1

I am taking CS420 - Operating Systems. My professor mentioned something in class the other day that peaked my interest and I have been researching it and have gotten close but cannot find out the answer.

He mentioned that him and another professor had wondered in the past if the creation of the ++/-- operators was originally intended to bind the ++/-- operators to the test-and-set operation/instruction.

He said that they weren't able to figure it out. I didn't ask him how much effort they had invested in this but would be very interested in the communities input.

S.S. Anne
  • 15,171
  • 8
  • 38
  • 76
Michael Riley
  • 67
  • 1
  • 10
  • 1
    I don't have any references handy, but I don't think this is the case. Many/most historical compilers generate code for `++` and `--` which is not atomic even on uni-processor machines. – R.. GitHub STOP HELPING ICE Oct 01 '15 at 04:47
  • 3
    https://www.bell-labs.com/usr/dmr/www/chist.html (search for “Thompson went a step further by inventing the `++` and `--` operators”) – cremno Oct 01 '15 at 04:48

1 Answers1

1

C is a platform independent language and designed as such. It was designed to compile against various machine languages and architectures, so it's difficult imagine they were counting on a test-and-set instruction or any CPU family-specific instruction in particular.

clearlight
  • 12,255
  • 11
  • 57
  • 75