Mailing List Archive

On Sunday 12 September 2004 01:30 pm, Andrew Gaydenko wrote:
> Hi,
>
> 'emerge -pvDu world' tells me I must move to gcc 3.4.1 (I have 3.3.4
> installed now). Does it mean gcc 3.4.1 is official now for ~x86?
> I have seen multiple problems with 3.4.x in different mail lists.
>
>
> Andrew
>

I haven't seen anything official yet, but since it compiles into a slot, it's
ok to install. JUST BE SURE that you use gcc-config to pickout your desired
compiler, BEFORE you compile something important. :')
--

******************************************************************************
Registered Linux User Number 185956
FSF Associate Member number 2340 since 05/20/2004
Join me in chat at #linux-users on irc.freenode.net
Buy an Xbox for $149.00, run linux on it and Microsoft loses $150.00!
6:58pm up 12 days, 19:11, 6 users, load average: 1.92, 1.75, 1.71

--
gentoo-user@gentoo.org mailing list

Andrew Gaydenko wrote:
> I have seen multiple problems with 3.4.x in different mail lists.

Many are fixed in portage. Actually, I've been using 3.3.4/3.4.2 for sometime
now, but since 3.4 just got unmasked in ~x86, I've unemerged gcc-3.3.4.

I have a pure gcc-3.4.2 system now. Of course, YMMV.

Best regards,
Norberto

--
gentoo-user@gentoo.org mailing list

Jerry McBride wrote:

> I haven't seen anything official yet, but since it compiles into a slot, it's
> ok to install. JUST BE SURE that you use gcc-config to pickout your desired
> compiler, BEFORE you compile something important. :')

And be sure to re-login, so that the changes to the environment
get picked up :) This took me a full 5 minutes yesterday to realize *G*

PS: Jerry, why are you using koi8-r charset, when all your characters
fit to at least iso-8859-1 or even us-ascii?

Alexander Skwar
--
<_Anarchy_> Argh.. who's handing out the paper bags 8)
¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯


--
gentoo-user@gentoo.org mailing list

Alexander Skwar wrote:
> Jerry McBride wrote:
> > I haven't seen anything official yet, but since it compiles into a slot,
> > it's ok to install. JUST BE SURE that you use gcc-config to pickout your
> > desired compiler, BEFORE you compile something important. :')
>
> And be sure to re-login, so that the changes to the environment
> get picked up :) This took me a full 5 minutes yesterday to realize *G*

Nah, simply do:

$ . /etc/profile


--
gentoo-user@gentoo.org mailing list

Norberto Bensa wrote:
> Alexander Skwar wrote:
>
>>Jerry McBride wrote:
>>
>>>I haven't seen anything official yet, but since it compiles into a slot,
>>>it's ok to install. JUST BE SURE that you use gcc-config to pickout your
>>>desired compiler, BEFORE you compile something important. :')
>>
>>And be sure to re-login, so that the changes to the environment
>>get picked up :) This took me a full 5 minutes yesterday to realize *G*
>
>
> Nah, simply do:
>
> $ . /etc/profile

Ja, sicher, but you'll need to that in each and every shell that
you start, don#t you?

Alexander Skwar
--
"I go on working for the same reason a hen goes on laying eggs."
- H. L. Mencken
¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯


--
gentoo-user@gentoo.org mailing list

Alexander Skwar wrote:
> Norberto Bensa wrote:
> > $ . /etc/profile
>
> Ja, sicher, but you'll need to that in each and every shell that
> you start, don#t you?
>

Of course. But it's faster than logout/login ;)

--
gentoo-user@gentoo.org mailing list

Norberto Bensa wrote:
> Alexander Skwar wrote:
>
>>Norberto Bensa wrote:
>>
>>> $ . /etc/profile
>>
>>Ja, sicher, but you'll need to that in each and every shell that
>>you start, don#t you?
>>
>
> Of course. But it's faster than logout/login ;)

