[gentoo-dev] gcc-3.3.5.20050110-r1 for stable

[gentoo-dev] gcc-3.3.5.20050110-r1 for stable

[Mike Frysinger]

can stable uses of gcc-3.3.5-r1 upgrade to gcc-3.3.5.20050110-r1 and see if 
they hit any fun and exciting bugs ?

if not i'd like to move this to stable this weekend
-mike
--
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] gcc-3.3.5.20050110-r1 for stable

[Dan Meltzer]

One thing... Maybe its just me... or maybe they are in no way related,
but I seem to have heard of a lot more 'libtool' problems when using a
snapshot version instead of a regularly numbered version, is there a
reason?

On Apr 7, 2005 11:46 PM, Mike Frysinger <vapier@gentoo.org> wrote:
> can stable uses of gcc-3.3.5-r1 upgrade to gcc-3.3.5.20050110-r1 and see if
> they hit any fun and exciting bugs ?
> 
> if not i'd like to move this to stable this weekend
> -mike
> --
> gentoo-dev@gentoo.org mailing list
> 
>
--
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] gcc-3.3.5.20050110-r1 for stable

[Ferris McCormick]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Thu, 7 Apr 2005, Mike Frysinger wrote:

> can stable uses of gcc-3.3.5-r1 upgrade to gcc-3.3.5.20050110-r1 and see if
> they hit any fun and exciting bugs ?
>

Uh, there isn't any such thing.  If you mean this:
Mon Mar 21 14:05:58 2005 >>> sys-devel/gcc-3.3.5.20050130-r1
it's been stable on sparc for 2.5 weeks.

Or did you mean this instead?
gcc-3.4.3.20050110-r1

> if not i'd like to move this to stable this weekend
> -mike
> --
> gentoo-dev@gentoo.org mailing list
>
>
Regards,
Ferris
- --
Ferris McCormick (P44646, MI) <fmccor@gentoo.org>
Developer, Gentoo Linux (Sparc)
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.1 (GNU/Linux)

iD8DBQFCVmx6Qa6M3+I///cRAsAoAKDiytqoV6v+6rqTucJVkPTBvgzgrQCfXhM0
QQlUcYo9Ti3z08i8gfwbzqc=
=tL0p
-----END PGP SIGNATURE-----
--
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] gcc-3.3.5.20050110-r1 for stable

[Stefan Sperling]

On Fri, Apr 08, 2005 at 07:19:06AM -0400, Dan Meltzer wrote:
> One thing... Maybe its just me... or maybe they are in no way related,
> but I seem to have heard of a lot more 'libtool' problems when using a
> snapshot version instead of a regularly numbered version, is there a
> reason?

Maybe you mean this?

 * Scanning libtool files for hardcoded gcc library paths...
 *   [1/6] Scanning /lib ...
 *   [2/6] Scanning /usr/lib ...
 *     FIXING: /usr/lib/bmp/Input/libmpg123.la ...[v]
 *     FIXING: /usr/lib/kde3/libk3boggvorbisdecoder.la ...[v]
 *     FIXING: /usr/lib/kde3/libk3bexternalencoder.la ...[v]

-- 
stefan                                               PGP Key: 0xF59D25F0
--
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] gcc-3.3.5.20050110-r1 for stable

[Stefan Sperling]

On Thu, Apr 07, 2005 at 10:46:35PM -0500, Mike Frysinger wrote:
> can stable uses of gcc-3.3.5-r1 upgrade to gcc-3.3.5.20050110-r1 and see if 
> they hit any fun and exciting bugs ?

it works nicely here. it compiles and can compile sed.
no bugs, no fun :(
-- 
stefan                                               PGP Key: 0xF59D25F0
--
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] gcc-3.3.5.20050110-r1 for stable

[Stefan Sperling]

