Possible optimization bug on i386

[Eric Botcazou]

> By this can I assume that my report was clear and unambiguous enough for it to 
> be useful to the development team?

Yes, you seem to have already narrowed down the problem to a small chunk 
of assembly code, which can take a significant amount of time. We now 
need a preprocessed testcase, even a not too small one provided that you 
can pinpoint the lines in the source code that produce the 
aforementioned chunk of assembly. And *specific* GCC version numbers 
(3.x is too vague) and configuration options as reported by 'gcc -v'. 
Then fill in a report with GNAT and I'll take a look at it.

> I don't see the point in reporting a bug that is not clearly defined enough for it to be investigated.

I wish every bug reporter made this reasoning!

-- 
Eric Botcazou