Installing Cuda 3.0 tools on new Macbook pro (GT330M)

elm · April 24, 2010, 2:46pm

As anybody had any luck installing the Cuda 3.0 tools on the new macbook pro with GT330M graphic card?
On my previous macbook pro with 9400/9600M, after installation, I would also see the driver info in the system preference in the “other” section.

By default the Intel is used and automatically switched to GT330M when needed. I reinstalled after having enabled GT300M fulltime … without much luck.
It seems that the installer does not detect the GT330M properly and therefore does not install the drivers properly.

Any other MAC guys around with similar problems?
Thanks
Eric

yifli · April 29, 2010, 3:55pm

Same problem here. Running deviceQuery gave :

./deviceQuery Starting…

CUDA Device Query (Runtime API) version (CUDART static linking)

There is no device supporting CUDA

deviceQuery, CUDA Driver = CUDART, CUDA Driver Version = 53931, CUDA Runtime Version = 0.0, NumDevs = 0

PASSED

yifli · April 29, 2010, 4:00pm

wait, I did manage to install the driver.

I can see the CUDA icon in ‘other’ section, and it says:

CUDA driver version: 3.0.14

GPU driver version: 1.6.10.34(19.5.2f18)

cgn1234 · May 6, 2010, 1:13am

How did you install the driver?

./deviceQuery Starting…

CUDA Device Query (Runtime API) version (CUDART static linking)

cudaGetDeviceCount FAILED CUDA Driver and Runtime version may be mismatched.

FAILED

The directions say:

Install the CUDA Driver by installing the appropriate CUDA driver package based on the GPU in the system (either NVIDIA GeForce GPU or NVIDIA Quadro GPU).

But they don’t say where you get the driver or how to install it.

renambot · May 9, 2010, 3:08am

After installing the CUDA driver, everything went fine.

Here are some results for those interested, from deviceQuery and bandwidthTest

% ./deviceQuery
./deviceQuery Starting…

CUDA Device Query (Runtime API) version (CUDART static linking)

There is 1 device supporting CUDA

Device 0: “GeForce GT 330M”
CUDA Driver Version: 3.0
CUDA Runtime Version: 3.0
CUDA Capability Major revision number: 1
CUDA Capability Minor revision number: 2
Total amount of global memory: 536543232 bytes
Number of multiprocessors: 6
Number of cores: 48
Total amount of constant memory: 65536 bytes
Total amount of shared memory per block: 16384 bytes
Total number of registers available per block: 16384
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
Maximum memory pitch: 2147483647 bytes
Texture alignment: 256 bytes
Clock rate: 1.10 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)

deviceQuery, CUDA Driver = CUDART, CUDA Driver Version = 53931, CUDA Runtime Version = 3.0, NumDevs = 1, Device = GeForce GT 330M

PASSED

% ./bandwidthTest --memory=pinned
[bandwidthTest]
./bandwidthTest Starting…

Running on…

Device 0: GeForce GT 330M
Quick Mode

Host to Device Bandwidth, 1 Device(s), Pinned memory, Write-Combined Memory Enabled
Transfer Size (Bytes) Bandwidth(MB/s)
33554432 2640.5

Device to Host Bandwidth, 1 Device(s), Pinned memory, Write-Combined Memory Enabled
Transfer Size (Bytes) Bandwidth(MB/s)
33554432 3206.5

Device to Device Bandwidth, 1 Device(s)
Transfer Size (Bytes) Bandwidth(MB/s)
33554432 8534.0

[bandwidthTest] - Test results:
PASSED

cgn1234 · May 9, 2010, 3:33pm

After installing the CUDA driver, everything went fine.

Here are some results for those interested, from deviceQuery and bandwidthTest

% ./deviceQuery

./deviceQuery Starting…

CUDA Device Query (Runtime API) version (CUDART static linking)

There is 1 device supporting CUDA

Device 0: “GeForce GT 330M”

CUDA Driver Version: 3.0

CUDA Runtime Version: 3.0

CUDA Capability Major revision number: 1

CUDA Capability Minor revision number: 2

Total amount of global memory: 536543232 bytes

Number of multiprocessors: 6

Number of cores: 48

Total amount of constant memory: 65536 bytes

Total amount of shared memory per block: 16384 bytes

Total number of registers available per block: 16384

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

Maximum memory pitch: 2147483647 bytes

Texture alignment: 256 bytes

Clock rate: 1.10 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)

deviceQuery, CUDA Driver = CUDART, CUDA Driver Version = 53931, CUDA Runtime Version = 3.0, NumDevs = 1, Device = GeForce GT 330M

PASSED

% ./bandwidthTest --memory=pinned

[bandwidthTest]

./bandwidthTest Starting…

Running on…

Device 0: GeForce GT 330M

Quick Mode

Host to Device Bandwidth, 1 Device(s), Pinned memory, Write-Combined Memory Enabled

Transfer Size (Bytes) Bandwidth(MB/s)

33554432 2640.5

Device to Host Bandwidth, 1 Device(s), Pinned memory, Write-Combined Memory Enabled

Transfer Size (Bytes) Bandwidth(MB/s)

33554432 3206.5

Device to Device Bandwidth, 1 Device(s)

Transfer Size (Bytes) Bandwidth(MB/s)

33554432 8534.0

[bandwidthTest] - Test results:

PASSED

How do you install the CUDA driver?

