Memory checker reports access errors with surface write

Hi

On Windows 8.1, Cuda 6.5 when i launch

simpleSurfaceWrite.cu in the samples directory

I get access violations, details follow:

The example works, however seems that the runtime tries to fetch more bytes from the global memory for the surface object itself. It tries to fetch 4 more bytes.

Driver issue? cuda runtime issue?

reported as:

CUDA context created : f18fc083f0
CUDA module loaded: f1911721b0 C:/ProgramData/NVIDIA Corporation/CUDA Samples/v6.5/0_Simple/simpleSurfaceWrite/simpleSurfaceWrite.cu

CUDA Memory Checker detected 32 threads caused an access violation:
Launch Parameters
CUcontext = f18fc083f0
CUstream = f19070ca40
CUmodule = f1911721b0
CUfunction = f1911b8c00
FunctionName = _Z18surfaceWriteKernelPfii
GridId = 24
gridDim = {64,64,1}
blockDim = {8,8,1}
sharedSize = 256
Parameters:
height = 512
gIData = 0x0000000700340000 0.63529414
width = 512
Parameters (raw):
0x00340000 0x00000007 0x00000200 0x00000200
GPU State:
Address Size Type Mem Block Thread blockIdx threadIdx PC Source

700060960 4 adr ld g 0 0 {0,0,0} {0,0,0} _Z18surfaceWriteKernelPfii+000e90 c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_functions.h:4437
700060960 4 adr ld g 0 1 {0,0,0} {1,0,0} _Z18surfaceWriteKernelPfii+000e90 c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_functions.h:4437
700060960 4 adr ld g 0 2 {0,0,0} {2,0,0} _Z18surfaceWriteKernelPfii+000e90 c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_functions.h:4437
700060960 4 adr ld g 0 3 {0,0,0} {3,0,0} _Z18surfaceWriteKernelPfii+000e90 c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_functions.h:4437
700060960 4 adr ld g 0 4 {0,0,0} {4,0,0} _Z18surfaceWriteKernelPfii+000e90 c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_functions.h:4437
700060960 4 adr ld g 0 5 {0,0,0} {5,0,0} _Z18surfaceWriteKernelPfii+000e90 c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_functions.h:4437
700060960 4 adr ld g 0 6 {0,0,0} {6,0,0} _Z18surfaceWriteKernelPfii+000e90 c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_functions.h:4437
700060960 4 adr ld g 0 7 {0,0,0} {7,0,0} _Z18surfaceWriteKernelPfii+000e90 c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_functions.h:4437
700060960 4 adr ld g 0 8 {0,0,0} {0,1,0} _Z18surfaceWriteKernelPfii+000e90 c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_functions.h:4437
700060960 4 adr ld g 0 9 {0,0,0} {1,1,0} _Z18surfaceWriteKernelPfii+000e90 c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_functions.h:4437
700060960 4 adr ld g 0 10 {0,0,0} {2,1,0} _Z18surfaceWriteKernelPfii+000e90 c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_functions.h:4437
700060960 4 adr ld g 0 11 {0,0,0} {3,1,0} _Z18surfaceWriteKernelPfii+000e90 c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_functions.h:4437
700060960 4 adr ld g 0 12 {0,0,0} {4,1,0} _Z18surfaceWriteKernelPfii+000e90 c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_functions.h:4437
700060960 4 adr ld g 0 13 {0,0,0} {5,1,0} _Z18surfaceWriteKernelPfii+000e90 c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_functions.h:4437
700060960 4 adr ld g 0 14 {0,0,0} {6,1,0} _Z18surfaceWriteKernelPfii+000e90 c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_functions.h:4437
700060960 4 adr ld g 0 15 {0,0,0} {7,1,0} _Z18surfaceWriteKernelPfii+000e90 c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_functions.h:4437
700060960 4 adr ld g 0 16 {0,0,0} {0,2,0} _Z18surfaceWriteKernelPfii+000e90 c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_functions.h:4437
700060960 4 adr ld g 0 17 {0,0,0} {1,2,0} _Z18surfaceWriteKernelPfii+000e90 c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_functions.h:4437
700060960 4 adr ld g 0 18 {0,0,0} {2,2,0} _Z18surfaceWriteKernelPfii+000e90 c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_functions.h:4437
700060960 4 adr ld g 0 19 {0,0,0} {3,2,0} _Z18surfaceWriteKernelPfii+000e90 c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_functions.h:4437
700060960 4 adr ld g 0 20 {0,0,0} {4,2,0} _Z18surfaceWriteKernelPfii+000e90 c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_functions.h:4437
700060960 4 adr ld g 0 21 {0,0,0} {5,2,0} _Z18surfaceWriteKernelPfii+000e90 c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_functions.h:4437
700060960 4 adr ld g 0 22 {0,0,0} {6,2,0} _Z18surfaceWriteKernelPfii+000e90 c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_functions.h:4437
700060960 4 adr ld g 0 23 {0,0,0} {7,2,0} _Z18surfaceWriteKernelPfii+000e90 c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_functions.h:4437
700060960 4 adr ld g 0 24 {0,0,0} {0,3,0} _Z18surfaceWriteKernelPfii+000e90 c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_functions.h:4437
700060960 4 adr ld g 0 25 {0,0,0} {1,3,0} _Z18surfaceWriteKernelPfii+000e90 c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_functions.h:4437
700060960 4 adr ld g 0 26 {0,0,0} {2,3,0} _Z18surfaceWriteKernelPfii+000e90 c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_functions.h:4437
700060960 4 adr ld g 0 27 {0,0,0} {3,3,0} _Z18surfaceWriteKernelPfii+000e90 c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_functions.h:4437
700060960 4 adr ld g 0 28 {0,0,0} {4,3,0} _Z18surfaceWriteKernelPfii+000e90 c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_functions.h:4437
700060960 4 adr ld g 0 29 {0,0,0} {5,3,0} _Z18surfaceWriteKernelPfii+000e90 c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_functions.h:4437
700060960 4 adr ld g 0 30 {0,0,0} {6,3,0} _Z18surfaceWriteKernelPfii+000e90 c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_functions.h:4437
700060960 4 adr ld g 0 31 {0,0,0} {7,3,0} _Z18surfaceWriteKernelPfii+000e90 c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_functions.h:4437

Summary of access violations:
c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_functions.h(4437): error MemoryChecker: #misaligned=0 #invalidAddress=32