Error -6 when creating context, Linux 64 bit, ION Error -6 occurs when creating a context, 64bit Fed

Accelerated Computing CUDA CUDA Programming and Performance
1

Hello.

I am running Fedora11 64 bit on an Atom330 with an ION chipset. i have the nvdrivers_2.3_linux_64_190.29 installed

oclDeviceQuery passes, returning the following:

oclDeviceQuery.exe Starting...

OpenCL SW Info:

CL_PLATFORM_NAME: 	NVIDIA

 CL_PLATFORM_VERSION: 	OpenCL 1.0 

 OpenCL SDK Version: 	4788711

OpenCL Device Info:

1 devices found supporting OpenCL:

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

 Device ION

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

  CL_DEVICE_NAME: 			ION

  CL_DEVICE_VENDOR: 			NVIDIA Corporation

  CL_DRIVER_VERSION: 			190.29

  CL_DEVICE_TYPE:			CL_DEVICE_TYPE_GPU

  CL_DEVICE_MAX_COMPUTE_UNITS:		2

  CL_DEVICE_MAX_WORK_ITEM_DIMENSIONS:	3

  CL_DEVICE_MAX_WORK_ITEM_SIZES:	512 / 512 / 64 

  CL_DEVICE_MAX_WORK_GROUP_SIZE:	512

  CL_DEVICE_MAX_CLOCK_FREQUENCY:	1100 MHz

  CL_DEVICE_ADDRESS_BITS:		32

  CL_DEVICE_MAX_MEM_ALLOC_SIZE:		128 MByte

  CL_DEVICE_GLOBAL_MEM_SIZE:		61 MByte

  CL_DEVICE_ERROR_CORRECTION_SUPPORT:	no

  CL_DEVICE_LOCAL_MEM_TYPE:		local

  CL_DEVICE_LOCAL_MEM_SIZE:		16 KByte

  CL_DEVICE_MAX_CONSTANT_BUFFER_SIZE:	64 KByte

  CL_DEVICE_QUEUE_PROPERTIES:		CL_QUEUE_OUT_OF_ORDER_EXEC_MODE_ENABLE

  CL_DEVICE_QUEUE_PROPERTIES:		CL_QUEUE_PROFILING_ENABLE

  CL_DEVICE_IMAGE_SUPPORT:		1

  CL_DEVICE_MAX_READ_IMAGE_ARGS:	128

  CL_DEVICE_MAX_WRITE_IMAGE_ARGS:	8

CL_DEVICE_IMAGE <dim>			2D_MAX_WIDTH	 8192

					2D_MAX_HEIGHT	 8192

					3D_MAX_WIDTH	 2048

					3D_MAX_HEIGHT	 2048

					3D_MAX_DEPTH	 2048

CL_DEVICE_EXTENSIONS:			cl_khr_byte_addressable_store

					cl_nv_compiler_options

					cl_nv_device_attribute_query

					cl_khr_global_int32_base_atomics

					cl_khr_global_int32_extended_atomics

CL_NV_DEVICE_COMPUTE_CAPABILITY:	1.1

  CL_NV_DEVICE_REGISTERS_PER_BLOCK:	8192

  CL_NV_DEVICE_WARP_SIZE:		32

  CL_NV_DEVICE_GPU_OVERLAP:		CL_FALSE

  CL_NV_DEVICE_KERNEL_EXEC_TIMEOUT:	CL_TRUE

  CL_NV_DEVICE_INTEGRATED_MEMORY:	CL_TRUE

  CL_DEVICE_PREFERRED_VECTOR_WIDTH_<t>	CHAR 1, SHORT 1, INT 1, FLOAT 1, DOUBLE 1

oclDeviceQuery, Platform Name = NVIDIA, Platform Version = OpenCL 1.0 , SDK Version = 4788711, NumDevs = 1, Device = ION

System Info: 

Local Time/Date =  14:36:19, 10/19/2009

 CPU Name: Intel(R) Atom(TM) CPU 330 @ 1.60GHz 

 # of CPU processors: 4

 Linux version 2.6.30.8-64.fc11.x86_64 (mockbuild@x86-5.fedora.phx.redhat.com) (gcc version 4.4.1 20090725 (Red Hat 4.4.1-2) (GCC) ) #1 SMP Fri Sep 25 04:43:32 EDT 2009

TEST PASSED

Press <Enter> to Quit...

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

However that’s the only success I get :(

oclBandwidthTest fails, as do oclConvolutionSeparable and oclNbody

./oclBandwidthTest Starting...

Running on...

Device ION

 Quick Mode

Failed to create OpenCL context!

TEST FAILED !!!

Press <Enter> to Quit...

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

./oclConvolutionSeparable Starting...

Allocating and initializing host memory...

Initializing OpenCL...

!!! Error # -6 at line 59 , in file src/main.cpp !!!

Exiting...

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

./oclNbody Starting...

!!! Error # -6 at line 178 , in file src/oclNbody.cpp !!!

Starting Cleanup...

oclNbody.exe Exiting...

Press <Enter> to Quit

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

Each time I am unable to create a context. Can someone help me with this please.

2

-6 is CL_OUT_OF_HOST_MEMORY. However, what I suspect is that the error happens in the first clGetContextInfo() call (one that is fetching the amount of memory needed for storing IDs for available devices), in both SDK examples, so I would try with checking error immediately after this call.

3

I have a test program where the first thing I do is create a context via clCreateContextFromType(0, CL_DEVIDE_TYPE_GPU, NULL, NULL, &ciErrNum) and that returns -6. I never get to call clGetContextInfo