Re: [ILUG] how does a gcc version effect the kernel build ?

From: Rick Moen (rick at domain linuxmafia.com)
Date: Tue 09 Jul 2002 - 05:17:49 IST


Quoting Brian O'Donoghue (Brian.ODonoghue at domain kbs.ie):

> Not using Red Hat I can't comment... there was some sort of fiasco
> where Red Hat released their own unsupported version of gcc.... 2.96
> they called it or something.

For those who didn't follow the flap at that time:

It was a temporary fork of gcc, built that way for excellent,
justifiable reasons. After all the hotheads had spoken and cooler heads
emerged, about the only questionable part of what they did was calling
it "gcc-2.96" without a qualifier making clear that this was _not_ an
official release from the GCC Steering Committee.

Essentially (eliding a ghastly amount of detail), Red Hat made the
(sound) judgement call that C++ support in the standard gcc codebase was
broken, and needed to be pushed (goaded) towards a reasonable long-term
solution. Their gcc 2.96 fork introduced much more stable C++ support
with a good binary object interface -- which was unavoidably
incompatible with both 2.95's and the projected 3.0 one. This was very
arguably a step forward despite immediate drawbacks. My view is that
they deserve not only support but also gratitude: The 3.x series came
sooner and better, as a result.

But anyone wanting to install additional (or different) versions can
do so trivially, anyway.

-- 
Cheers,            There are only 10 types of people in this world -- 
Rick Moen          those who understand binary arithmetic and those who don't.
rick at domain linuxmafia.com


This archive was generated by hypermail 2.1.6 : Thu 06 Feb 2003 - 13:17:44 GMT