Yeah, it surely is, I grant you that *G*

Alexander Skwar
--
Strategy:
A long-range plan whose merit cannot be evaluated until sometime
after those creating it have left the organization.
¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯


--
gentoo-user@gentoo.org mailing list

> Many are fixed in portage. Actually, I've been using 3.3.4/3.4.2 for sometime
> now, but since 3.4 just got unmasked in ~x86, I've unemerged gcc-3.3.4.
>
> I have a pure gcc-3.4.2 system now. Of course, YMMV.
>

I just did an emerge syc, and I'm not seeing 3.4 unmasked:

oberon portage # emerge -p gcc

These are the packages that I would merge, in order:

Calculating dependencies ...done!
[ebuild R ] sys-devel/gcc-3.3.4-r1

Any ideas?
--
Chris Bare
chris@bareflix.com

--
gentoo-user@gentoo.org mailing list

It's for ~x86.

Andrew

======= On Wednesday 15 September 2004 02:45, Chris Bare wrote: =======
> Many are fixed in portage. Actually, I've been using 3.3.4/3.4.2 for sometime
> now, but since 3.4 just got unmasked in ~x86, I've unemerged gcc-3.3.4.
>
> I have a pure gcc-3.4.2 system now. Of course, YMMV.
>

I just did an emerge syc, and I'm not seeing 3.4 unmasked:

oberon portage # emerge -p gcc

These are the packages that I would merge, in order:

Calculating dependencies ...done!
[ebuild R ] sys-devel/gcc-3.3.4-r1

Any ideas?
--
Chris Bare
chris@bareflix.com

--
gentoo-user@gentoo.org mailing list



--
gentoo-user@gentoo.org mailing list

On Tue, 14 Sep 2004, Chris Bare wrote:

> > Many are fixed in portage. Actually, I've been using 3.3.4/3.4.2 for sometime
> > now, but since 3.4 just got unmasked in ~x86, I've unemerged gcc-3.3.4.
> >
> > I have a pure gcc-3.4.2 system now. Of course, YMMV.
> >
>
> I just did an emerge syc, and I'm not seeing 3.4 unmasked:
>
> oberon portage # emerge -p gcc
>
> These are the packages that I would merge, in order:
>
> Calculating dependencies ...done!
> [ebuild R ] sys-devel/gcc-3.3.4-r1
>
> Any ideas?

And I'm using the gcc34-x86-2004.2 profile and already upgraded to
gcc-3.4.2-r1 as it is ~x86 too.

But after emerge sync today portage wants to downgrade it again to
3.4.1-r2.

According to the ebuild, 3.4.2-r2 is ~x86 too... What gives?

--
T.G.

--
gentoo-user@gentoo.org mailing list

On Wed, 15 Sep 2004, Peter Ruskin wrote:

> On Tuesday 14 September 2004 23:56, Tero Grundström wrote:
> > And I'm using the gcc34-x86-2004.2 profile and already upgraded to
> > gcc-3.4.2-r1 as it is ~x86 too.
> >
> > But after emerge sync today portage wants to downgrade it again to
> > 3.4.1-r2.
> >
> > According to the ebuild, 3.4.2-r2 is ~x86 too... What gives?
>
> It seems to have been masked. If you want to keep it, add this
> to /etc/portage/package.unmask:
> =sys-devel/gcc-3.4.2-r1

Sure I'll keep it - cannot see why this bug fix release would be more
unstable than 3.4.1, which has been very stable this far.

Anyways this kind of mistakes are plain wrong even on ~x86 IMHO.

Thanks

--
T.G.

--
gentoo-user@gentoo.org mailing list

On 04/09/15(Wed) 04:32, Tero Grundström wrote:
> On Wed, 15 Sep 2004, Peter Ruskin wrote:
> > On Tuesday 14 September 2004 23:56, Tero Grundström wrote:
> > > According to the ebuild, 3.4.2-r2 is ~x86 too... What gives?
> >
> > It seems to have been masked. If you want to keep it, add this
> > to /etc/portage/package.unmask:
> > =sys-devel/gcc-3.4.2-r1
>
> Sure I'll keep it - cannot see why this bug fix release would be more
> unstable than 3.4.1, which has been very stable this far.
>
> Anyways this kind of mistakes are plain wrong even on ~x86 IMHO.

