Posts by Keith Myers

1) Message boards : The Lounge : The Seti is Down Cafe (Message 92275)
Posted 16 hours ago by Profile Keith Myers
Post:
You are most welcome Richard. Great meeting you and Uli. Hope you see many redwoods today and get to hug a tree.

Wonder if murae makes it into the server stack today as the new upload server. And thanks for the photo of Amigos. Good to know that eventually we will see Parkes data.

I do hope that we can get some hint as to what the next hardware bits are needed for this year's fundraiser.
2) Message boards : The Lounge : The Seti is Down Cafe (Message 92072)
Posted 16 days ago by Profile Keith Myers
Post:
Wanted to see the graphs at Haveland but the site appears down now for a couple of hours. Anyone else have better luck? Have the assimilators and deleters finally peaked and started heading downward?
3) Message boards : Questions and problems : It appears that CPU tasks take way too long to crunch when GPU is active (Message 91996)
Posted 22 days ago by Profile Keith Myers
Post:
It depends on the application whether it needs just a little bit of cpu to support the gpu task or A LOT of cpu to support the gpu task. The SoG app at Seti expects a full cpu core to support the gpu task.
Same for Einstein but not for MW. If all depends on how the app developer wrote the application code. At Seti for example, the older CUDA42 or CUDA50 applications used just a fraction of cpu_time compared to the run_time. Another example is the "special app" at Seti which also uses very little of cpu_time compared to the run_time in its stock configuration. If you add the -nobs (no blocking sync) parameter to the cmdline in either the app_info or app_config, the application will use a full cpu thread to support the gpu task and the run_times will match the cpu_times and shave 5-10 seconds off the crunching time.
4) Message boards : Questions and problems : It appears that CPU tasks take way too long to crunch when GPU is active (Message 91951)
Posted 26 days ago by Profile Keith Myers
Post:
Keith: Perhaps, but this was not a problem before. I did update the bios, so perhaps with recent security patches, that has affected the processor. I've backed down to 75% of CPUs and 90% of CPU time, we will see how that affects things.

You mentioned security patches. The recent security mitigations for Intel processors have reduced their performance by 30% according to many online testing reviews of the patches. You might be observing this factor.

Updating the BIOS could have changed the setup of the processor clocks and boost settings and/or the memory speeds and the motherboard settings could have changed also from previous setup.

I would avoid using the internal iGPU for crunching as that normally severely increases the cpu crunching times.
5) Message boards : Questions and problems : It appears that CPU tasks take way too long to crunch when GPU is active (Message 91934)
Posted 28 days ago by Profile Keith Myers
Post:
Not enough cpu to support all the work attempted and not enough time slices per thread. Drop cpu utilization by either reducing cpu work or reducing gpu work that needs cpu attention.
6) Message boards : Questions and problems : Where is the SETI gpu application source located? (Message 91674)
Posted 30 May 2019 by Profile Keith Myers
Post:
Thanks Richard, no wonder I couldn't find it. I just assumed that directory was the cpu app.
I am trying to build the gpu apps for aarch64-unknown-linux-gnu for my Jetson Nano single board computer. I have an Einstein gpu app currently running on it as proof of concept.
7) Message boards : Questions and problems : Where is the SETI gpu application source located? (Message 91672)
Posted 30 May 2019 by Profile Keith Myers
Post:
I am having difficulties locating the gpu application sources. I have looked in Xbranch and sah_v7_opt and can't find any of the gpu applications. I can find the cpu and AP applications. But only Petri's gpu app is obvious.
8) Message boards : Questions and problems : AMD driver discovery problem revisited (Message 91533)
Posted 17 May 2019 by Profile Keith Myers
Post:
I'm not really familiar with AMD graphics having only run Nvidia. I see weekly updates to the Radeon drivers for Linux on news articles at Phoronix.com. When I searched for AMD drivers for Linux, I landed on
https://drivers.amd.com/drivers/linux/amdgpu-pro-19.10-782345-ubuntu-18.04.tar.xz
which covers the S9000 and S9100. But then I see a note down in Driver Details which mentions
Note: Customers who have upgraded to the latest 4.15 Kernel for Ubuntu will need to use an 18.20 based driver such as Radeon™ Pro Software Adrenalin Edition 18.Q3 for Linux.

which doesn't make sense because Ubuntu 18.04.2 has the 4.15 Kernel. That Radeon™ Pro Software Adrenalin Edition 18.Q3 for Linux doesn't cover the S9000 or S9100 cards. Huh?

