all: Configuration/System Configuration/ Always run Fusion on CPU

is not ticked, Resolve instantly crashes when adding a background and a merge node.

See this post here:
viewtopic.php?f=32&t=73490&p=408724#p408724

It would be really interesting if other people have the same crash on other Linux distros.
I am really worried that Fusion will die and ResolveFusion will never really work on Linux.

Sat Jul 07, 2018 9:29 pm

Daniel Tufvesson wrote:The latest version of MakeResolveDeb:
makeresolvedeb_15.0b5-1.sh.tar.gz
Or download from here for the latest updates and more info:
http://www.danieltufvesson.com/makeresolvedeb

Just to say,

Makeresolvedeb has been updated to 15.0b6.

Daniel's got it availabe on his website, but must've slipped his mind to post it here.

Sun Jul 08, 2018 9:19 am

Thank you for the reminder Solu

Yes, it did indeed slip my mind. Sorry about that. Work has been crazy this week.

The latest version of MakeResolveDeb:

makeresolvedeb_15.0b6-1.sh.tar.gz: (2.31 KiB) Downloaded 1156 times

It's also here on my website along with some more detailed instructions.

This time my testing took a bit longer that usual. Mostly because of other work but also because I discovered an audio related problem in one of my test setups that I in the end realized I've had all along everywhere.

Resolve appears to be constantly creating and destroying it's audio connection when I run Pulseaudio+ALSA. This can happen hundreds of times per second and may cause audio glitches and impact real-time playback performance. I'm currently having a bit of a struggle deciphering the Pulseaudio debug log.

Has anyone else seen this?

