[CUDPP][1120] Simple cudpp program error

[谢乃闻]

Hi, all.
I am new to cudpp. My demo program "simple Cudpp" does not run well.
My error is just like the previous
one:https://groups.google.com/forum/#!topic/cudpp/w2WQod7nKnM

"Failed to use CUDPP 2.0 on my desktop, windows7 sp1, GTX 660 Ti card,
CUDA 5.0, VS2010
I have built CUDPP 2.0 successfully using CMAKE and VS2010.
But when I try to lauch the simpleCUDPP or the cudpp_testrig program,
they hung up and keep stuck there.
For example, the simpleCUDPP seems to be trapped in the "cudaMalloc"
routine between the two breakpoint in the image above.".


My working platform is "win 7 sp1, GTX 660 TI card, cuda 5.5 vs 2008"
Different from the previous question, I want do build the program in
sm_2.0 architecture. but I do not know why the program fails when the
configuration is already set in cMakeList,.
Thanks.

[John Owens]

Well, first, do try to let the program run for a while if it's trying to possibly JIT kernels.
Second, can you successfully build and run other programs (the CUDA samples for instance)?
Third, we are not testing CUDPP any more with such an old version of CUDA. It would be a lot easier to try to reproduce your case if you were up to date on CUDA releases (7.0).

JDO

Post by 谢乃闻
Hi, all.
I am new to cudpp. My demo program "simple Cudpp" does not run well.
My error is just like the previous one:https://groups.google.com/forum/#!topic/cudpp/w2WQod7nKnM
"Failed to use CUDPP 2.0 on my desktop, windows7 sp1, GTX 660 Ti card, CUDA 5.0, VS2010
I have built CUDPP 2.0 successfully using CMAKE and VS2010.
But when I try to lauch the simpleCUDPP or the cudpp_testrig program, they hung up and keep stuck there.
For example, the simpleCUDPP seems to be trapped in the "cudaMalloc" routine between the two breakpoint in the image above.".
My working platform is "win 7 sp1, GTX 660 TI card, cuda 5.5 vs 2008"
Different from the previous question, I want do build the program in sm_2.0 architecture. but I do not know why the program fails when the configuration is already set in cMakeList,.
Thanks.
--
You received this message because you are subscribed to the Google Groups "CUDPP" group.
Visit this group at http://groups.google.com/group/cudpp.
For more options, visit https://groups.google.com/d/optout.

[谢乃闻]

Thanks. I already solve it by the solution provided by yizhi tang in
https://groups.google.com/forum/#!topic/cudpp/w2WQod7nKnM

圚 2015幎5月17日星期日 UTC+8䞋午11:10:43John Owens写道

Post by John Owens
Well, first, do try to let the program run for a while if it's trying to
possibly JIT kernels.
Second, can you successfully build and run other programs (the CUDA samples for instance)?
Third, we are not testing CUDPP any more with such an old version of CUDA.
It would be a lot easier to try to reproduce your case if you were up to
date on CUDA releases (7.0).
JDO

Post by 谢乃闻
Hi, all.
I am new to cudpp. My demo program "simple Cudpp" does not run well.

https://groups.google.com/forum/#!topic/cudpp/w2WQod7nKnM

Post by 谢乃闻
"Failed to use CUDPP 2.0 on my desktop, windows7 sp1, GTX 660 Ti card,

CUDA 5.0, VS2010

Post by 谢乃闻
I have built CUDPP 2.0 successfully using CMAKE and VS2010.
But when I try to lauch the simpleCUDPP or the cudpp_testrig program,

they hung up and keep stuck there.

Post by 谢乃闻
For example, the simpleCUDPP seems to be trapped in the "cudaMalloc"

routine between the two breakpoint in the image above.".

Post by 谢乃闻
My working platform is "win 7 sp1, GTX 660 TI card, cuda 5.5 vs 2008"
Different from the previous question, I want do build the program in

sm_2.0 architecture. but I do not know why the program fails when the
configuration is already set in cMakeList,.

Post by 谢乃闻
Thanks.
--
You received this message because you are subscribed to the Google

Groups "CUDPP" group.

Post by 谢乃闻
To unsubscribe from this group and stop receiving emails from it, send

<javascript:>.

Post by 谢乃闻
Visit this group at http://groups.google.com/group/cudpp.
For more options, visit https://groups.google.com/d/optout.