integer arithmetic overflow

Accelerated Computing CUDA CUDA Programming and Performance
1

Is it defined how integer arithmetic overflows?

For instance, is it guaranteed that adding or multiplying two large unsigned ints will “gracefully” overflow like modulus 2^32?

I imagine this is somewhat hardware specific, so I’ll take GeForce 8800 as the example platform.

2

p.s. integer math appears to overflow “gracefully” on GeForce 8800 GTS. That is, the low 32 bits are correct.