Emulation mode ? all correct ? / how to access the GPU mode ?

Accelerated Computing CUDA CUDA Programming and Performance
1

Hi all,

I have a MacBookPro in the second latest release, which (I guess) has a 8600M GT card. If I run the CUDA SDK examples, I am not sure if everything runs properly, which is why I would like to post three results, which hopefully help to clarify some issues:

alex:release jazzalex$ ./transpose
Using device 0: Device Emulation (CPU)
Transposing a 256 by 4096 matrix of floats…
Naive transpose average time: 0.067 ms
Optimized transpose average time: 0.039 ms

Test FAILED

alex:release jazzalex$ ./bandwidthTest
Quick Mode
Host to Device Bandwidth for Pageable memory
.
Transfer Size (Bytes) Bandwidth(MB/s)
33554432 1250000.0

Quick Mode
Device to Host Bandwidth for Pageable memory
.
Transfer Size (Bytes) Bandwidth(MB/s)
33554432 1787709.5

Quick Mode
Device to Device Bandwidth
.
Transfer Size (Bytes) Bandwidth(MB/s)
33554432 3076923.0

&&&& Test PASSED

alex:release jazzalex$ ./clock
Using device 0: Device Emulation (CPU)
Test PASSED
time = 0

Since most tests tell me that they are running in device emulation mode, I am wondering how I can switch to GPU mode ? Or is there anything incorrectly configured or simply wrong ? E.g. the bandwidth test confuses me due to the extreme high values.

Thanks in advance, any hint appreciated,
best

– A l e x

2

… ups, I just saw this has been discussed already:

[url=“http://forums.nvidia.com/index.php?showtopic=92647”]http://forums.nvidia.com/index.php?showtopic=92647[/url]

I did a device query now, which clearly showed that I am running in emulation mode:

alex:release jazzalex$ ./deviceQuery
There is no device supporting CUDA.

Device 0: “Device Emulation (CPU)”
Major revision number: 9999
Minor revision number: 9999
Total amount of global memory: 4294967295 bytes
Number of multiprocessors: 16
Number of cores: 128
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: 1
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
Maximum memory pitch: 262144 bytes
Texture alignment: 256 bytes
Clock rate: 1.35 GHz
Concurrent copy and execution: No

Test PASSED

Are still some news or would I simply need to check the “personalize” mode in the installer ?

Thanks

– A l e x

3

yeah, I’d make sure CUDA.kext was installed.