Specification of cc 2.0 inconsistent with cudaDeviceQuer results?

> /usr/local/cuda/sdk/C/bin/linux/release$ ./deviceQuery

[deviceQuery] starting...

./deviceQuery Starting...

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

Found 2 CUDA Capable device(s)

Device 0: "Tesla C2070"

 CUDA Driver Version / Runtime Version 	4.0 / 4.0

 CUDA Capability Major/Minor version number:	2.0

 Total amount of global memory: 	6143 MBytes (6441598976 bytes)

 (14) Multiprocessors x (32) CUDA Cores/MP: 	448 CUDA Cores

 GPU Clock Speed: 	1.15 GHz

 Memory Clock rate: 	1494.00 Mhz

 Memory Bus Width: 	384-bit

 L2 Cache Size: 	786432 bytes

 Max Texture Dimension Size (x,y,z) 	1D=(65536), 2D=(65536,65535), 3D=(2048,2048,2048)

 Max Layered Texture Size (dim) x layers 	1D=(16384) x 2048, 2D=(16384,16384) x 2048


Now look at the "Max Texture Dimension Size (x,y,z) " entry for 1D: the SDK program returns 65536.

Compare it with NVidia CUDA Programming Guide, ver. 4.0, page 159, Appendix F (Compute capability).

The entry “Maximum width for a 1D texture reference bound to a CUDA array” for cc 2.x reads 32768, whereas

the entry “Maximum width for a 1D texture reference bound to linear memory” reads 2[sup]27[/sup].

None of these two numbers equals 65536, as reported by deviceQuery through cudaDeviceQuery.

Is it a (documentation?) bug or am I misunderstanding something?