Re: Any further work on superoptimizer?

In article 92-01-078 Mark Smotherman writes:
>Henry Massalin published a paper on a superoptimizer in ASPLOS II, 1987
>[1]. One important aspect of this work was the generation of *branchless*
>run-time library routines for simple functions (e.g., absolute value, max)
>that could be in-line substituted. Thus the overhead of procedure call
>and the problem of branch stalls / delay slot scheduling were avoided.


I don't remember if it was the same paper, but at about that time I read
an article about a "superoptimizer" that supposedly tried every possible
code sequence shorter than the original, and with a semantic analyzer
decided which if any effectively did the same thing. It would often come
up with surprising sequences using the XOR trick and other obscurities.
However, using no computer help, I coded up sequences that beat two of his
"superoptimized" ones. I sent the code-fragments to the author, but never
received a reply.


--

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