On Sat, 6 Oct 2001, Gregor N. Purdy wrote: > After the bytecode is loaded, but before it is executed, put it through > a stage of processing that requires about as much information as a > disassembler would (which is why my op_info stuff from one of my previous > patches is required). > > This process converts opcodes into pointers to the op functions, and > arguments to pointers to the constant values or register entries. This > means that we amortize the dereferences over all invocations of the > op at each PC, which when tight loops are involved should make for > noticable savings.
It seems that this would interfere with trying to share the bytecode using mmap(). We might gain a small performance increase but lose a significant advantage in memory reduction in large systems. Of course, if it's configurable then people can make that trade-off for themselves. -sam