On Fri, Apr 08, 2005 at 11:35:19AM +0000, Ferris McCormick wrote:
> Uh, there isn't any such thing.  If you mean this:
> Mon Mar 21 14:05:58 2005 >>> sys-devel/gcc-3.3.5.20050130-r1
> it's been stable on sparc for 2.5 weeks.
> 
> Or did you mean this instead?
> gcc-3.4.3.20050110-r1

mike, I was testing version gcc-3.3.5.20050130-r1.
-- 
stefan                                               PGP Key: 0xF59D25F0
--
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] gcc-3.3.5.20050110-r1 for stable

[Mike Frysinger]

On Thursday 07 April 2005 11:46 pm, Mike Frysinger wrote:
> gcc-3.3.5.20050110-r1

sorry, ive been hacking on both the 3.3.5 and 3.4.3 snapshots lately and i 
thought they were both made 20050110 ...

i am looking for 3.3.5.20050130-r1 to be tested for stable
-mike
--
gentoo-dev@gentoo.org mailing list

[gentoo-dev] List-Unsubscribe

[ren tie'nan]

--
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] gcc-3.3.5.20050110-r1 for stable

[Spider]

[-- Attachment #1: Type: text/plain, Size: 1054 bytes --]

On Thu, 2005-04-07 at 22:46 -0500, Mike Frysinger wrote:
> can stable uses of gcc-3.3.5-r1 upgrade to gcc-3.3.5.20050110-r1 and see if 
> they hit any fun and exciting bugs ?
> 
> if not i'd like to move this to stable this weekend
> -mike


 A touch late, but this uncovered a bug in libtool.  


Basically..  you just fucked all c++ code completely  :p
grep 3.3.5 /usr/bin/libtool
predep_objects="/usr/lib/gcc-lib/i686-pc-linux-gnu/3.3.5/../../../crti.o /usr/lib/gcc-lib/i686-pc-linux-gnu/3.3.5/crtbeginS.o"
postdep_objects="/usr/lib/gcc-lib/i686-pc-linux-gnu/3.3.5/crtendS.o /usr/lib/gcc-lib/i686-pc-linux-gnu/3.3.5/../../../crtn.o"
compiler_lib_search_path="-L/usr/lib/gcc-lib/i686-pc-linux-gnu/3.3.5
-L/usr/lib/gcc-lib/i686-pc-linux-gnu/3.3.5/../../../../i686-pc-linux-gnu/lib -L/usr/lib/gcc-lib/i686-pc-linux-gnu/3.3.5/../../.."



*Sigh* 

https://bugs.gentoo.org/show_bug.cgi?id=88596

//Spider


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


[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-dev] gcc-3.3.5.20050110-r1 for stable

[Mike Frysinger]

On Sunday 10 April 2005 11:31 am, Spider wrote:
>  A touch late, but this uncovered a bug in libtool.

re-emerge libtool and it'll fix itself
-mike
--
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] gcc-3.3.5.20050110-r1 for stable

[Spider]

[-- Attachment #1: Type: text/plain, Size: 618 bytes --]

On Sun, 2005-04-10 at 16:22 -0400, Mike Frysinger wrote:
> On Sunday 10 April 2005 11:31 am, Spider wrote:
> >  A touch late, but this uncovered a bug in libtool.
> 
> re-emerge libtool and it'll fix itself


Yep, I know that. 

Unfortunately, thats -not- going to be a "solution" for the people who
just installed 2005.0, got a gcc update and suddenly everything broke.

the ~x86 version doesn't exhibit this problem, btw.  stabilizing this
version might be prudent.

//Spider

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


[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-dev] gcc-3.3.5.20050110-r1 for stable

[M. Edward (Ed) Borasky]

Should I *not* emerge this gcc? I usually hold off on gcc updates when 
I'm in the middle of other testing. Right now, I'm doing a lot of beta 
testing with R and Atlas, so I held off when the latest gcc showed up 
after "emerge sync".