Excerpt from /usr/portage/profiles/package.mask:

# <lv@gentoo.org> (14 Sep 2004)
# This release has so many issues it's not even funny. With the insanity
# caused by unmasking gcc 3.4 on x86, I want to make sure the bugs fixed
# are actual gcc 3.4 bugs, and not just "ebuild rewrite for 3.4.2" bugs.
>=sys-devel/gcc-3.4.2


Patrick Börjesson

--
Public key ID: 4C5AB0BF
Public key available at wwwkeys.pgp.net

On Wed, 15 Sep 2004, Patrick Börjesson wrote:

> On 04/09/15(Wed) 04:32, Tero Grundström wrote:
> > Sure I'll keep it - cannot see why this bug fix release would be more
> > unstable than 3.4.1, which has been very stable this far.
> >
> > Anyways this kind of mistakes are plain wrong even on ~x86 IMHO.
>
> Excerpt from /usr/portage/profiles/package.mask:
>
> # <lv@gentoo.org> (14 Sep 2004)
> # This release has so many issues it's not even funny. With the insanity
> # caused by unmasking gcc 3.4 on x86, I want to make sure the bugs fixed
> # are actual gcc 3.4 bugs, and not just "ebuild rewrite for 3.4.2" bugs.
> >=sys-devel/gcc-3.4.2

Well this makes the mistake even worse...

--
T.G.

--
gentoo-user@gentoo.org mailing list

I have several problems with the "new" gcc.
Compiling with distcc failed.
Mozilla, can't compile. php5.0.1, giflib, also failed.
I'm getting angry, because I want to set up my epia 800 mobo as server,
but it takes a very long time :/

Maybe all my problems depends on gcc. It's not so funny.
But, that's life :))))))
I have time to play my children :)))))))))

István





2004-09-15, sze keltezéssel 17:46-kor Tero Grundström ezt írta:
> On Wed, 15 Sep 2004, Patrick Börjesson wrote:
>
> > On 04/09/15(Wed) 04:32, Tero Grundström wrote:
> > > Sure I'll keep it - cannot see why this bug fix release would be more
> > > unstable than 3.4.1, which has been very stable this far.
> > >
> > > Anyways this kind of mistakes are plain wrong even on ~x86 IMHO.
> >
> > Excerpt from /usr/portage/profiles/package.mask:
> >
> > # <lv@gentoo.org> (14 Sep 2004)
> > # This release has so many issues it's not even funny. With the insanity
> > # caused by unmasking gcc 3.4 on x86, I want to make sure the bugs fixed
> > # are actual gcc 3.4 bugs, and not just "ebuild rewrite for 3.4.2" bugs.
> > >=sys-devel/gcc-3.4.2
>
> Well this makes the mistake even worse...
>
> --
> T.G.
>
> --
> gentoo-user@gentoo.org mailing list
>
>


--
gentoo-user@gentoo.org mailing list

On 04/09/15(Wed) 19:55, Mac wrote:
> I have several problems with the "new" gcc.
[snip]
> I'm getting angry, because I want to set up my epia 800 mobo as server,
> but it takes a very long time :/

One recommendation would be to _not_ use ~x86 if you want stable
software (as you should on a server)...


Patrick Börjesson

--
Public key ID: 4C5AB0BF
Public key available at wwwkeys.pgp.net

begin quote
On Thu, 16 Sep 2004 00:14:31 +0200
Patrick Börjesson <psycho@rift.ath.cx> wrote:

> On 04/09/15(Wed) 19:55, Mac wrote:
> > I have several problems with the "new" gcc.
> [snip]
> > I'm getting angry, because I want to set up my epia 800 mobo as
> > server,
> > but it takes a very long time :/
>
> One recommendation would be to _not_ use ~x86 if you want stable
> software (as you should on a server)...


