Re: g++.old-deja/g++.mike/p7325.C - suspected bogus test case

[Mike Stump <mrs at apple dot com>]

On Friday, June 25, 2004, at 11:00  PM, Zack Weinberg wrote:
> Test case g++.old-deja/g++.mike/p7325.C (appended to this message, for
> reference) appears to be invalid.

Yes, 7325 refers to (now old) cygnus bug reporting system.  Jason could 
probably confirm my memory or elaborate.

We want to ensure short life times of things that have short lifetimes 
and that we reuse the space quickly.  This is not an unreasonable thing 
to test or expect and it is important to test for such things.  I don't 
know of an easy or better way to test for this.  I suppose that a 
statistical approach might be better, essentially take the address of 
an automatic and just have many, many large objects and then check to 
see how we are doing on space.

I believe it used to work and pass, and someone broke it and didn't fix 
it.

You want to get rid of a test case that found a regression instead of 
enticing someone to fix the compiler?  This seems odd.  Usually, we 
find who introduced the regression and ask them to fix it.  I'd say the 
test case is doing its job.

Why not just fix the compiler?

I object to removing tests that actually reliably find regressions.