Spider wrote:

>On Sun, 2005-04-10 at 16:22 -0400, Mike Frysinger wrote:
>  
>
>>On Sunday 10 April 2005 11:31 am, Spider wrote:
>>    
>>
>>> A touch late, but this uncovered a bug in libtool.
>>>      
>>>
>>re-emerge libtool and it'll fix itself
>>    
>>
>
>
>Yep, I know that. 
>
>Unfortunately, thats -not- going to be a "solution" for the people who
>just installed 2005.0, got a gcc update and suddenly everything broke.
>
>the ~x86 version doesn't exhibit this problem, btw.  stabilizing this
>version might be prudent.
>
>//Spider
>
>  
>
--
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] gcc-3.3.5.20050110-r1 for stable

[Mike Frysinger]

On Sunday 10 April 2005 05:22 pm, Spider wrote:
> the ~x86 version doesn't exhibit this problem, btw.  stabilizing this
> version might be prudent.

1.5.14 doesnt have any open issues for it so i've pushed it to stable
-mike
--
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] gcc-3.3.5.20050110-r1 for stable

[Francesco Riosa]

Mike Frysinger wrote:

>On Sunday 10 April 2005 05:22 pm, Spider wrote:
>  
>
>>the ~x86 version doesn't exhibit this problem, btw.  stabilizing this
>>version might be prudent.
>>    
>>
>
>1.5.14 doesnt have any open issues for it so i've pushed it to stable
>  
>


emerge -pv gcc
[ebuild     U ] sys-devel/gcc-3.3.5.20050130-r1 [3.3.5-r1] (-altivec)
-bootstrap -boundschecking -build -debug -fortran -gcj -gtk -hardened
(-ip28) (-multilib) -multislot (-n32) (-n64) +nls -nocxx -objc -static
(-uclibc) 0 kB

[...]

 * Configuring GCC with:
        --enable-version-specific-runtime-libs
        --prefix=/usr
        --bindir=/usr/i686-pc-linux-gnu/gcc-bin/3.3.5-20050130
        
--includedir=/usr/lib/gcc-lib/i686-pc-linux-gnu/3.3.5-20050130/include
        --datadir=/usr/share/gcc-data/i686-pc-linux-gnu/3.3.5-20050130
        --mandir=/usr/share/gcc-data/i686-pc-linux-gnu/3.3.5-20050130/man
        --infodir=/usr/share/gcc-data/i686-pc-linux-gnu/3.3.5-20050130/info
        
--with-gxx-include-dir=/usr/lib/gcc-lib/i686-pc-linux-gnu/3.3.5-20050130/include/g++-v3
        --host=i686-pc-linux-gnu
        --disable-altivec
        --enable-nls
        --without-included-gettext
        --with-system-zlib
        --disable-checking
        --disable-werror
        --disable-libunwind-exceptions
        --disable-multilib
        --disable-libgcj
        --enable-languages=c,c++
        --enable-shared
        --enable-threads=posix
        --enable-__cxa_atexit
        --enable-clocale=gnu

*** This configuration is not supported in the following subdirectories:
     target-libffi target-boehm-gc target-zlib target-libjava 
target-libf2c zlib fastjar target-libobjc
    (Any other directories should still work fine.)
Created "Makefile" in /var/tmp/portage/gcc-3.3.5.20050130-r1/work/build 
using "mt-frag"
gcc: installation problem, cannot exec `as': No such file or directory
*** The command 'gcc -o conftest -O2 -march=i686   conftest.c' failed.
*** You must set the environment variable CC to a working compiler.

!!! ERROR: sys-devel/gcc-3.3.5.20050130-r1 failed.
!!! Function gcc_do_configure, Line 1048, Exitcode 1
!!! failed to run configure
!!! If you need support, post the topmost build error, NOT this status 
message.



