Re: [gentoo-user] gcc failed to compile, mpfr-4 compiled successfully...

2018-06-25 Thread Peter Humphrey
On Sunday, 24 June 2018 19:49:02 BST Rich Freeman wrote:
> On Sun, Jun 24, 2018 at 2:28 PM  wrote:
> > Since the unstable packages of today are the stable packages of
> > tommorrow one needs to bite the bullet (and hopefully this sentence
> > is not complete nonsense...I am no native speaker...not at all!)
> 
> Presumably if you're running stable on a host you probably wouldn't
> want to be biting the bullet on that particular host.
> 
> In any case, it seems likely that the news item shouldn't have been
> displayed to stable users, or should have had instructions pertaining
> to who it was targeted at.

Well, on my stable system I went and keyworded mpfr-4.0.1 and mpc-1.1.0-r1, 
then followed the instructions. I did that because I didn't want to find, 
later, that I'd been left behind by those versions becoming stable and not 
remembering the news item. I saw a good chance of inexplicable failures in 
that case.

> Of course, it shouldn't have been displayed for anybody considering it
> was never reviewed as required by GLEP 42.  Stuff like this is the
> reason these are supposed to be reviewed...

I'll leave those concerned to wrestle with that one...;)

-- 
Regards,
Peter.






Re: [gentoo-user] gcc failed to compile, mpfr-4 compiled successfully...

2018-06-25 Thread Arve Barsnes
On 25 June 2018 at 04:19,   wrote:
> I think, my attempt to play with words despite the fact that I am no
> native speaker ;) has spent some confusion (or I understand something
> wrong right now:
>
> In my make,conf this is set
>
> ACCEPT_KEYWORDS="~amd64"
>
> so I am on unstable.

The actual error, in case anyone knows the fix:

/var/tmp/portage/sys-devel/gcc-7.3.0-r3/work/gcc-7.3.0/libsanitizer/sanitizer_common/sanitizer_platform_limits_posix.cc:157:10:
fatal error:  sys/ustat.h: No such file or directory
 #include  

So maybe rebuild glibc or something?

Cheers,
Arve



Re: [gentoo-user] gcc failed to compile, mpfr-4 compiled successfully...

2018-06-24 Thread tuxic
On 06/24 07:01, Philip Webb wrote:
> 180624 Andreas K. Huettel wrote:
> > Am Sonntag, 24. Juni 2018, 09:01:53 CEST schrieb tu...@posteo.de:
> >> The bad news: mpfr-4 was installed after a successful compilation
> >> and gcc failed to compile.
> > This was a collective thinko of the toolchain team.
> > The news item should never have been shown like this to stable users,
> > who dont have to do anything now.
> 
> I read the news & immediately noticed that Mpfr-4 is still 'testing'.
> After a bit of reflection, as a stable user,
> I upgraded to Gcc-7, but left Mpfr + Mpc alone.
> Gcc-7 compiled successfully & I wb using it for next week's upgrade.
> 
> -- 
> ,,
> SUPPORT ___//___,   Philip Webb
> ELECTRIC   /] [] [] [] [] []|   Cities Centre, University of Toronto
> TRANSIT`-O--O---'   purslowatchassdotutorontodotca
> 
> 

Hi,

I think, my attempt to play with words despite the fact that I am no
native speaker ;) has spent some confusion (or I understand something
wrong right now:

In my make,conf this is set

ACCEPT_KEYWORDS="~amd64"

so I am on unstable.

Cheers
Meino





Re: [gentoo-user] gcc failed to compile, mpfr-4 compiled successfully...

2018-06-24 Thread Philip Webb
180624 Andreas K. Huettel wrote:
> Am Sonntag, 24. Juni 2018, 09:01:53 CEST schrieb tu...@posteo.de:
>> The bad news: mpfr-4 was installed after a successful compilation
>> and gcc failed to compile.
> This was a collective thinko of the toolchain team.
> The news item should never have been shown like this to stable users,
> who dont have to do anything now.

I read the news & immediately noticed that Mpfr-4 is still 'testing'.
After a bit of reflection, as a stable user,
I upgraded to Gcc-7, but left Mpfr + Mpc alone.
Gcc-7 compiled successfully & I wb using it for next week's upgrade.

