state of 3.2.1-pre: how far from release?

[Joe Buck]

> > In any case: Mark, what's your thinking?  Are there other bugs that you
> > consider blockers?
> 
> That's a good question, and one I was just looking at.  I am concerned
> about 8453 which is a silent code-gen failure in C++, and 8467 which
> is a silent code-gen failure with sibcalls, and 5351 which is a
> silent code-gen failure with structure passing.

I didn't have 8467 in my list, it has just been raised to "high".
Could it have the same underlying causes as 8362?

Also, I have just verified that a simple patch from HJ fixes 5351 (see his
earlier message of today on gcc).  Also, 7591 is a duplicate of 5351, so
at least two folks ran into it.

> I am also concerned about 8146, the bootstrap failure with GCC 2.95.3.  Do
> we yet have information about what is causing the problem?

This bug also appears to be an i586/i686 difference, suggesting that it
might be the same issue as 5351.  (that is, folks report that if gcc 3.2
is told to tune for i586 the bug does not occur).  I plan to test as well
to see if HJ's patch addresses that one.  In any case, I'm asking you
and/or RTH to look at HJ's patch again.  Evidently 5351 does not occur on
the trunk, so maybe even if you're not thrilled about the concept behind
HJ's patch it might be well worth doing on the branch assuming no new
regressions (I'll run tests on that score soon).

> I don't see that we have many options; people are more interested in
> working on BIB than 3.2.1, I have cajoled about as much as I am able,
> and I have no other tools at my disposal.

You have one tool at your disposal: you could set a deadline.  As I've
said, even though 3.2.1 still contains nasties it appears to have far
fewer nasties than 3.2.

> If the RS6000 maintainers do not want to do that, then David's
> work-around for abi_check.cc is OK.
> 
> I would like the RS6000 maintainers to decide one way or the other,
> and check in one or the other approach ASAP.

Agreed.