bandwidthTest.cu(657) : cudaSafeCall() Runtime API error 2: out of memory.
I get the following error. Does anybody know what is wrong with my installation?

/Developer/GPU Computing/C/bin/darwin/release$ ./bandwidthTest
[bandwidthTest] starting...
./bandwidthTest Starting...

Running on...

Device 0: GeForce 9400M
Quick Mode

bandwidthTest.cu(657) : cudaSafeCall() Runtime API error 2: out of memory.


Here is my graphics card info and cpu info.

Hardware Overview:

Model Name: MacBook Pro
Model Identifier: MacBookPro5,5
Processor Name: Intel Core 2 Duo
Processor Speed: 2.53 GHz
Number Of Processors: 1
Total Number Of Cores: 2
L2 Cache: 3 MB
Memory: 8 GB
Bus Speed: 1.07 GHz
Boot ROM Version: MBP55.00AC.B03
SMC Version (system): 1.47f2
Sudden Motion Sensor:
State: Enabled


NVIDIA GeForce 9400M:

Chipset Model: NVIDIA GeForce 9400M
Type: GPU
Bus: PCI
VRAM (Total): 256 MB
Vendor: NVIDIA (0x10de)
Device ID: 0x0863
Revision ID: 0x00b1
ROM Revision: 3427
Displays:
Color LCD:
Resolution: 1280 x 800
Pixel Depth: 32-Bit Color (ARGB8888)
Mirror: Off
Online: Yes
Built-In: Yes
ASUS VW266H:
Resolution: 1920 x 1200 @ 60 Hz
Pixel Depth: 32-Bit Color (ARGB8888)
Main Display: Yes
Mirror: Off
Online: Yes
Rotation: Supported
Adapter Type: Mini DisplayPort To VGA Adapter
Adapter Firmware Version: 1.03
I get the following error. Does anybody know what is wrong with my installation?



/Developer/GPU Computing/C/bin/darwin/release$ ./bandwidthTest

[bandwidthTest] starting...

./bandwidthTest Starting...



Running on...



Device 0: GeForce 9400M

Quick Mode



bandwidthTest.cu(657) : cudaSafeCall() Runtime API error 2: out of memory.





Here is my graphics card info and cpu info.



Hardware Overview:



Model Name: MacBook Pro

Model Identifier: MacBookPro5,5

Processor Name: Intel Core 2 Duo

Processor Speed: 2.53 GHz

Number Of Processors: 1

Total Number Of Cores: 2

L2 Cache: 3 MB

Memory: 8 GB

Bus Speed: 1.07 GHz

Boot ROM Version: MBP55.00AC.B03

SMC Version (system): 1.47f2

Sudden Motion Sensor:

State: Enabled





NVIDIA GeForce 9400M:



Chipset Model: NVIDIA GeForce 9400M

Type: GPU

Bus: PCI

VRAM (Total): 256 MB

Vendor: NVIDIA (0x10de)

Device ID: 0x0863

Revision ID: 0x00b1

ROM Revision: 3427

Displays:

Color LCD:

Resolution: 1280 x 800

Pixel Depth: 32-Bit Color (ARGB8888)

Mirror: Off

Online: Yes

Built-In: Yes

ASUS VW266H:

Resolution: 1920 x 1200 @ 60 Hz

Pixel Depth: 32-Bit Color (ARGB8888)

Main Display: Yes

Mirror: Off

Online: Yes

Rotation: Supported

Adapter Type: Mini DisplayPort To VGA Adapter

Adapter Firmware Version: 1.03

#1
Posted 12/17/2011 05:27 AM   
bandwidthTest by default tries to transfer 32MB of data, which might be more than fits on your GPU. Try smaller transfers with an invocation similar to this one, which will test 1MB to 16MB in 1MB increments:

[font="Courier New"]./bandwidthTest --mode=range --start=1048576 --end=16777216 --increment=1048576[/font]

BTW, the 9400M is an integrated GPU that does not need any host<->device transfers at all. Use mapped memory to avoid the unnecessary copy.
bandwidthTest by default tries to transfer 32MB of data, which might be more than fits on your GPU. Try smaller transfers with an invocation similar to this one, which will test 1MB to 16MB in 1MB increments:



./bandwidthTest --mode=range --start=1048576 --end=16777216 --increment=1048576



BTW, the 9400M is an integrated GPU that does not need any host<->device transfers at all. Use mapped memory to avoid the unnecessary copy.

Always check return codes of CUDA calls for errors. Do not use __syncthreads() in conditional code unless the condition is guaranteed to evaluate identically for all threads of each block. Run your program under cuda-memcheck to detect stray memory accesses. If your kernel dies for larger problem sizes, it might exceed the runtime limit and trigger the watchdog timer.

#2
Posted 12/19/2011 01:06 PM   
hhm, why are the examples not tuned to use the smallest ammount of data or even better shouldn't they detect the available amount of RAM like any other application?
hhm, why are the examples not tuned to use the smallest ammount of data or even better shouldn't they detect the available amount of RAM like any other application?

#3
Posted 03/26/2012 01:10 PM   
Scroll To Top