please clarify criteria for gcc-4.0.0 patches

[Benjamin Kosnik <bkoz at redhat dot com>]

Mark, this patch has just hit my mainline gcc tree.

http://gcc.gnu.org/ml/gcc-patches/2004-09/msg01650.html

I have some issue with this. For instance, this patch is not appropriate
for stage 3: you are clearly adding new functionality. In general, I
find the rationale for patch inclusion in gcc-4.0 increasingly bizarre:
patches fixing typos in comments are rejected, while whole new
functionality is being approved.

This is what I am using to understand the gcc dev process:
http://gcc.gnu.org/develop.html

Can you elaborate on what, exactly, you expect from gcc developers in
the coming months? Would it be possible to explain, in detail, what
kinds of patches should go in, and apply this uniformly to all gcc
developers at all participating companies?  IMHO, gcc mainline is a
mess. Do you have a plan for stabilizing gcc for stage 3? Would you be
willing to detail it?

The other part of your email I find most discouraging. It is, and I quote:

"Since we've decided to fix ABI issues for GCC 4.0"

I assume you mean arm-eabi only, correct? Can you point me at this
discussion please?

best,
benjamin