Let it break, its educational.

//S

--
begin .signature
Tortured users / Laughing in pain
See Microsoft KB Article Q265230 for more information.
end

On Thu, 16 Sep 2004, Patrick Börjesson wrote:

> On 04/09/15(Wed) 19:55, Mac wrote:
> > I have several problems with the "new" gcc.
> [snip]
> > I'm getting angry, because I want to set up my epia 800 mobo as server,
> > but it takes a very long time :/
>
> One recommendation would be to _not_ use ~x86 if you want stable
> software (as you should on a server)...

Oh really? I'm tired of you smart asses saying not to use ~arch. I'm sure
everybody running ~arch knows the risks.

Everybody who runs ~arch is testing software for *you* too. If nobody
would test them how would you softballs know it's stable?

However sometimes software is brought to ~arch too soon like in this case
- how could you run a testing system as a user if the software breaks your
system?

--
T.G.

--
gentoo-user@gentoo.org mailing list

> Oh really? I'm tired of you smart asses saying not to use ~arch. I'm sure
> everybody running ~arch knows the risks.
>
> Everybody who runs ~arch is testing software for *you* too. If nobody
> would test them how would you softballs know it's stable?

Ouch! But,understandable. I never heard the term *softballs* before,
but, my wife, an accomplished firmware engineer, often refers to many
engineers/cs as having *paper_balls*. According to her, those are the
engineers/cs that just fold their manhood up so as to not get in the
way....

--still laughing....

James





--
gentoo-user@gentoo.org mailing list

On Thu, 16 Sep 2004 13:02:56 +0300 (EEST) Tero Grundström
<tero@vuosaari.hai.fi> wrote:
| Oh really? I'm tired of you smart asses saying not to use ~arch. I'm
| sure everybody running ~arch knows the risks.

Nothing wrong with ~arch, so long as you don't use it on a box that has
to stay up reliably.

| However sometimes software is brought to ~arch too soon like in this
| case- how could you run a testing system as a user if the software
| breaks your system?

Expect ~arch to be totally broken sometimes. Unfortunately, our QA on
~arch has been far too good lately, so some people are getting the idea
that it's fine to run it on production kit...

--
Ciaran McCreesh : Gentoo Developer (Sparc, MIPS, Vim, Fluxbox)
Mail : ciaranm at gentoo.org
Web : http://dev.gentoo.org/~ciaranm

On Thursday 16 September 2004 10:13 am, Ciaran McCreesh wrote:
> On Thu, 16 Sep 2004 13:02:56 +0300 (EEST) Tero Grundström
>
> <tero@vuosaari.hai.fi> wrote:
> | Oh really? I'm tired of you smart asses saying not to use ~arch. I'm
> | sure everybody running ~arch knows the risks.
>
> Nothing wrong with ~arch, so long as you don't use it on a box that has
> to stay up reliably.
>
> | However sometimes software is brought to ~arch too soon like in this
> | case- how could you run a testing system as a user if the software
> | breaks your system?
>
> Expect ~arch to be totally broken sometimes. Unfortunately, our QA on
> ~arch has been far too good lately, so some people are getting the idea
> that it's fine to run it on production kit...

"Getting the idea"??? I've been doing it for quite some time and I have to
agree with your comment about Q&A. It is quite good. Thank you.

About the only thing I keep a close tab on is GCC. I allow all of the 3.3.x
updates through, but the 3.4.x stuff I keep on a development box in my
office. Probably the only thing that keeps me from going to 3.4.x is that a
few of the sources won't build with it. Once the source catches up with the
compiler... I'll move it up too.


Cheers.