Code: Select all: Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] sink-input.c: Created input 2 "ALSA Playback" on alsa_output.usb-Yamaha_Corporation_MG-XU-00.analog-stereo with sample spec float32le 2ch 48000Hz and channel map front-left,front-right Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] sink-input.c: media.name = "ALSA Playback" Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] sink-input.c: application.name = "ALSA plug-in [resolve]" Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] sink-input.c: native-protocol.peer = "UNIX socket client" Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] sink-input.c: native-protocol.version = "32" Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] sink-input.c: application.process.id = "17914" Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] sink-input.c: application.process.user = "dtu" Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] sink-input.c: application.process.host = "resolve-deb9-ci" Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] sink-input.c: application.process.binary = "resolve" Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] sink-input.c: application.language = "C" Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] sink-input.c: window.x11.display = ":0.0" Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] sink-input.c: application.process.machine_id = "9ffe47c1f6ad47d6a9dd213aaf47dda2" Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] sink-input.c: application.process.session_id = "1" Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] sink-input.c: module-stream-restore.id = "sink-input-by-application-name:ALSA plug-in [resolve]" Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] protocol-native.c: Requested tlength=30.00 ms, minreq=7.50 ms Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] protocol-native.c: Final latency 37.50 ms = 15.00 ms + 2*7.50 ms + 7.50 ms Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] sink-input.c: Freeing input 2 "ALSA Playback" Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] sink-input.c: Trying to change sample rate Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] sink-input.c: Rate changed to 48000 Hz Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] module-stream-restore.c: Restoring mute state for sink input sink-input-by-application-name:ALSA plug-in [resolve]. Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] resampler.c: Forcing resampler 'copy', because of fixed, identical sample rates. Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] sink-input.c: Created input 3 "ALSA Playback" on alsa_output.usb-Yamaha_Corporation_MG-XU-00.analog-stereo with sample spec float32le 2ch 48000Hz and channel map front-left,front-right Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] sink-input.c: media.name = "ALSA Playback" Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] sink-input.c: application.name = "ALSA plug-in [resolve]" Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] sink-input.c: native-protocol.peer = "UNIX socket client" Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] sink-input.c: native-protocol.version = "32" Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] sink-input.c: application.process.id = "17914" Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] sink-input.c: application.process.user = "dtu" Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] sink-input.c: application.process.host = "resolve-deb9-ci" Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] sink-input.c: application.process.binary = "resolve" Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] sink-input.c: application.language = "C" Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] sink-input.c: window.x11.display = ":0.0" Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] sink-input.c: application.process.machine_id = "9ffe47c1f6ad47d6a9dd213aaf47dda2" Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] sink-input.c: application.process.session_id = "1" Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] sink-input.c: module-stream-restore.id = "sink-input-by-application-name:ALSA plug-in [resolve]" Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] protocol-native.c: Requested tlength=30.00 ms, minreq=7.50 ms Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] protocol-native.c: Final latency 37.50 ms = 15.00 ms + 2*7.50 ms + 7.50 ms Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] sink-input.c: Freeing input 3 "ALSA Playback" Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] client.c: Freed 6 "ALSA plug-in [resolve]" Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] protocol-native.c: Connection died. Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] client.c: Created 7 "Native client (UNIX socket client)" Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] client.c: Freed 7 "Native client (UNIX socket client)" Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] protocol-native.c: Connection died. Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] client.c: Created 8 "Native client (UNIX socket client)" Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] protocol-native.c: Got credentials: uid=1000 gid=1000 success=1 Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] sink-input.c: Trying to change sample rate Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] sink-input.c: Rate changed to 48000 Hz Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] module-stream-restore.c: Restoring mute state for sink input sink-input-by-application-name:ALSA plug-in [resolve]. Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] resampler.c: Forcing resampler 'copy', because of fixed, identical sample rates. Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] sink-input.c: Created input 4 "ALSA Playback" on alsa_output.usb-Yamaha_Corporation_MG-XU-00.analog-stereo with sample spec float32le 2ch 48000Hz and channel map front-left,front-right Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] sink-input.c: media.name = "ALSA Playback" Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] sink-input.c: application.name = "ALSA plug-in [resolve]" Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] sink-input.c: native-protocol.peer = "UNIX socket client" Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] sink-input.c: native-protocol.version = "32" Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] sink-input.c: application.process.id = "17914" Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] sink-input.c: application.process.user = "dtu" Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] sink-input.c: application.process.host = "resolve-deb9-ci" Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] sink-input.c: application.process.binary = "resolve" Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] sink-input.c: application.language = "C" Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] sink-input.c: window.x11.display = ":0.0" Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] sink-input.c: application.process.machine_id = "9ffe47c1f6ad47d6a9dd213aaf47dda2" Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] sink-input.c: application.process.session_id = "1" Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] sink-input.c: module-stream-restore.id = "sink-input-by-application-name:ALSA plug-in [resolve]" Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] protocol-native.c: Requested tlength=30.00 ms, minreq=7.50 ms Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] protocol-native.c: Final latency 37.50 ms = 15.00 ms + 2*7.50 ms + 7.50 ms Jul 6 23:35:53 resolve-deb9-ci pulseaudio[18340]: [pulseaudio] sink-input.c: Freeing input 4 "ALSA Playback"

Mon Jul 09, 2018 8:52 am

Hi,

I got Resolve 15B6 Studio working on my ubuntu 18.04 thanks to Daniel !!

So far two issues :
Sound glitches/scratch randomly but quite often....don't know how to check alsa/pulseaudio issues
no Cinema viewer mode (ctrl+F goes to a black screen).

Any clues to solve this ?

Fusion tab seems to work as well as color tab.

I am on an Dell XPS 15 9570

[EDIT] Also : each time I quit the app, it crashes. I'll be glad to send a report, but you'll have to tell me how ;-)

And is there a way to use the audio with something else than Alsa or Pulseaudio, if it is the issue ? (When I open the sound pref, in the Application tab, I can see the Alsa Resolve plug, but it seems to be unstable : the icon is flickering...). Which could be also related to the issue noted by Daniel.

But anyway, it seems to be pretty close to a working solution for Resolve on a main distro...