-- 
,,
SUPPORT ___//___,   Philip Webb
ELECTRIC   /] [] [] [] [] []|   Cities Centre, University of Toronto
TRANSIT`-O--O---'   purslowatchassdotutorontodotca




Re: [gentoo-user] gcc failed to compile, mpfr-4 compiled successfully...

2018-06-24 Thread Andreas K. Huettel
Am Sonntag, 24. Juni 2018, 09:01:53 CEST schrieb tu...@posteo.de:
> Hi,
> 
> I think I am in trouble here...especially after reading the news
> item...
> 
> After the daily sunc routine, a new news items was announced which
> I read.
> 
> It saus, that the update to mpfr-4 could possibly break gcc. Therefore
> I had to recompile a combo of sources including mpfr and gcc.
> 
> The bad news: mpfr-4 was installed after a successful compilation and
> gcc failed to compile.
> 

This was a collective thinko of the toolchain team. The news item should never 
have been shown (like this) to stable users (who dont have to do anything 
now). 

Emergency instructions ** if you now have problems building any of these 
packages ** and have a stable system ** (untested, but not that much can be 
wrong):

1) Make sure in your config files that both mpfr and mpc are either keyworded 
stable or ~arch, not any mixed combination

2) rebuild in this order, with emerge -1a --ignore-built-slot-operator-deps y
  * mpfr 
  * mpc
  * gcc



-- 
Andreas K. Hüttel
dilfri...@gentoo.org
Gentoo Linux developer
(council, toolchain, perl, libreoffice, comrel)

signature.asc
Description: This is a digitally signed message part.


Re: [gentoo-user] gcc failed to compile, mpfr-4 compiled successfully...

2018-06-24 Thread Rich Freeman
On Sun, Jun 24, 2018 at 2:28 PM  wrote:
>
> Since the unstable packages of today are the stable packages of
> tommorrow one needs to bite the bullet (and hopefully this sentence
> is not complete nonsense...I am no native speaker...not at all!)
>

Presumably if you're running stable on a host you probably wouldn't
want to be biting the bullet on that particular host.

In any case, it seems likely that the news item shouldn't have been
displayed to stable users, or should have had instructions pertaining
to who it was targeted at.

Of course, it shouldn't have been displayed for anybody considering it
was never reviewed as required by GLEP 42.  Stuff like this is the
reason these are supposed to be reviewed...

-- 
Rich



Re: [gentoo-user] gcc failed to compile, mpfr-4 compiled successfully...

2018-06-24 Thread tuxic
On 06/24 11:12, Daniel Frey wrote:
> On 06/24/18 10:21, Dale wrote:
> > Plus it does say to include that info.  If it wasn't provided, it would
> > likely be asked for at some point unless someone has ran into that
> > specific problem and can recognize it with nothing but the bare info. 
> > 
> > My email program downloaded the whole thing but I don't hold it against
> > someone that is just following instructions.  I just wish I knew how to
> > help.  :/
> > 
> > Dale
> > 
> > :-)  :-) 
> > 
> 
> Mine downloaded the whole thing too, but I have 150 mbps symmetrical and
> it only took one second to load the message.
> 
> Meino, mpfr-4 is marked unstable/testing, is there a specific reason
> you're trying to use it?
> 
> You could also refer to:
> 
> https://wiki.gentoo.org/wiki/Mpfr4-update-guide
> 
> Failing that, you'll probably have to open a bug.
> 
> Dan
> 

Hi Dan,

thanks for your offered help!
In the past I run Gentoo stable with more headaches than
now running on "un"stable.

mpfr is compiling fine...thats not the problem.

sys-devel/gcc-7.3.0 compiled fine
sys-devel/gcc-7.3.0-r1 compiled fine
sys-devel/gcc-7.3.0-r3 compiled fine (Mon May 21 15:02:26 2018
  according to qlop -l

But now it fails in stage 3 as far as I can see for an unknown
(at least for me) reason.

I switched off any optimization in make.conf and tried again
but with no success.