I installed the cudatoolkit_3.0_macos_32.pkg.

/Chris

Thanks,

/Chris

aterrel · May 11, 2010, 4:31pm

FYI: I had the exact same problems as above and it didn’t resolve itself until the machine was rebooted.

cgn1234 · May 12, 2010, 2:27pm

Thanks. I tried rebooting, but I’ll try it again.

/Chris

cgn1234 · May 15, 2010, 2:29pm

No dice. I still get:

./deviceQuery Starting…

CUDA Device Query (Runtime API) version (CUDART static linking)

cudaGetDeviceCount FAILED CUDA Driver and Runtime version may be mismatched.

FAILED

Just to be clear, all I should have to do is install the cudatoolkit_3.0_macos_32.pkg and then the gpucomputingsdk_3.0_macros_32.pkg and run the make file?

Thanks,

/Chris

renambot · May 18, 2010, 4:40am

did you install the driver ?

First entry in the table of MacOS download:

Developer Drivers for MacOS

http://developer.nvidia.com/object/cuda_3_0_downloads.html

Luc

kcjin000 · July 2, 2010, 12:14pm

Dear,

MacBook Pro 6.2 2.66Ghz Intel Core i7, MacOSX: 10.6.4,

NVIDIA GeForce GT 330M, Memory: 8GB 1067 MHz DDR3

“dmesg” command shows,

npvhash=4095

PAE enabled

64 bit mode enabled

Darwin Kernel Version 10.4.0: Fri Apr 23 18:28:53 PDT 2010; root:xnu-1504.7.4~1/RELEASE_I386

vm_page_bootstrap: 1989319 free pages and 107833 wired pages

I installed cudadriver_3.0.14_macos.dmg
Reboot.
In CUDA Preferences, dialog shows
- CUDA Driver version: 3.0.14
- GPU Driver Version: 1.6.16.11 (19.5.8f01)
Next, installed cudatoolkit_3.0_macos_32.pkg
Next, installed gpucomputingsdk_3.0_macox_32.pkg
run make command
The messages of “deviceQuery” command show

There is no device supporting CUDA

deviceQuery, CUDA Driver=CUDART, CUDA Driver Version=55683, CUDA Runtime Version=0.0 NumDevs=0

PASSED

Any comment would be appreciated ~

Best Regards,

kcjin000@hotmail.com

kcjin000 · July 2, 2010, 12:14pm

Dear,

MacBook Pro 6.2 2.66Ghz Intel Core i7, MacOSX: 10.6.4,

NVIDIA GeForce GT 330M, Memory: 8GB 1067 MHz DDR3

“dmesg” command shows,

npvhash=4095

PAE enabled

64 bit mode enabled

Darwin Kernel Version 10.4.0: Fri Apr 23 18:28:53 PDT 2010; root:xnu-1504.7.4~1/RELEASE_I386

vm_page_bootstrap: 1989319 free pages and 107833 wired pages

I installed cudadriver_3.0.14_macos.dmg
Reboot.
In CUDA Preferences, dialog shows
- CUDA Driver version: 3.0.14
- GPU Driver Version: 1.6.16.11 (19.5.8f01)
Next, installed cudatoolkit_3.0_macos_32.pkg
Next, installed gpucomputingsdk_3.0_macox_32.pkg
run make command
The messages of “deviceQuery” command show

There is no device supporting CUDA

deviceQuery, CUDA Driver=CUDART, CUDA Driver Version=55683, CUDA Runtime Version=0.0 NumDevs=0

PASSED

Any comment would be appreciated ~

Best Regards,

kcjin000@hotmail.com

oldmanjank · July 8, 2010, 6:44pm

I had all the same issued above with CUDA 3.1 on my i7 MBP with the GT330M.

Apparently you have to force the use of the nvidia drivers after they’re installed. This handy tool [url=“gfxCardStatus by cody krieger”]http://codykrieger.com/gfxCardStatus/[/url] allows you to do just that. Forcing the nvidia card to be used (not intel or dynamic switching) had deviceQuery showing the right device and had my CUDA programs running again. Hope this works for you too.

aranaman · October 2, 2010, 9:14pm

Thanks @oldmanjank! This really helped me out for my MBP 2010 i5.

