Re: how to avoid a memset() optimization

I have little insight to add to this discussion, but I do wish to
reply to something Charles Bryant wrote:
> In fact, it must be the linker which does this, since at compilation
> time you might not have written clrmem(). And even if such a linker
> conspires with a compiler to eliminate dead code, you can always
> write clrmem() in assembly.


The MIPS compiler suite had exactly this behavior. The compiler spit
out "intermediate" code that the linker linked, optimized (including
dead-code elimination and value propagation), and then translated into
machine language. Not only did the MIPS (now SGI) compilers use this
technology, but the early Unix compilers for DEC Alpha's used it too.
(I don't know how the current Compaq compilers deal with this issue.)


Later the DEC group I consulted for(*), took this apporach even further,
and wrote "OM" (object modifier) which took executable images (of
machine language instructions) and optimized them. Thus, when OM was
used to eliminate your dead-code, there was no hope (unless you wanted
to construct your dead code in the data segment and jump into it),
since even assembly language routines were susceptible to its reach.


(*Technically, the development of OM was done by DEC WRL (Western
Reasearch Labs), but the Unix group released it as part of their
standard compiler suite.)


Hope this helps,
-Chris


*****************************************************************************
Chris Clark Internet : compres@world.std.com
Compiler Resources, Inc. Web Site : http://world.std.com/~compres
3 Proctor Street voice : (508) 435-5016
Hopkinton, MA 01748 USA fax : (508) 435-4847 (24 hours)

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