Programming is great because your mental model is a directed acyclic graph of data flow, but you write code that is sequential and linear, which the compiler converts into a DAG so it can write efficient sequential machine code, which the CPU converts into a DAG it can execute
OK this tweet kind of took off.
So, joking aside, a natural question to ask is whether this switching back and forth is a waste of everybody's time
The compiler DAG -> CPU DAG step makes a lot more sense if you view machine code as simply a graph serialization format.
So, joking aside, a natural question to ask is whether this switching back and forth is a waste of everybody's time
The compiler DAG -> CPU DAG step makes a lot more sense if you view machine code as simply a graph serialization format.
Registers are temporary labels that we assign to outputs from graph nodes.
"Instruction scheduling" is the process of deciding what piece of the graph to write next, so that the CPU deserializes the right piece at the right time.
"Instruction scheduling" is the process of deciding what piece of the graph to write next, so that the CPU deserializes the right piece at the right time.
Higher up the stack: the programmer head => code-in-editor gap
Programming languages are UIs for compilers. It strikes me that humanity may be capable of compiler UX better than "edit a bunch of text files, and get your punctuation right, damn it". That remains to be seen though
Programming languages are UIs for compilers. It strikes me that humanity may be capable of compiler UX better than "edit a bunch of text files, and get your punctuation right, damn it". That remains to be seen though
