Re: m68k structure packing

[Jim Wilson <wilson at cygnus dot com>]

I think you are suffering from some misconceptions here.

	I don't think we should force authors of code to prove to kenner that
	a change is good enough for the FSF's compiler, and then force them to
	go through the pain again to get it into the egcs compiler.

We don't.  Code that is accepted by Kenner into the FSF's compiler is
automatically accepted into EGCS unless there is something obviously wrong
with it.  In this case, there was something wrong with your patch that was
obvious to me but which Kenner apparently missed.  I don't think it is
unreasonable for me to point out problems which Kenner has apparently missed.

It turns out that it was a documentation problem and not a problem with your
patch, but the fact remains that there was a real problem here that needed to
be addressed.

	I think
	we risk faster divergence that way, and keeping them converged is
	better.

There was no chance of a divergence here.  If there was a problem with your
patch, I would have reported it to Kenner so that gcc2 and egcs could stay
in sync.

	I thought that kenner would be generally be harder to deal
	with, and while I may still generally think that is true, it wasn't in
	this case.

I am trying to be cooperative.  I realized that you didn't fully understand
what I wanted to see, so as soon as I could I spent several hours investigating
the problem myself and I posted the results of what I found to the list.
It was only 24 hours after I brought up the issue before I resolved it.
I don't think that 24 hours is an unresonable delay for this patch.

Jim