Mon Jul 09, 2018 2:13 pm

Daniel Tufvesson wrote:I'm afraid you still need to have CUDA installed for Resolve to run properly even if you are running OpenCL and vice versa.

How Resolve behaves after finding and initializing a CUDA board.

Code: Select all: [0x7f69122f4a40] | UI.GLContext | INFO | 2018-07-09 16:42:20,274 | Initialized OpenGL 4.6 (requested 2.0) on device 'NVIDIA

Are we sure it's OpenCL instead of OpenGL?

Mon Jul 09, 2018 7:05 pm

The audio situation described by Alexandre sounds exactly like what I'm seeing also. The flickering is very familiar. I'm actually not sure when it started but I have verified that the problem exist in both b5 and b6.

As far as I know Resolve only uses ALSA for system audio. The Decklink output works fine though.

The "Resolve quit unexpectedly" situation is known. It happens on all distributions. I think the best workaround for now is to uncheck "Send report when application quits unexpectedly" in general preferences.

The full screen viewer goes black for me as well in b6. In b5 it works.

Tue Jul 10, 2018 6:58 am

I’m experiencing the same random audio problems as Alexandre mentions, after installing beta 6 on Ubuntu 18.04. Thanks Daniel for the deb.

Tue Jul 10, 2018 7:24 am

So Resolve doesn't like h.264 on Linux. Seems like this is unavoidable until BMD decides to add support.

One of my cameras (a Canon T3i) records h.264 and there's no getting out of it. Anyone have a good, reliable ffmpeg command to run so I can transcode footage to DNxHD or ProRes? I've tried fiddling with the settings and I can never get it to work right.

Tue Jul 10, 2018 11:42 am

Sam Weston wrote:So Resolve doesn't like h.264 on Linux. Seems like this is unavoidable until BMD decides to add support.

Are you using Studio or the free version? And what container format is your footage? H.264 is supported, just not in the free version.

Cheers,
Mike

Tue Jul 10, 2018 2:38 pm

Hi folks.
I'm testing out Resolve 15b6 (and if I get it to work well enough I may even invest in this), but have trouble getting it to start from Ubuntu Studio 18.04. (Sales guys, please: Good video editors for Linux are rare, and I'm not very fond of the alternatives. Which I believe to be a market segment, and yes people like me will pay if it's good enough.)

But I get segmentation fault when I'm trying to run it. My GPU is an Nvidia Quadro K5200 (i really needed the drivers), running on top of an AMD APU (which has it's own small video support) but that shouldn't run now with the NVIDIA GPU.

I ran Daniels makeresolvedeb, latest edition, and installed the deb. I also read the start of this thread and ldd claims there are no missing libraries.
When it comes to logs, rollinglogs.txt reports some hundred lines of

[0x7f5478ab7480] | Main | INFO | 2018-07-10 15:40:15,189 | Updating display GPU information...

before it gives up, while .... reports this:

OpenCL error -1001, GPUPropertiesUtilUnix.cpp:338.
terminating

Some required information is missing for email crash report, the crash archive will not be updated

==========[CRASH DUMP]==========
Please send this to support:

#TIME Tue Jul 10 15:40:15 2018 - Uptime 00:00:03 (hh:mm:ss)
#PROGRAM_NAME DaVinci Resolve v15.0.0b.065 (Linux/Clang)