Masking and waiting for the next update does not help since emerge
wants this package and stops any update until I unmask gcc it. That
gives me a nearly full three-stage compilation intro of gcc with
a grande finale: Failed with each sync.

H

Since the unstable packages of today are the stable packages of
tommorrow one needs to bite the bullet (and hopefully this sentence
is not complete nonsense...I am no native speaker...not at all!)

Cheers!
Meino







Re: [gentoo-user] gcc failed to compile, mpfr-4 compiled successfully...

2018-06-24 Thread Daniel Frey
On 06/24/18 10:21, Dale wrote:
> Plus it does say to include that info.  If it wasn't provided, it would
> likely be asked for at some point unless someone has ran into that
> specific problem and can recognize it with nothing but the bare info. 
> 
> My email program downloaded the whole thing but I don't hold it against
> someone that is just following instructions.  I just wish I knew how to
> help.  :/
> 
> Dale
> 
> :-)  :-) 
> 

Mine downloaded the whole thing too, but I have 150 mbps symmetrical and
it only took one second to load the message.

Meino, mpfr-4 is marked unstable/testing, is there a specific reason
you're trying to use it?

You could also refer to:

https://wiki.gentoo.org/wiki/Mpfr4-update-guide

Failing that, you'll probably have to open a bug.

Dan



Re: [gentoo-user] gcc failed to compile, mpfr-4 compiled successfully...

2018-06-24 Thread Dale
zless wrote:
> În ziua de duminică, 24 iunie 2018, la 11:55:52 EEST, Zoltán Kócsi a scris:
>> Hi,
>>
>>> Here are the outputs, which I should include if I need support
>>> according to emerge ;)
>> Still, if you didn't attach 9MB worth of stuff, that'd be appreciated.
>>
>> As a rough guess, the Gentoo list probably has at least 5 thousand
>> different subscribers, so your simple email generated some 45GB
>> outgoing traffic on the server.
>>
>> On a personal note, I'm in Australia where, thanks to our Liberal
>> government building a National Broadband Network that was outdated
>> about 10 years before its inception, we have the speed of a string and
>> tin cans, but without their reliability.
>>
> On a brighter note use KMail which will not download the attachments but
> will allow to read the message itself.
>
> I'm sure there are other well behaved email programs like this.
>

Plus it does say to include that info.  If it wasn't provided, it would
likely be asked for at some point unless someone has ran into that
specific problem and can recognize it with nothing but the bare info. 

My email program downloaded the whole thing but I don't hold it against
someone that is just following instructions.  I just wish I knew how to
help.  :/

Dale

:-)  :-) 



Re: [gentoo-user] gcc failed to compile, mpfr-4 compiled successfully...

2018-06-24 Thread zless
În ziua de duminică, 24 iunie 2018, la 11:55:52 EEST, Zoltán Kócsi a scris:
> Hi,
> 
> > Here are the outputs, which I should include if I need support
> > according to emerge ;)
> 
> Still, if you didn't attach 9MB worth of stuff, that'd be appreciated.
> 
> As a rough guess, the Gentoo list probably has at least 5 thousand
> different subscribers, so your simple email generated some 45GB
> outgoing traffic on the server.
> 
> On a personal note, I'm in Australia where, thanks to our Liberal
> government building a National Broadband Network that was outdated
> about 10 years before its inception, we have the speed of a string and
> tin cans, but without their reliability.
> 

On a brighter note use KMail which will not download the attachments but
will allow to read the message itself.

I'm sure there are other well behaved email programs like this.





Re: [gentoo-user] gcc failed to compile, mpfr-4 compiled successfully...

2018-06-24 Thread Zoltán Kócsi
Hi,

> Here are the outputs, which I should include if I need support
> according to emerge ;)

Still, if you didn't attach 9MB worth of stuff, that'd be appreciated.

As a rough guess, the Gentoo list probably has at least 5 thousand
different subscribers, so your simple email generated some 45GB
outgoing traffic on the server.

On a personal note, I'm in Australia where, thanks to our Liberal
government building a National Broadband Network that was outdated
about 10 years before its inception, we have the speed of a string and
tin cans, but without their reliability.

Regards,

Zoltan