Just to add to the points given above, I understand this was one of the reasons for the long distance walk on the second EVA of Apollo 14 - to test the ability of the astronauts to travel back to the LM - I suspect both in terms of navigation and endurance.
As Dalhousie points out there was a buddy breathing system available, so that if one PLSS failed the astronaut could plug himself into his mate's PLSS and the two would use the water and oxygen. Obviously this would use the consumables faster, but it meant that they'd be able to get back to the LM for an abort...
...except if the rover had failed as well while they were at their walk-back limit, as this limit was based on the two astronauts both having their own supplies.
So in fact if the rover and one PLSS failed while the astronauts were a long way from the LM, then it was a non-survivable situation. Somewhat remarkably, given their insistence on safety margins in so many aspects of Apollo, NASA seems to have accepted this possibility. After all, it strikes me that an accident with the rover (say, a rollover on a hillside) would provide a distinct possibility of damaging a PLSS too.
Incidentally, if you read the ALSJ transcripts for the rover missions, you'll find that walk-back limit is often mentioned as the reason for getting the astronauts to wrap up their investigations of a site and moving on to the next site.