Content deleted Content added
→Bytecode vs. machine code: In 1970s VM usually meant a VM under CP-67 or VM/370 |
→Bytecode vs. machine code: VAX, the ultimate byte code |
||
Line 102:
::'''Keep Separate''' — But this article needs to be refined and its category has to be determined. Is it related to virtual machines, interpreters, or process virtual machines? The page on VMs has to explain the type of code --[[User:Melab-1|Melab±1]] [[User_talk:Melab-1|☎]] 22:45, 19 June 2010 (UTC)
:::'''Keep Separate''' The term '''Bytecode''' has been in use ever since its [[canonical]] example (or maybe even earlier) of a byte-code in the form of the [[p-code machine|p-code]] used by [[UCSD Pascal]], which was one of the [[IBM_PC#OS_support|contenders]] for an operating system for the upcoming [[IBM PC]]. So at the very least the term has historical significance. [[User:Mahjongg|Mahjongg]] ([[User talk:Mahjongg|talk]]) 17:52, 21 June 2010 (UTC)
::::
::::This is a little strange. I don't know that anything deserves to be called bytecode more than VAX machine code. (And there are plenty of software implementations of VAX.) But also, I suspect the reason VAX went away, was that it uses a bytecode. VAX instructions can have from one to at least six operands. Each operand has a byte indicating the address mode for that argument. Depending on the address mode, that byte is followed by an appropriate number of bytes for that mode. It might be one to 60 bytes for an instruction. Like JVM code, it is well designed for processing one byte at a time. It is, however, extremely difficult to decode instructions in parallel. Very convenient for the microcoded VAX machines. The important distinction, then, is that it is convenient for processing one byte at a time, based on the state determined by earlier bytes. And especially, a large number of different modes depending on those bytes. [[User:Gah4|Gah4]] ([[User talk:Gah4|talk]]) 09:23, 30 November 2022 (UTC)
== Bytecode v8 ==
| |||