Runs on older, fails on newer GSO9600 fine, GTX 460 fails

My app running on GPU with these params:

Device 0: “GeForce 9600 GSO”
CUDA Driver Version: 3.20
CUDA Runtime Version: 3.20
CUDA Capability Major/Minor version number: 1.1
Total amount of global memory: 402325504 bytes
Multiprocessors x Cores/MP = Cores: 12 (MP) x 8 (Cores/MP) = 96 (Cores)
Total amount of constant memory: 65536 bytes
Total amount of shared memory per block: 16384 bytes
Total number of registers available per block: 8192
Warp size: 32
Maximum number of threads per block: 512
Maximum sizes of each dimension of a block: 512 x 512 x 64
Maximum sizes of each dimension of a grid: 65535 x 65535 x 1
Clock rate: 1.70 GHz
Concurrent copy and execution: Yes
Run time limit on kernels: Yes
Integrated: No
Support host page-locked memory mapping: Yes
Compute mode: Default (multiple host threads can use th
usly)
Concurrent kernel execution: No
Device has ECC support enabled: No
Device is using TCC driver mode: No

but fails with error 30, unknown error, on GPU with such params:

Device 0: “GeForce GTX 460”
CUDA Driver Version: 4.0
CUDA Runtime Version: 3.20
CUDA Capability Major/Minor version number: 2.1
Total amount of global memory: 1008271360 bytes
Multiprocessors x Cores/MP = Cores: 7 (MP) x 48 (Cores/MP) = 336 (Cores)
Total amount of constant memory: 65536 bytes
Total amount of shared memory per block: 49152 bytes
Total number of registers available per block: 32768
Warp size: 32
Maximum number of threads per block: 1024
Maximum sizes of each dimension of a block: 1024 x 1024 x 64
Maximum sizes of each dimension of a grid: 65535 x 65535 x 65535
Clock rate: 1.60 GHz
Concurrent copy and execution: Yes
Run time limit on kernels: Yes
Integrated: No
Support host page-locked memory mapping: Yes
Compute mode: Default (multiple host threads can use this device simultaneously)
Concurrent kernel execution: Yes
Device has ECC support enabled: No
Device is using TCC driver mode: No

Can it be driber version (4.0) CUDA runtime version (3.2) mismatch? And what is this “30” error that app reports?
CUDA call ‘cudaStreamSynchronize(0)’ failed (30) in file ‘d:/R/SETI6/AKv8/client/MB_CUDA_kernels.cu’ in line 3025: unknown error.

You could upgrade to toolkit 4.0, rebuild your application and it would probably be fixed.

Otherwise please post more code.

Unfortunately upgrading to CUDA 4 not an option. 275 (particularly 275.33) drivers changed synching way for OpenCL app [and I develop both CUDA and OpenCL], now whole CPU core used for synching, it’s too wasteful for high-performance CPU+GPU computing we do.

I’ll post more info about this error after some internal debugging probably.

Hi,

I’ve got the same error. I have an application with streams which works perfectly with CUDA 3.1 in a GTX-275. Now I have changed to CUDA 4.0 and GTX-580,I have installed all the drivers and when the application arrives to the line “cudaStreamSynchronize”. It returns to me Unknown Error with code cudaError_t = 30.

Thank you in advance
Jose