Re: Optimizing in assembly language

> Transmeta must have solved the problem of 'morphing data tables by
> mistake':-) but I have not seen reports of how they cope with the most


> paranoid self modifying code often found in dongles ?


Both solutions are simple:
1) the code looks at x86 data, not at translated data, nobody knows
about translated data in fact. And the morpher never morphs
unreachable paths (being unreachable is a transitive relationship).


2) paranoid self modifying code will be slow; Transmeta says the
morpher will spend less time working on paranoid code, possibly
by interpreting it instead of compiling it. The problem is, that JITs
often create code that the morpher will think of as self modifying
(for example JITs that cache native code as there is need to run
it, instead of compiling everything at startup)

Return to the comp.compilers page.
Search the comp.compilers archives again.