Re: [Gnu-arch-users] GCC v. Arch address@hidden: Regressions on mainline]

[Charles Duffy]

On Wed, 2004-06-23 at 02:24, Jan Hudec wrote:
> Things can still break when two people
> commit conflicting changes, which is fixed ex-post.

...as opposed to the case where the second change never gets committed
in the first place because the pre-commit hook rejects it. As long as
issues of compile time can be resolved, this is arguably the better
solution. I'll grant that it would require a rather substantial amount
of hardware to keep up with the commit rate -- but quite a few companies
have an interest in GCC development, one of my former employers among
them.