I would stick to the the long term releases like 16.04 or 18.04. I think Ubuntu 18.04 is a very stable release. I had a test partition with 18.10 for a while which I updated to the 19.04 release. The release is very fast compared to 18.04 but it also ships with a GLIBC 2.29 versus the GLIBC 2.27 of 18.04 and threw a monkey wrench into my normal reason for testing and compiling in that test partition with any executable not being compatible with the earlier Linux versions. Based on the speed of 19.04, I think that when the 20.04 LTS release debuts next year, I will be doing a distribution upgrade.
9) Message boards : Questions and problems : AMD driver discovery problem revisited (Message 91530)
Posted 17 May 2019 by Profile Keith Myers
Post:
What project or app is complaining of using the four year old opencl.dll? I believe almost all the projects are written to use OpenCL 1.2 and I don't know of any that require OpenCL 2.0.
10) Message boards : Questions and problems : Need help in retrieving the sah_v7_opt svn repository (Message 91526)
Posted 16 May 2019 by Profile Keith Myers
Post:
I can view those pages without any issues. I just can't retrieve or download anything on those pages. The only way I can see for downloading any part of the svn repository is from the "Download in other formats:
Zip Archive" link on the bottom of any page. Using that link only gets the errors I posted previously. I don't see any other mechanism for downloading the branches on those pages unless I am completely blind. I even looked at the page source in developer mode looking for a download link and found none.

I posted the errors when I try to use the svn checkout command from the command line Terminal in the previous messages. That doesn't work either.

I need the sah_v7_opt directory so I can compile the science gpu apps for aarch64 platform.

Since I am unable to retrieve anything from the svn repository, I will take you up on your offer.

[Edit] OK. Finally figured out the correct URL to access the branches via checkout. The web interface still does not work but I just retrieved the seti_boinc and sah_v7_opt repositories via command line with svn checkout.

Thanks Richard for the offer. Not necessary now.
11) Message boards : Questions and problems : AMD driver discovery problem revisited (Message 91517)
Posted 16 May 2019 by Profile Keith Myers
Post:
The warning about no Nvidia library found is just part of DA's code. He tests for everything and the warning is benign. I asked a similar question because of a reported warning and found the answer in the code once Richard pointed me at the part that I walked. It does and can throw someone off on a tangent without knowing that is the normal response for not finding a library.
12) Message boards : Questions and problems : Need help in retrieving the sah_v7_opt svn repository (Message 91514)
Posted 15 May 2019 by Profile Keith Myers
Post:
This is the result with the insecure URL:

svn checkout http://setisvn.ssl.berkeley.edu/trac/browser/branches/sah_v7_opt
svn: E170011: Repository moved temporarily to 'https://setisvn.ssl.berkeley.edu/trac/browser/branches/sah_v7_opt'
13) Message boards : Questions and problems : Need help in retrieving the sah_v7_opt svn repository (Message 91513)
Posted 15 May 2019 by Profile Keith Myers
Post:
I tried the insecure version with checkout too.http://setisvn.ssl.berkeley.edu/trac/browser/branches/sah_v7_opt and get the same exact result.

How did you get a browser to accept http:// instead of https:// ??
I've tried both of the available browsers Chrome 73 and Firefox 66. Neither of them will allow me to enter just the unsecure version of the URL and always autocorrect it to https. Even though I tell the browser I am willing to enter a unsecure site.

