Pending GCC-3.1 upgrade - do or don't

Ian Molton spyro at armlinux.org
Tue May 7 17:10:54 PDT 2002


On Tue, 7 May 2002 16:17:30 -0700 (PDT)
Chad Simmons <polpak at yahoo.com> wrote:

> > 'If it aint broke, dont fix it' ?
> 
> It is 'broke' otherwise why would they be releasing a new version?
> The most obvious example of gcc 2.95.3 breakage is in it's processing
> of C++. We are therefore looking to move to the 'fixed' version which
> is the newer release. I'm sure that the gcc developers wouldn't have 
> spent all this time and energy on this new version, if the old one
> wasn't 'broke' in some way.

I wont disagree that 2.95.3 is 'broke', but we are talking degree.

3.x is currently (perhaps with the excpetion of c++) more broke than
2.95.3

now, when 3.x shapes up to being BETTER, rather than a bit 'less broke',
THEN we should switch, not before.

And before anyone says 'my 3.x built system 'feels' faster, let me
remind them that 'feels' is very dangerous. I remember once building a
pre-empt kernel, and being absolutely CONVINCED my system was running
'smoother than ever before', until I discovered I had never 'make
installed' it.

(it was actually smoother (demonstrably) after installing it.

'feels smoother' is a very dangerous thing.
-- 
Unsubscribe: send email to listar at linuxfromscratch.org
and put 'unsubscribe lfs-dev' in the subject header of the message



More information about the lfs-dev mailing list