It's a sensationalist headline, but the problem is one that needs to be addressed -- whether it's acceptable for the kernel to just assume that the RDRAND instr

It's a sensationalist headline, but the problem is one that needs to be addressed -- whether it's acceptable for the kernel to just assume that the RDRAND instruction does what Intel claims it does.