/opt/resolve/bin/resolve() [0x3ef09e9]
/opt/resolve/bin/resolve(_ZN7BtDebug14ReportSegfaultEiP7siginfoPv+0x6c) [0x3eeffec]
/lib/x86_64-linux-gnu/libpthread.so.0(+0x12890) [0x7f5463412890]
/lib/x86_64-linux-gnu/libc.so.6(gsignal+0xc7) [0x7f546141ee97]
/lib/x86_64-linux-gnu/libc.so.6(abort+0x141) [0x7f5461420801]
/opt/resolve/bin/../libs/libc++abi.so.1(+0x939c) [0x7f547b1c139c]
/opt/resolve/bin/../libs/libc++abi.so.1(+0x94c6) [0x7f547b1c14c6]
/opt/resolve/bin/../libs/libc++abi.so.1(+0x22036) [0x7f547b1da036]
/opt/resolve/bin/../libs/libc++abi.so.1(+0x223a1) [0x7f547b1da3a1]
/opt/resolve/bin/../libs/libc++abi.so.1(+0x228f7) [0x7f547b1da8f7]
/opt/resolve/bin/../libs/libc++abi.so.1(__gxx_personality_v0+0x6c) [0x7f547b1daddc]
/lib/x86_64-linux-gnu/libgcc_s.so.1(_Unwind_RaiseException+0x14b) [0x7f54617e8b6b]
/opt/resolve/bin/../libs/libc++abi.so.1(__cxa_throw+0x66) [0x7f547b1d99c6]
/opt/resolve/bin/resolve() [0x92f034]
/opt/resolve/bin/resolve() [0x92e829]
/opt/resolve/bin/resolve() [0x9268fa]
/opt/resolve/bin/resolve() [0x8f053c]
/opt/resolve/bin/resolve() [0x8edd7e]
/opt/resolve/bin/resolve() [0x8ec935]
/opt/resolve/bin/resolve() [0x8ec218]
/opt/resolve/bin/resolve() [0x928343]
/opt/resolve/bin/resolve(_ZN7BtDebug14ReportSegfaultEiP7siginfoPv+0x4d7) [0x3ef0457]
/lib/x86_64-linux-gnu/libpthread.so.0(+0x12890) [0x7f5463412890]
/lib/x86_64-linux-gnu/libc.so.6(gsignal+0xc7) [0x7f546141ee97]
/lib/x86_64-linux-gnu/libc.so.6(abort+0x141) [0x7f5461420801]
/opt/resolve/bin/../libs/libc++abi.so.1(+0x939c) [0x7f547b1c139c]
/opt/resolve/bin/../libs/libc++abi.so.1(+0x94c6) [0x7f547b1c14c6]
/opt/resolve/bin/../libs/libc++abi.so.1(+0x22036) [0x7f547b1da036]
/opt/resolve/bin/../libs/libc++abi.so.1(+0x223a1) [0x7f547b1da3a1]
/opt/resolve/bin/../libs/libc++abi.so.1(+0x228f7) [0x7f547b1da8f7]
/opt/resolve/bin/../libs/libc++abi.so.1(__gxx_personality_v0+0x6c) [0x7f547b1daddc]
/lib/x86_64-linux-gnu/libgcc_s.so.1(_Unwind_RaiseException+0x14b) [0x7f54617e8b6b]
/opt/resolve/bin/../libs/libc++abi.so.1(__cxa_throw+0x66) [0x7f547b1d99c6]
/opt/resolve/bin/resolve() [0x92f034]
/opt/resolve/bin/resolve() [0x92e829]
/opt/resolve/bin/resolve() [0x9268fa]
/opt/resolve/bin/resolve() [0x8f053c]
/opt/resolve/bin/resolve() [0x8edd7e]
/opt/resolve/bin/resolve() [0x8ec935]
/opt/resolve/bin/resolve() [0x8ec218]
/opt/resolve/bin/resolve() [0x928343]
/opt/resolve/bin/resolve(_ZN7BtDebug14ReportSegfaultEiP7siginfoPv+0x4d7) [0x3ef0457]
/lib/x86_64-linux-gnu/libpthread.so.0(+0x12890) [0x7f5463412890]
/lib/x86_64-linux-gnu/libc.so.6(gsignal+0xc7) [0x7f546141ee97]
/lib/x86_64-linux-gnu/libc.so.6(abort+0x141) [0x7f5461420801]
/opt/resolve/bin/../libs/libc++abi.so.1(+0x939c) [0x7f547b1c139c]
/opt/resolve/bin/../libs/libc++abi.so.1(+0x94c6) [0x7f547b1c14c6]
/opt/resolve/bin/../libs/libc++abi.so.1(+0x22036) [0x7f547b1da036]
/opt/resolve/bin/../libs/libc++abi.so.1(+0x223a1) [0x7f547b1da3a1]
/opt/resolve/bin/../libs/libc++abi.so.1(+0x228f7) [0x7f547b1da8f7]
Signal Number = 6