I had exactly the same error message you described above:

./deviceQuery Starting...

CUDA Device Query (Runtime API) version (CUDART static linking)

There is no device supporting CUDA

deviceQuery, CUDA Driver = CUDART, CUDA Driver Version = 53931, CUDA Runtime Version = 0.0, NumDevs = 0

PASSED

Press <Enter> to Quit...

-----------------------------------------------------------

Then I installed gfxCardStatus, switched to “NVIDIA only” mode, reinstalled the driver and surprise surprise I’m getting this:

./deviceQuery Starting...

CUDA Device Query (Runtime API) version (CUDART static linking)

There is 1 device supporting CUDA

Device 0: "GeForce GT 330M"

  CUDA Driver Version:						   3.10

  CUDA Runtime Version:						  3.0

  CUDA Capability Major revision number:		 1

  CUDA Capability Minor revision number:		 2

  Total amount of global memory:				 268107776 bytes

  Number of multiprocessors:					 6

  Number of cores:							   48

  Total amount of constant memory:			   65536 bytes

  Total amount of shared memory per block:	   16384 bytes

  Total number of registers available per block: 16384

  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

  Maximum memory pitch:						  2147483647 bytes

  Texture alignment:							 256 bytes

  Clock rate:									1.10 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)

deviceQuery, CUDA Driver = CUDART, CUDA Driver Version = 53931, CUDA Runtime Version = 3.0, NumDevs = 1, Device = GeForce GT 330M

PASSED

Press <Enter> to Quit...

-----------------------------------------------------------

Good software!

aranaman · October 2, 2010, 9:14pm

Thanks @oldmanjank! This really helped me out for my MBP 2010 i5.

I had exactly the same error message you described above:

./deviceQuery Starting...

CUDA Device Query (Runtime API) version (CUDART static linking)

There is no device supporting CUDA

deviceQuery, CUDA Driver = CUDART, CUDA Driver Version = 53931, CUDA Runtime Version = 0.0, NumDevs = 0

PASSED

Press <Enter> to Quit...

-----------------------------------------------------------

Then I installed gfxCardStatus, switched to “NVIDIA only” mode, reinstalled the driver and surprise surprise I’m getting this:

./deviceQuery Starting...

CUDA Device Query (Runtime API) version (CUDART static linking)

There is 1 device supporting CUDA

Device 0: "GeForce GT 330M"

  CUDA Driver Version:						   3.10

  CUDA Runtime Version:						  3.0

  CUDA Capability Major revision number:		 1

  CUDA Capability Minor revision number:		 2

  Total amount of global memory:				 268107776 bytes

  Number of multiprocessors:					 6

  Number of cores:							   48

  Total amount of constant memory:			   65536 bytes

  Total amount of shared memory per block:	   16384 bytes

  Total number of registers available per block: 16384

  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

  Maximum memory pitch:						  2147483647 bytes

  Texture alignment:							 256 bytes

  Clock rate:									1.10 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)

deviceQuery, CUDA Driver = CUDART, CUDA Driver Version = 53931, CUDA Runtime Version = 3.0, NumDevs = 1, Device = GeForce GT 330M

PASSED

Press <Enter> to Quit...

-----------------------------------------------------------

Good software!

Topic		Replies	Views
CUDA Installation on 2013 Macbook Pro (NVIDIA GeForce GT 750M) - Driver version incompatibility CUDA Setup and Installation	10	10562	March 8, 2020
NVIDIA 320M on MB Pro 13" NVIDIA 320M Graphic Card on the new MacBook Pro 13" CUDA Programming and Performance	11	16508	November 5, 2010
CUDA still shows as "Update Required" after installing latest package for macOS CUDA Setup and Installation	86	87415	June 19, 2018
CUDA driver on Mac OS Mavericks CUDA Setup and Installation	23	25214	May 2, 2015
MAC CUDA driver fully compatible with macOS High Sierra 10.13 (error CUDA Setup and Installation	254	334800	February 25, 2020
No CUDA device after rebooting... CUDA Programming and Performance	7	10977	December 7, 2011
Problems with CUDA 9.1 in Ubuntu 16.04 CUDA Setup and Installation	36	24291	May 15, 2018
CUDA Driver for Lion CUDA Programming and Performance	18	211587	June 27, 2012
10.6.4 Update & CUDA GPU driver Following updating of OS X to 10.6.4 now asked to update CUDA dr CUDA Programming and Performance	36	15494	June 29, 2010
Setting up CUDA on MacBook Pro with NVIDIA GeForce GT 650M CUDA Setup and Installation	6	12669	November 7, 2017

Installing Cuda 3.0 tools on new Macbook pro (GT330M)

Related topics