Re: how to avoid a memset() optimization

Lars Duening wrote:
>
> Alex Colvin <alexc@world.std.com> wrote:
>
> > "Francis Wai" <fwai@rsasecurity.com> writes:
> >
> > ...the case of a memory scrub optimized away by the compiler...
> >
> > >Various suggestions have been made, such as declaring the variable
> > >volatile and having a scrub memory function in a file of its own. I'm
> > >wondering if there are better ways such as telling the compiler not to
> > >optimize away a function call.
> >
>
> #pragma eliminate_dead_code=no
> memset(key, 0, sizeof key);
> #pragma eliminate_dead_code=restore


Maybe I'm just silly, but isn't changing memset() so that it can be
optimized away just plain pointless?


The point of an automated optimizer is to perform optimizations that
can't be done in the source language, or that would be too ugly to do.
How often do programmers accidentally leave "dead" memset() calls in
their code, and is it worth the compiler's effort to remove these?


Maybe its to improve performance of badly generated code, which contains
a lot of dead memset() calls?


I suppose its possible that not just memset(), but a whole class of
functions are being targetted here. Perhaps even a solution like:


void *safe_memset(void *buf, int c, size_t len) { memset(buf, c, len); }


would be detected and removed?


CU
Dobes
[No need. See later messages. -John]

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