SETI MB CUDA for Linux

[sunu]

sunu reported the url, but a reminder you need to install the 2.2 cuda sdk and tools too and update the libs in every directory you have them from the sdk lib dir. especially /usr/lib64, the boinc dir and  the seti project dir and make sure those paths are in ld.so.conf and rerun ldconfig.

Oh yes, forgot about that. You'll also need a cuda 2.2 compatible nvidia driver. Get it in http://www.nvidia.com/object/linux_display_amd64_185.18.14.html

[pp]

Any noticeable speedup?

[sunu]

Any noticeable speedup?

There is a small one.

[Tye]

Will it work with the new 190 driver and CUDA 2.3?

With the 185 driver and 2.2 libraries, I tried the 2.2 VLARkill app and putting the libraries in /usr/lib, the boinc folder, the proj*/setiat* folder, and my ~/bin but I kept getting computation errors with the 2.2 VLARkill app.  Argh.

I couldn't download the tools since they are now offering the 2.3 tools, hence the 2.3 question...

[Tye]

BTW, this is the error info that shows up with such a workunit:

<core_client_version>6.6.11</core_client_version>
<![CDATA[
<message>
process exited with code 193 (0xc1, -63)
</message>
<stderr_txt>

SETI@home MB CUDA_2.2 608 Linux 64bit SM 1.0 - r12 by Crunch3r :p
VLAR autokill mod

setiathome_CUDA: Found 1 CUDA device(s):
   Device 1 : GeForce 9600 GSO
           totalGlobalMem = 805044224
           sharedMemPerBlock = 16384
           regsPerBlock = 8192
           warpSize = 32
           memPitch = 262144
           maxThreadsPerBlock = 512
           clockRate = 1350000
           totalConstMem = 65536
           major = 1
           minor = 1
           textureAlignment = 256
           deviceOverlap = 1
           multiProcessorCount = 12
SIGSEGV: segmentation violation
Stack trace (16 frames):
setiathome-6.08.CUDA_2.2_x86_64-pc-linux-gnu(boinc_catch_signal+0x43)[0x485ef3]
/lib/libpthread.so.0[0x7fc1f6ee9080]
/usr/lib/libcuda.so.1[0x7fc1f7ebb020]
/usr/lib/libcuda.so.1[0x7fc1f7ec0d84]
/usr/lib/libcuda.so.1[0x7fc1f7e8a10f]
/usr/lib/libcuda.so.1[0x7fc1f7c15b3b]
/usr/lib/libcuda.so.1[0x7fc1f7c2646b]
/usr/lib/libcuda.so.1[0x7fc1f7c0e211]
/usr/lib/libcuda.so.1(cuCtxCreate+0xaa)[0x7fc1f7c07faa]
setiathome-6.08.CUDA_2.2_x86_64-pc-linux-gnu[0x5bf335]
setiathome-6.08.CUDA_2.2_x86_64-pc-linux-gnu[0x413c5b]
setiathome-6.08.CUDA_2.2_x86_64-pc-linux-gnu[0x41f68d]
setiathome-6.08.CUDA_2.2_x86_64-pc-linux-gnu[0x42b54d]
setiathome-6.08.CUDA_2.2_x86_64-pc-linux-gnu[0x408707]
/lib/libc.so.6(__libc_start_main+0xe6)[0x7fc1f6b865a6]
setiathome-6.08.CUDA_2.2_x86_64-pc-linux-gnu(__gxx_personality_v0+0x219)[0x408349]

Exiting...

</stderr_txt>
]]>

[sunu]

Will it work with the new 190 driver and CUDA 2.3?

I'm not too sure. I'm having some bad results.

With the 185 driver and 2.2 libraries, I tried the 2.2 VLARkill app and putting the libraries in /usr/lib, the boinc folder, the proj*/setiat* folder, and my ~/bin but I kept getting computation errors with the 2.2 VLARkill app.  Argh.