Can you access it via the command line with svn checkout?
14) Message boards : Questions and problems : Need help in retrieving the sah_v7_opt svn repository (Message 91511)
Posted 15 May 2019 by Profile Keith Myers
Post:
I need help in retrieving the sah_v7_opt svn repository. I can't retrieve it via the command line svn checkout command:
keith@Serenity:~$ svn checkout https://setisvn.ssl.berkeley.edu/trac/browser/branches/sah_v7_opt
svn: E170013: Unable to connect to a repository at URL 'https://setisvn.ssl.berkeley.edu/trac/browser/branches/sah_v7_opt'
svn: E130005: XML Parsing failed: Unexpected root element 'html'


Nor via the browser interface by using the get zip archive function on the branch web page:
Oops…
Trac detected an internal error:
TimeoutError: Unable to get database connection within 0 seconds. (OperationalError: unable to open database file)
There was an internal error in Trac. It is recommended that you notify your local Trac administrator with the information needed to reproduce the issue.

To that end, you could  a ticket.

The action that triggered the error was:

GET: /changeset/4031/branches/sah_v7_opt


I created a ticket but have not received any response yet.
15) Message boards : GPUs : radeon R7 240 -- ubuntu 18.04 -- boinc not computing on it (Message 91471)
Posted 13 May 2019 by Profile Keith Myers
Post:
I've installed from amd site the official driver, but still not working.

The AMD Linux drivers do not automatically install the OpenCL component that is required for BOINC.
https://einsteinathome.org/content/quick-guide-how-install-opencl-amd-gpus-linux-kubuntu-1804-and-similar-distro?page=1
16) Message boards : GPUs : Is there a way in Linux Mint 19.1 to delay boinc start at boot? (Message 91470)
Posted 13 May 2019 by Profile Keith Myers
Post:
I would first try to add a SLEEP 60 to the beginning of the startup boinc-client script in /etc/init.d/boinc-client
sleep 60
set -e
. /lib/lsb/init-functions

If that doesn't work I would then try to add an AFTER statement in the service startup script in /lib/systemd/system/boinc-client.service

[Unit]
Description=Berkeley Open Infrastructure Network Computing Client
After=network.target
After=multi.user.target


The first suggestion simply tells the system to wait 60 seconds before executing the main parts of the boinc-client script. Disadvantage is that it will always take 60 seconds to restart the service after stopping it and restarting it.

The second suggestion tells the service not to start until the video drivers are loaded which the multi.user.target is defined for.
17) Message boards : The Lounge : The Seti is Down Cafe (Message 91433)
Posted 7 May 2019 by Profile Keith Myers
Post:
...and unborked :-)

and borked again!

and unborked again!
18) Message boards : The Lounge : The Seti is Down Cafe (Message 91432)
Posted 7 May 2019 by Profile Keith Myers
Post:
...and unborked :-)

and borked again!
19) Message boards : Questions and problems : Old reliable host can't connect to E@H (Message 91416)
Posted 7 May 2019 by Profile Keith Myers
Post:
indicates that google.com could be contacted, not as stated. So there's still a query there.

Thanks for the comment Richard, I see that I have been misinterpreting the client responses. I always read each line as the action:response.
Now I see that each line is self-contained, in this case the first line is the response:action of contacting the reference site and the response is in the next line.
Thanks for furthering my understanding of the client.
20) Message boards : Questions and problems : Old reliable host can't connect to E@H (Message 91399)
Posted 5 May 2019 by Profile Keith Myers
Post:
Great news, glad that fixed it. I am constantly surprised that people running clients as old as the 6.10.60 client are still able to interact with any project. Think those are only running the Seti project though. I had a hunch that Einstein with its aggressive stance toward personal information security enforced to be compliant with GDPR rules may have come into play.

I would have liked to have seen what the http_debug output for a connection request would have shown.


Next 20

Copyright © 2019 University of California. Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation.