Does it work yet, running one task per GPU, with multiple GPUs on Windows? Back in April, I think it was still being worked on. But I haven't heard any update since then.
I have sent a PM to the dev. At the moment no.
Got it. I will stay tuned for more info.
____________
Reno, NV
Team: SETI.USA
/ 
Reply
Quote
rebirtherVolunteer moderator
Project administrator
Project developer
Project tester
Project scientist
Send message
Joined: 2 Jan 13
Posts: 7479
Credit: 43,584,303
RAC: 35,116
cuda11/11.1 linux apps are up, could still need some tweaking. If you have a RTX3080/90 pls test if you are getting work depends on driver and cuda version, thx to DeleteNull to made it possible!
11.1 is slightly faster then 11.0
/ Reply
Quote
rebirtherVolunteer moderator
Project administrator
Project developer
Project tester
Project scientist
Send message
Joined: 2 Jan 13
Posts: 7479
Credit: 43,584,303
RAC: 35,116
After some tweaking the plan_class_spec is working now correctly.
/ Reply
Quote
rebirtherVolunteer moderator
Project administrator
Project developer
Project tester
Project scientist
Send message
Joined: 2 Jan 13
Posts: 7479
Credit: 43,584,303
RAC: 35,116
./mfaktc.exe: error while loading shared libraries: libcudart.so.10.1: cannot open shared object file: No such file or directory (for 2xxx cards)
Note: If you are using a higher cuda11 driver you must copy the missing file in /usr/lib64 (for linux), you will get a mix of cuda100 and cuda111 apps
You can find the info also in FAQ
/ Reply
Quote
Boy, points sure went to shit the past day or 2.
/ Reply
Quote
Boy, points sure went to shit the past day or 2.
Something odd in the points awarding is happening.
See my message:
http://srbase.my-firewall.org/sr5/forum_thread.php?id=1498&postid=7130#7130
/ Reply
Quote
Boy, points sure went to shit the past day or 2.
Something odd in the points awarding is happening.
See my message:
http://srbase.my-firewall.org/sr5/forum_thread.php?id=1498&postid=7130#7130
UPDATE: I was seeing a WU results delay and likely not related to your concern.
/ Reply
Quote
./mfaktc.exe: error while loading shared libraries: libcudart.so.10.1: cannot open shared object file: No such file or directory (for 2xxx cards)
Note: If you are using a higher cuda11 driver you must copy the missing file in /usr/lib64 (for linux), you will get a mix of cuda100 and cuda111 apps
You can find the info also in FAQ
Hi, I get these errors. Where i can find libcudart.so.10.1?
Linux Mint 19.3
Thanx for hints
/ Reply
Quote
rebirtherVolunteer moderator
Project administrator
Project developer
Project tester
Project scientist
Send message
Joined: 2 Jan 13
Posts: 7479
Credit: 43,584,303
RAC: 35,116
./mfaktc.exe: error while loading shared libraries: libcudart.so.10.1: cannot open shared object file: No such file or directory (for 2xxx cards)
Note: If you are using a higher cuda11 driver you must copy the missing file in /usr/lib64 (for linux), you will get a mix of cuda100 and cuda111 apps
You can find the info also in FAQ
Hi, I get these errors. Where i can find libcudart.so.10.1?
Linux Mint 19.3
Thanx for hints
see FAQ
/ Reply
Quote
This has probably been answered before but i hate boinc forum search, is it possible to select a particular cuda app version? You said earlier that planclass is working i think, can that be used to specify i want cuda100 only?
/ Reply
Quote
rebirtherVolunteer moderator
Project administrator
Project developer
Project tester
Project scientist
Send message
Joined: 2 Jan 13
Posts: 7479
Credit: 43,584,303
RAC: 35,116
This has probably been answered before but i hate boinc forum search, is it possible to select a particular cuda app version? You said earlier that planclass is working i think, can that be used to specify i want cuda100 only?
If you are under these specs then yes but cant specified by yourself:
/ Reply
Quote
rebirtherVolunteer moderator
Project administrator
Project developer
Project tester
Project scientist
Send message
Joined: 2 Jan 13
Posts: 7479
Credit: 43,584,303
RAC: 35,116
New linux cuda100 TF app is up which included the libcudart.so.10.1 in the zip. I hope this fixes the linking issue (only affecting users which have cuda11 installed)
/ Reply
Quote
Got 1 time the cuda 100 TF Workunits and all other times the non working 0.12 Cuda 111 Workunits. Is there an option for selecting the cuda 100 Version?
/ Reply
Quote
rebirtherVolunteer moderator
Project administrator
Project developer
Project tester
Project scientist
Send message
Joined: 2 Jan 13
Posts: 7479
Credit: 43,584,303
RAC: 35,116
Got 1 time the cuda 100 TF Workunits and all other times the non working 0.12 Cuda 111 Workunits. Is there an option for selecting the cuda 100 Version?
no, its defined in the plan_class
/ Reply
Quote
<core_client_version>7.9.3</core_client_version>
<![CDATA[
<message>
process exited with code 195 (0xc3, -61)</message>
<stderr_txt>
14:18:59 (20946): wrapper (7.2.26012): starting
14:18:59 (20946): wrapper: running ./mfaktc.exe ( --device 0)
./mfaktc.exe: /lib/x86_64-linux-gnu/libm.so.6: version `GLIBC_2.29' not found (required by ./mfaktc.exe)
14:19:00 (20946): ./mfaktc.exe exited; CPU time 0.000622
14:19:00 (20946): app exit status: 0x100
14:19:00 (20946): called boinc_finish
</stderr_txt>
]]>
A team member said that i have to Install a newer Ubuntu version because of GLIBC_2.29 error.
I choose the LTS version because i don't want to install every year a new OS.
/ Reply
Quote
rebirtherVolunteer moderator
Project administrator
Project developer
Project tester
Project scientist
Send message
Joined: 2 Jan 13
Posts: 7479
Credit: 43,584,303
RAC: 35,116
7.9.3
process exited with code 195 (0xc3, -61)
14:18:59 (20946): wrapper (7.2.26012): starting
14:18:59 (20946): wrapper: running ./mfaktc.exe ( --device 0)
./mfaktc.exe: /lib/x86_64-linux-gnu/libm.so.6: version `GLIBC_2.29' not found (required by ./mfaktc.exe)
14:19:00 (20946): ./mfaktc.exe exited; CPU time 0.000622
14:19:00 (20946): app exit status: 0x100
14:19:00 (20946): called boinc_finish
]]>
A team member said that i have to Install a newer Ubuntu version because of GLIBC_2.29 error.
I choose the LTS version because i don't want to install every year a new OS.
It must be possible to install only a newer glibc version
/ Reply
Quote
rebirtherVolunteer moderator
Project administrator
Project developer
Project tester
Project scientist
Send message
Joined: 2 Jan 13
Posts: 7479
Credit: 43,584,303
RAC: 35,116
The range 72-73bit is nearly done. The rest of the work was uploaded on the server.
/ Reply
Quote
The range 72-73bit is nearly done. The rest of the work was uploaded on the server.
So does that mean that we move onto the 73-74 bit range next? If so, does that have a large effect on TF as a project?
/ Reply
Quote
So does that mean that we move onto the 73-74 bit range next? If so, does that have a large effect on TF as a project?
The impact of trial factoring from 73 bit to 74 bit, will mean following:
Tasks to factor: 16,865,230
Likeliness of factor: 1.435 percent (maybe little higher)
First time tests saved by factoring: 242,016 tests.
The impact is as you can see very high. A 12 core 2.3 GHz Xeon (not modern at all) uses about 48 CPU days per test at n=100M. It then uses 192 days at n=200M. It then uses 768 days at n=400M. It then uses 3,072 days at n=800M.
From n=100M to n=200M tests saved 16,232 (1,947,840 CPU days saved)
From n=200M to n=400M tests saved 51,851 (24,888,480 CPU days saved)
From n=400M to n=800M tests saved 116,577 (223,827,840 CPU days saved)
So the impact is a whooping 250,664,160 saved CPU days of testing, on a 12 core 2.3 GHz Xeon and by all means very huge and meaningfull.
Thanks for asking the question :)
Take care.
/ Reply
Quote
The impact of trial factoring from 73 bit to 74 bit, will mean following:
<snips a stunning amount of precise technical information.>
Thank you, that’s fascinating!
/ Reply
Quote