Are you sure that you've done everything from http://lunatics.kwsn.net/linux/seti-mb-cuda-for-linux.msg19014.html#msg19014 except for number1 if you have a multi-gpu setup?

[Tye]

Argh - sorry, I'm a dumbass!  Forgot to put the executable in my $PATH (in my case ~/bin).  Double-argh.  Will test again once I get some WUs, but I think that should do the trick (it did last time - you'd think I'd learn!).  Triple-argh.

Still only single GPU, but I've got the CPU and GPU both doing MB stuff thanks to you guys now.  Will pick up another GPU at some point though.

[sunu]

Still only single GPU, but I've got the CPU and GPU both doing MB stuff thanks to you guys now.  Will pick up another GPU at some point though.

Multi-gpu problems of the past with seti CUDA are pretty much solved now in linux.

[Tye]

Still only single GPU, but I've got the CPU and GPU both doing MB stuff thanks to you guys now.  Will pick up another GPU at some point though.

Multi-gpu problems of the past with seti CUDA are pretty much solved now in linux.

Yep, seems like 6.6.11 is the way to go - already using it to be prepared.

[riofl]

stupid question.. what  sets boinc's timing to report completed workunits? is it based on the number of days cache of work? it seems every time i look at it there are more than 100 wu ready to report and it doesn't seem to want to do that so i keep hitting manual update. i am presently keeping 6 days of work to do because of the recent problems at setihome. maybe i should drop it back to 4 or something?

plus i don't think that V5 script is working for me to grab VLARS as there are always 4 or 5 cuda computation error wu every time i look at it. i have never seen the script report them to me.

[Richard Haselgrove]

stupid question.. what  sets boinc's timing to report completed workunits?

There's a whole long list of possible triggers (see John McLeod VII's post at SETI for all the gory details), but in general no 'ready to report' task should hand around on your computer for longer than 24 hours. So they should disappear at least once per day without you doing anything.

[riofl]

stupid question.. what  sets boinc's timing to report completed workunits?

There's a whole long list of possible triggers (see John McLeod VII's post at SETI for all the gory details), but in general no 'ready to report' task should hand around on your computer for longer than 24 hours. So they should disappear at least once per day without you doing anything.

ok i wont touch it and wait till this time tomorrow to look to see if it has reported.
will check out that reference. thanks!

[Josef W. Segur]

The BOINC FAQ Service has very similar info in http://boincfaq.mundayweb.com/index.php?language=1&view=68, though jm7 has been more directly involved in the scheduling code development and lists things a little differently.
                                                                                    Joe

[koschi]

Thanks a lots for your tips sunu,

it run very well know i have got my firsts valids WUs on my GTS250 (about 5~10 min for a WU).It use about 2 % of CPU time to run so that i can crunch 2 other seti WU on my C2D E7300.

Bye

After crunching other projects for some months, I restarted SETI@GPU just his morning, but using the initial CUDA build for Linux which uses 100% of one CPU core as well...
So these new versions, that can be found in crunch3rs board, will use only few % of one CPU? That would be awsome...

I'm currently still at 180.60, some 2.6.30 rc and 6.6.17, but willing to update if I could free up that core with a never version of the app

[pp]

After crunching other projects for some months, I restarted SETI@GPU just his morning, but using the initial CUDA build for Linux which uses 100% of one CPU core as well...
So these new versions, that can be found in crunch3rs board, will use only few % of one CPU? That would be awsome...

I'm currently still at 180.60, some 2.6.30 rc and 6.6.17, but willing to update if I could free up that core with a never version of the app

I can confirm that setiathome-6.08.CUDA_2.2_x86_64-pc-linux-gnu no longer takes 100% CPU (only about 2-4% now on this PC). That's with CUDA 2.2 - I haven't been brave enough  to upgrade to CUDA 2.3 yet. Also using nVidia 185.18.14 and BOINC 6.9.0. The only downside so far is that the CPU time column now only shows actual CPU time used which is only a couple of minutes during the 19 minutes CUDA run. So no good way of checking exactly how long a WU takes now unless I monitor the clock manually.