================================
[0x7f5478ab7480] | Main | INFO | 2018-07-10 15:40:15,189 | Updating display GPU information...

It's clear this is a GPU issue, but does any of you have an idea of exactly what it is, and how to solve it?
That'd be great, thank you.

Tue Jul 10, 2018 9:28 pm

Try installing the nvidia-opencl-icd package for your distribution and start Resovle again. Does that change anything?

Wed Jul 11, 2018 6:42 am

Daniel Tufvesson wrote:Try installing the nvidia-opencl-icd package for your distribution and start Resovle again. Does that change anything?

Hmm. nvidia-opencl-icd is a virtual package pointing to either libnvidia-compute-390 390.48-0ubuntu3, that I have already installed, or nvidia-opencl-icd-340 340.106-0ubuntu3, which seems like a deprecated version.
It seems ubuntu has renamed the package, for whatever reason.

I can try reinstalling libnvidia-compute-390, as I have a faulty drive (and the new one is still in the mail), so I may have lost some files. It breaks the whole package hierarchy for the nvidia drivers however, so I'll wait till I'm ready for reboot.

Wed Jul 11, 2018 5:03 pm

haraldthi wrote:[0x7f5478ab7480] | Main | INFO | 2018-07-10 15:40:15,189 | Updating display GPU information...
before it gives up, while .... reports this:
OpenCL error -1001, GPUPropertiesUtilUnix.cpp:338.
terminating

that's a quite useful indicator.

the error -1001 (=CL_PLATFORM_NOT_FOUND_KHR) does indeed signify: "No valid ICDs found" in the clGetPlatform* functions.

you should perhaps try the 'clinfo' command, to get some report about the available/working OpenCL services on your system.

it maybe related to nvideo opencl installation, but it could be also very likely caused by the AMD components. mixed GPU setups are always rather error prone.

Thu Jul 12, 2018 5:30 pm

Martin Schitter wrote:
haraldthi wrote:[0x7f5478ab7480] | Main | INFO | 2018-07-10 15:40:15,189 | Updating display GPU information...
before it gives up, while .... reports this:
OpenCL error -1001, GPUPropertiesUtilUnix.cpp:338.
terminating

that's a quite useful indicator.

the error -1001 (=CL_PLATFORM_NOT_FOUND_KHR) does indeed signify: "No valid ICDs found" in the clGetPlatform* functions.

you should perhaps try the 'clinfo' command, to get some report about the available/working OpenCL services on your system.

it maybe related to nvideo opencl installation, but it could be also very likely caused by the AMD components. mixed GPU setups are always rather error prone.

That was interesting, yes.
Running clinfo gave
Number of platforms 0
no matter if I ran it as clinfo -a, clinfo -a --offline or using sudo.
Seems there was nothing registered to run opencl, despite the whole nvidia driver repository being installed (sans some i386 packages).

I tried to install the mesa-opencl-icd package as an alternative, and when running clinfo this gave:
Number of platforms 1
Platform Name Clover
Platform Vendor Mesa
Platform Version OpenCL 1.1 Mesa 18.0.5
Platform Profile FULL_PROFILE
Platform Extensions cl_khr_icd
Platform Extensions function suffix MESA

Platform Name Clover
Number of devices 1
Device Name NVF1
Device Vendor NVIDIA
Device Vendor ID 0x10de
Device Version OpenCL 1.1 Mesa 18.0.5
Driver Version 18.0.5
Device OpenCL C Version OpenCL C 1.1
Device Type GPU
Device Profile FULL_PROFILE
Device Available Yes
Compiler Available Yes
Max compute units 12
Max clock frequency 512MHz
Max work item dimensions 3
Max work item sizes 1024x1024x64
Max work group size 1024
=== CL_PROGRAM_BUILD_LOG ===
invalid source Preferred work group size multiple <getWGsizes:675: build program : error -11>
Preferred / native vector sizes
char 16 / 16
short 8 / 8
int 4 / 4
long 2 / 2
half 0 / 0 (n/a)
float 4 / 4
double 2 / 2 (cl_khr_fp64)
Half-precision Floating-point support (n/a)
Single-precision Floating-point support (core)
Denormals No
Infinity and NANs Yes
Round to nearest Yes
Round to zero No
Round to infinity No
IEEE754-2008 fused multiply-add No
Support is emulated in software No
Correctly-rounded divide and sqrt operations No
Double-precision Floating-point support (cl_khr_fp64)
Denormals Yes
Infinity and NANs Yes
Round to nearest Yes
Round to zero Yes
Round to infinity Yes
IEEE754-2008 fused multiply-add Yes
Support is emulated in software No
Address bits 64, Little-Endian
Global memory size 1099511627776 (1024GiB)
Error Correction support No
Max memory allocation 1099511627776 (1024GiB)
Unified memory for Host and Device No
Minimum alignment for any data type 128 bytes
Alignment of base address 32768 bits (4096 bytes)
Global Memory cache type None
Image support No
Local memory type Local
Local memory size 49152 (48KiB)
Max number of constant args 15
Max constant buffer size 65536 (64KiB)
Max size of kernel argument 4096 (4KiB)
Queue properties
Out-of-order execution No
Profiling Yes
Profiling timer resolution 0ns
Execution capabilities
Run OpenCL kernels Yes
Run native kernels No
Device Extensions cl_khr_byte_addressable_store cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_local_int32_base_atomics cl_khr_local_int32_extended_atomics cl_khr_fp64

NULL platform behavior
clGetPlatformInfo(NULL, CL_PLATFORM_NAME, ...) Clover
clGetDeviceIDs(NULL, CL_DEVICE_TYPE_ALL, ...) Success [MESA]
clCreateContext(NULL, ...) [default] Success [MESA]
clCreateContextFromType(NULL, CL_DEVICE_TYPE_DEFAULT) Success (1)
Platform Name Clover
Device Name NVF1
clCreateContextFromType(NULL, CL_DEVICE_TYPE_CPU) No devices found in platform
clCreateContextFromType(NULL, CL_DEVICE_TYPE_GPU) Success (1)
Platform Name Clover
Device Name NVF1
clCreateContextFromType(NULL, CL_DEVICE_TYPE_ACCELERATOR) No devices found in platform
clCreateContextFromType(NULL, CL_DEVICE_TYPE_CUSTOM) No devices found in platform
clCreateContextFromType(NULL, CL_DEVICE_TYPE_ALL) Success (1)
Platform Name Clover
Device Name NVF1

ICD loader properties
ICD loader Name OpenCL ICD Loader
ICD loader Vendor OCL Icd free software
ICD loader Version 2.2.11
ICD loader Profile OpenCL 2.1

telling me that it perhaps could work as a glue layer to the Nvidia hardware, but running /opt/resolve/bin/resolve it froze and had to be stopped by a kill -9.

