[PATCH, tree-dce]: DCE enh to handle dead calls (not pure)

[Steven Bosscher]

On Sat, May 17, 2008 at 11:44 AM, Steven Bosscher <stevenb.gcc@gmail.com> wrote:
> The idea is of this patch is really nice. But please try to avoid
> submitting patches with so many silly little issues!

Wow, you checked this in?!

I see this patch was checked in as r135463.  But as far as I can tell,
it was not approved.  Please revert it.

This was one of the poorest patch submissions I have ever seen.  The
quality of the patch itself was way below par and the whole submission
procedure was done incorrectly.

Your patch is not approved, and should not be approved in this state.
Even with the fixes, you still don't properly use the GCC coding
style, and you haven't addressed all comments like e.g. my
optimize_size_comment.

I do not understand why you get SVN write access if you:
a) do not understand the process of committing patches
b) submit patches of clearly inferior quality
c) recklessly commit patches without approval.

In the past, commit rights for the GCC repository were only given to
people who regularly submitted good patches, and who had shown with
prior patch submissions that they understood the GCC patch process.  I
understand that nowadays commit rights are given away much more
easily, but you have *never* submitted a patch before, and you clearly
don't understand the process either.  You just abused your commit
rights. If you don't know how things work, you should just not have
commit rights yet.

Diego, why do you sponsors people from your team if you are not sure
they deserve commit rights and know what is expected of them?

Please revert r135463, and also your follow-up r135493.

Gr.
Steven