[e-lang] GridLock

[Mark S. Miller]

Mark S. Miller wrote:
> [...] we can explain the loss of progress of a deadlocking 
> program within its scarcity-free computational model. We cannot so explain the 
> loss of progress of a gridlocking program.

My apologies, my explanation here is confused and confusing.

Within a scarcity-free computational model, one can of course model 
scarcity-bounds, and so model and explain gridlock. For example, one can 
introduce an explicit reservoir of a limited number of tokens to model some 
actual scarce resource, and so model gridlocking behavior.

More to the point, in a scarcity-free computational model, one can choose not 
to model scarcity if one wishes, and so provide an account of stuck-free 
behavior of programs that have gridlock bugs. Of course, the scarcity-free 
assumptions of these models are probably[1] counter-factual, so these programs 
still suffer from lost-progress bugs when run on realizable hardware.

[1] http://www.aleph.se/Trans/Global/Omega/dyson.txt

-- 
Text by me above is hereby placed in the public domain

     Cheers,
     --MarkM