rollinglog.txt gives:
[0x7f99f3d3d480] | Undefined | INFO | 2018-07-12 18:54:27,151 | --------------------------------------------------------------------------------
[0x7f99f3d3d480] | Undefined | INFO | 2018-07-12 18:54:27,162 | Loaded log config from /opt/resolve/configs/log-conf.xml
[0x7f99f3d3d480] | Undefined | INFO | 2018-07-12 18:54:27,162 | --------------------------------------------------------------------------------
[0x7f99f3d3d480] | Main | INFO | 2018-07-12 18:54:27,167 | Running DaVinci Resolve v15.0.0b.065 (Linux/Clang)
[0x7f99f3d3d480] | Main | INFO | 2018-07-12 18:54:27,442 | Updating display GPU information...

and Resolvedebug.txt gives:
[0x7f99f3d3d480] | Main | INFO | 2018-07-12 18:54:27,167 | Running DaVinci Resolve v15.0.0b.065 (Linux/Clang)
[0x7f99f3d3d480] | Main | INFO | 2018-07-12 18:54:27,442 | Updating display GPU information...
double free or corruption (out)

So I wonder what has happened here.
The package nvidia-opencl-icd-384 is a transitional package depending on nvidia-headless-390, and that's the last we've seen of that. Has the Nvidia Opencl library been lost somehow?

Thu Jul 12, 2018 6:34 pm

haraldthi wrote:
Martin Schitter wrote:you should perhaps try the 'clinfo' command, to get some report about the available/working OpenCL services on your system.

That was interesting, yes.
Running clinfo gave
Number of platforms 0

yes -- that's definitely an indicator for a serious system missconfiguration.

haraldthi wrote:I tried to install the mesa-opencl-icd package as an alternative, and when running clinfo this gave:
Number of platforms 1
Platform Name Clover
Platform Vendor Mesa
Platform Version OpenCL 1.1 Mesa 18.0.5

i don't think, this mesa OpenCl 1.1 support on the CPU will work for resolve...

haraldthi wrote: Platform Name Clover
Number of devices 1
Device Name NVF1
Device Vendor NVIDIA
Device Vendor ID 0x10de
Device Version OpenCL 1.1 Mesa 18.0.5

and this section also doesn't look like the proprietary nvidia-driver.

maybe you didn't blacklist the open source "nouveau" driver module and it blocks the initialization of the proprietary nvidia-driver? this is usually not necessary, if you use the distribution specific nvidia-driver installation packages, because they handle most of this tasks for you, but it could be necessary, if you used the nvidia original installer or messed up your system by any similar manual installation action. anyhow you'll need the proprietary nvidia driver to run resolve, not noveau!

haraldthi wrote:ICD loader properties
ICD loader Name OpenCL ICD Loader
ICD loader Vendor OCL Icd free software
ICD loader Version 2.2.11
ICD loader Profile OpenCL 2.1

this section looks o.k. resp. very similar to my (working) setup.

haraldthi wrote:The package nvidia-opencl-icd-384 is a transitional package depending on nvidia-headless-390, and that's the last we've seen of that. Has the Nvidia Opencl library been lost somehow?

i use debian testing instead of ubuntu, therefore i can not tell you the exact packages names and working versions, but on debian you only need:

nvidia-kernel-dkms libcudart9.1 nvidia-opencl-icd

and all their "required" dependencies... -- and you can even avoid the installation of the nvidia specific xserver packages by more restrictive installation commands, if you don't use the nvidea cards for the actual video output.

Sat Jul 21, 2018 11:14 am

Hi Everyone,

I was just wondering if there was any news with the sound issue and the black full screen viewer with the 15b6 ? Any workaround ?

And thanks again to Daniel for his makeresolvedeb !

Sat Jul 21, 2018 5:34 pm

Hi Alexandre!

Unfortunately I don't think there are any workarounds for the current sound and full screen viewer problems. As far as I know it affects all distributions. Right now I think our best solution is to revert back to 15.0b5 until b7 is out that hopefully fixes these issues.

Since you are using makeresolvedeb, you can make a new 15.0b5 *.deb and install that on top of 15.0b6 and your Resolve config and database will stay intact!

Who is online