[mythtv] gcc3.2 + mythTV = trouble...

[Lonnie Borntreger]

On Mon, 2002-12-09 at 09:35, Chris Liscio wrote:
> I may continue using gcc 3.2 in the end, but right now I'd be happy to
> find out if I can modulate easily between the two compilers (on *any*
> distro) to figure out what bugs get caused by gcc3.2, and what issues
> are actual bugs that may be important to *this* list.  I can't really
> provide all that much help if I keep making shots in the dark assuming
> all weird bugs are compiler issues.  :)

It's very easy to switch between using 3.2 and 2.9 compilers - as long
as you're not talking about c++, and then you need to read Matt's post
(I won't repeat it).  Of course that doesn't help you with mythtv.

> Besides, nobody here (and correct me if I'm wrong) really wants to be
> chasing down bugs that are caused by a compiler that isn't really
> considered 100% complete/stable by the majority of the install base
> out there.  But that's just my own opinion.  Rest assured that grains
> of salt are readily available to accompany all my opinions.  :D

You're making an assumption on where the bug lives.  "sleep" is so
common that I doubt it is a bug in the compiler, I would think that
someone would have caught it long ago.  Anyway, I wouldn't consider 3.2
any less complete than 2.9 (it still continues to get fixes), and AFAIK
all the major distributions are using it in their releases.  It would
make more sense to identify the cause and get it fixed (compiler or
not), then trying to work around it by using another compiler version
that many/most people will not have available.

My two cents.
Lonnie Borntreger