SETI MB CUDA for Linux

[Tye]

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.

Yup, that's been bothering me too.  I'm wondering if there's a way to trick it into reporting clock time rather than cpu time...  I'm using nvidia 185.18.14 and BOINC 6.6.11 btw since I'd like to do multi-GPU here soon.

[riofl]

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.

Yup, that's been bothering me too.  I'm wondering if there's a way to trick it into reporting clock time rather than cpu time...  I'm using nvidia 185.18.14 and BOINC 6.6.11 btw since I'd like to do multi-GPU here soon.

6.6.37 was reporting proper cpu/gpu times, but when i went back to 6.6.11 to use multiple devices that time reporting broke. i am not sure if adding a flops statement in app_info.xml  will help with that or not.

for the ability to use multiple devices not having proper reporting time is an irritant but one i will gladly put up with until there is a newer version with fixed device code.

[koschi]

Follow all steps (1-4) below:

1)  Use a newer boinc version. The latest is 6.6.36, http://boinc.berkeley.edu/download_all.php . I haven't checked it, I use 6.6.20, direct download link http://boinc.berkeley.edu/dl/boinc_6.6.20_x86_64-pc-linux-gnu.sh
2)  Make sure all the appropriate cuda libs from 2.2 toolkit

libcudart.so
libcudart.so.2
libcudart.so.2.2
libcufft.so
libcufft.so.2
libcufft.so.2.2

are in the projects/setiathome.berkeley.edu directory.

3)  Edit accordingly your ld.so.conf or the corresponding ld-something file of your distro with the above location of the cuda libs.

4)  Place a copy of the cuda client in one of the following locations:

/usr/local/sbin
/usr/local/bin
/usr/sbin
/usr/bin
/sbin
/bin
/usr/games

Thanks a lot, especially number 4 I would have never tried, its a little insane  
One unit was already successfully validated. The CPU usage on my Q6600 is at roughly 1-2s/1min wall clock time.
The default priority of nice 10 seems to slow the process down on my box, once I switched it to 0 or -5, it processed much faster and collected up CPU time quicker.

[riofl]

hmm maybe i should try that. im still using nice 19. im only using 3 of my cpus leaving 1 to handle both gpu apps and my desktop. that seems to be most efficient, running cpus at 100% i found the times dropped from an average of 1 hr 25 min to an hour or under. and thats at nice 19.. maybe i should play a bit in priorities.

[riofl]

off topic but i thought if anyone would know it would be here.

is there any utility for x86_64 besides lm_sensors that will properly monitor the GTX285 fans?

I assume this is done with the ADT7473 chip. it is enabled in the kernel but lm_sensors reports back "no driver for Analog Devices ADT7473 yet" and also "unknown adapter NVIDIA i2c adapter".

temps are monitoring just fine i just cannot see the fans.

[koschi]

Try nvclock: http://www.linuxhardware.org/nvclock/#download
The current 0.8b4 is already part of Ubuntu Jaunty, if you are using some older version or another flavour of Linux you might want to compile it yourself, if it isn't available at that level. Around December/January quite soem progress has been made in that tool, though its still not supporting all nvidia cards.
My 8800GTS also uses the ADT7473, temperature readings show an offset of 8°C to the nvidia-settings temperature reading, fan controlling is working.

[riofl]

Try nvclock: http://www.linuxhardware.org/nvclock/#download
The current 0.8b4 is already part of Ubuntu Jaunty, if you are using some older version or another flavour of Linux you might want to compile it yourself, if it isn't available at that level. Around December/January quite soem progress has been made in that tool, though its still not supporting all nvidia cards.
My 8800GTS also uses the ADT7473, temperature readings show an offset of 8°C to the nvidia-settings temperature reading, fan controlling is working.

i have 8b4 and i set the fans manually with it but it cannot handle the gtx285 yet:
nvclock --info
It seems your card isn't officialy supported in NVClock yet.
The reason can be that your card is too new.
If you want to try it anyhow [DANGEROUS], use the option -f to force the setting(s).
NVClock will then assume your card is a 'normal', it might be dangerous on other cards.
Also please email the author the pci_id of the card for further investigation.
[Get that value using the -i option].


i am looking for something with support for the ADT7473 as a module or interface so  i can display  them in gkrellm on my desktop. i dont know how gkrellm gets the gpu temps but it reports them accurately. maybe ill send bill (the author of gkrellm) a note to ask if he knows how to monitor them directly.

[Kunin]

I see this line:

<file_name>setiathome-6.08.CUDA_2.2_x86_64-pc-linux-gnu</file_name>

But I only have the old app without the 2.2 part - where are you getting this CUDA app?

Get it from http://calbe.dw70.de/mb/viewtopic.php?f=9&t=110

Will that work better with CUDA 2.3 than the one at the start of this thread?

I tried to go to the link, but every time I go to calbe.dw70.de I get access denied... is there another place to get it?

[koschi]

Earlier versions of nvclock I had to convince with -f to read the data of my card, but it worked quite well...

Has anyone tried the CUDA 2.2 client together with 190.xx drivers and the CUDA 2.3 dlls, if there is some speed-up like under Windows?

[Kunin]

I'm currently running CUDA 2.3 with the app from the start of this thread., does seem to be a speed up.

[riofl]

does anyone  know if there is a cuda 2.3 vlarkill x86_64 app available yet? i am switching everything to 2.3 and the 190 driver today.

[letni]

I'm trying to get CUDA working (with the 32 bit binary posted at message 1 of this thread) with my new 8600GTS in Slackware Linux and I'm having issues.. I have run the nvidia installer, etc, but I get some weird errors..

1. The output shows I have a cuda device, however, it says I have revision 0 of the driver installed, even though I have installed the 185.18.14 and updaged to the 185.18.31..
 CUDA device: GeForce 8600 GTS (driver version 0, comp
ute capability 1.1, 255MB, est. 18GFLOPS).
2. I have modified my app_info.xml to allow both AK_V8_SSE3 (32bit) and the cuda to run simultaneously (included .xml file).. I have 3 active tasks being worked on, two (for my dual CPU) say setiathome_enhanced 6.03 and run just fine.  The third is the CUDA which setiathome_enhanced 6.08 (cuda), and the status NEVER goes past Ready to start. It will eventuall error out with Computation error.
3. I have tried the CUDA toolkit 2.0 and 2.2 based libs (put them in the projects/setiathome.berkeley.edu directory.

4. I'm not using XWindows at all.  This is all console based only.. Is Xorg required to be running to utilize CUDA?

Anyone have any thoughts or advice on how to debug this?


[attachment deleted by admin]

[riofl]

will look things over more closely later after work but what version of boinc are you running?

x is not required. the only requirement there is that the nvidia driver loads properly.

you placed the libs in the project dir but did you make sure the project dir is included in your system ld.so.conf file? and did you re-run ldconfig?

try ldd <setiathome application name> and make sure there are no errors and every line points to a library.

[koschi]

Could you provide a link to your host, or some failed work units?
If you follow the instructions provided by sunu and make sure the Nvidia modules are loaded, then it will also work for you

I'm running the 190.18 driver with CUDA 2.3 libraries and 2.2VLARkill app now on two machines with G92 chips, so far no isses.

[Tye]

Could you provide a link to your host, or some failed work units?
If you follow the instructions provided by sunu and make sure the Nvidia modules are loaded, then it will also work for you

And whatever you do, don't forget to do step #4 - that's gotten me twice now.  ;-)