--

******************************************************************************
Registered Linux User Number 185956
FSF Associate Member number 2340 since 05/20/2004
Join me in chat at #linux-users on irc.freenode.net
Buy an Xbox for $149.00, run linux on it and Microsoft loses $150.00!
11:00am up 16 days, 11:13, 6 users, load average: 2.72, 2.90, 2.95

--
gentoo-user@gentoo.org mailing list

On Thu, 16 Sep 2004 11:05:36 -0400 Jerry McBride <mcbrides9@comcast.net>
wrote:
| > Expect ~arch to be totally broken sometimes. Unfortunately, our QA
| > on~arch has been far too good lately, so some people are getting the
| > idea that it's fine to run it on production kit...
|
| "Getting the idea"??? I've been doing it for quite some time and I
| have to agree with your comment about Q&A. It is quite good. Thank
| you.

Yeah, and that's a problem. ~arch *does* break every now and again (see,
for example, the first bash 3 release), and people assuming that it's
safe because we don't usually break things too badly leads to people
getting veeeerrry upset when ~arch does break.

Not that I'm saying avoid ~arch, or anything like that. Just don't use
it where you're not prepared to have bugs which could totally screw up
your system. Also remember that it's very easy to selectively run ~arch
via package.keywords, which may be a better option for many people.

Summary: if something in ~arch breaks your system, don't expect too much
sympathy :)

--
Ciaran McCreesh : Gentoo Developer (Sparc, MIPS, Vim, Fluxbox)
Mail : ciaranm at gentoo.org
Web : http://dev.gentoo.org/~ciaranm

This is a home server for my own, behind a firewall.
I separate the workstation and the server functions between two machines
(my wife must use windowser, some file/scanner/printer/internet sharing
and some other stuff.

By the way, what is the challenge in x86? :))))))))))


2004-09-16, cs keltezéssel 04:03-kor Spider ezt írta:
> begin quote
> On Thu, 16 Sep 2004 00:14:31 +0200
> Patrick Börjesson <psycho@rift.ath.cx> wrote:
>
> > On 04/09/15(Wed) 19:55, Mac wrote:
> > > I have several problems with the "new" gcc.
> > [snip]
> > > I'm getting angry, because I want to set up my epia 800 mobo as
> > > server,
> > > but it takes a very long time :/
> >
> > One recommendation would be to _not_ use ~x86 if you want stable
> > software (as you should on a server)...
>
>
> Let it break, its educational.
>
> //S


--
gentoo-user@gentoo.org mailing list

On Thu, 16 Sep 2004, Ciaran McCreesh wrote:

> Summary: if something in ~arch breaks your system, don't expect too much
> sympathy :)

With all due respect Ciaran, those running ~arch *deserve* at least some
sympathy. They do, after all, take the risks to verify that packages are
stable. They are also the ones who report most of the bugs and also help
devs to fix them before the packages are marked stable. Normal users can
then, thanks to these testers, just enjoy the fruits with the comfort of
knowing the fruits are good.

Well.. But I refuse to believe that Gentoo devs don't appreciate the value
of testers...

--
T.G.

--
gentoo-user@gentoo.org mailing list

On Fri, 17 Sep 2004 00:59:44 +0300 (EEST) Tero Grundström
<tero@vuosaari.hai.fi> wrote:
| On Thu, 16 Sep 2004, Ciaran McCreesh wrote:
|
| > Summary: if something in ~arch breaks your system, don't expect too
| > much sympathy :)
|
| With all due respect Ciaran, those running ~arch *deserve* at least
| some sympathy. They do, after all, take the risks to verify that
| packages are stable. They are also the ones who report most of the
| bugs and also help devs to fix them before the packages are marked
| stable. Normal users can then, thanks to these testers, just enjoy the
| fruits with the comfort of knowing the fruits are good.
|
| Well.. But I refuse to believe that Gentoo devs don't appreciate the
| value of testers...

Oh, we do, just not people who test really flaky stuff on production
systems.

--
Ciaran McCreesh : Gentoo Developer (Sparc, MIPS, Vim, Fluxbox)
Mail : ciaranm at gentoo.org
Web : http://dev.gentoo.org/~ciaranm