--
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] gcc-3.3.5.20050110-r1 for stable

[Francesco Riosa]

Francesco Riosa wrote:

> Mike Frysinger wrote:
>
>> On Sunday 10 April 2005 05:22 pm, Spider wrote:
>>  
>>
>>> the ~x86 version doesn't exhibit this problem, btw.  stabilizing this
>>> version might be prudent.
>>>   
>>
>>
>> 1.5.14 doesnt have any open issues for it so i've pushed it to stable
>>  
>>
>
>
> emerge -pv gcc
> [ebuild     U ] sys-devel/gcc-3.3.5.20050130-r1 [3.3.5-r1] (-altivec)
> -bootstrap -boundschecking -build -debug -fortran -gcj -gtk -hardened
> (-ip28) (-multilib) -multislot (-n32) (-n64) +nls -nocxx -objc -static
> (-uclibc) 0 kB
>
> [...]
>
> * Configuring GCC with:
>        --enable-version-specific-runtime-libs
>        --prefix=/usr
>        --bindir=/usr/i686-pc-linux-gnu/gcc-bin/3.3.5-20050130
>        
> --includedir=/usr/lib/gcc-lib/i686-pc-linux-gnu/3.3.5-20050130/include
>        --datadir=/usr/share/gcc-data/i686-pc-linux-gnu/3.3.5-20050130
>        --mandir=/usr/share/gcc-data/i686-pc-linux-gnu/3.3.5-20050130/man
>        
> --infodir=/usr/share/gcc-data/i686-pc-linux-gnu/3.3.5-20050130/info
>        
> --with-gxx-include-dir=/usr/lib/gcc-lib/i686-pc-linux-gnu/3.3.5-20050130/include/g++-v3 
>
>        --host=i686-pc-linux-gnu
>        --disable-altivec
>        --enable-nls
>        --without-included-gettext
>        --with-system-zlib
>        --disable-checking
>        --disable-werror
>        --disable-libunwind-exceptions
>        --disable-multilib
>        --disable-libgcj
>        --enable-languages=c,c++
>        --enable-shared
>        --enable-threads=posix
>        --enable-__cxa_atexit
>        --enable-clocale=gnu
>
> *** This configuration is not supported in the following subdirectories:
>     target-libffi target-boehm-gc target-zlib target-libjava 
> target-libf2c zlib fastjar target-libobjc
>    (Any other directories should still work fine.)
> Created "Makefile" in 
> /var/tmp/portage/gcc-3.3.5.20050130-r1/work/build using "mt-frag"
> gcc: installation problem, cannot exec `as': No such file or directory
> *** The command 'gcc -o conftest -O2 -march=i686   conftest.c' failed.
> *** You must set the environment variable CC to a working compiler.
>
> !!! ERROR: sys-devel/gcc-3.3.5.20050130-r1 failed.
> !!! Function gcc_do_configure, Line 1048, Exitcode 1
> !!! failed to run configure
> !!! If you need support, post the topmost build error, NOT this status 
> message.
>
and this are the packages emerged few seconds before (gmt times)

     Mon Apr 11 10:27:19 2005 >>> sys-apps/texinfo-4.8
     Mon Apr 11 10:27:43 2005 >>> sys-devel/gcc-config-1.3.10-r2
     Mon Apr 11 10:27:56 2005 >>> sys-devel/binutils-config-1.8-r2
     Mon Apr 11 10:37:17 2005 >>> sys-devel/binutils-2.15.92.0.2-r7

can anyone explain me why updating this fundamentals packages is so 
fragile also on a *stable* x86 installation ?

--
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] gcc-3.3.5.20050110-r1 for stable

[Mike Frysinger]

On Monday 11 April 2005 07:24 am, Francesco Riosa wrote:
> gcc: installation problem, cannot exec `as': No such file or directory

run `binutils-config 1`

there's a bug open atm where portage doesnt seem to run pkg_postinst() for 
some reason and thus when you upgrade your binutils, the proper symlinks 
arent installed automatically
-mike
--
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] gcc-3.3.5.20050110-r1 for stable

[Francesco Riosa]

Mike Frysinger wrote:

>On Monday 11 April 2005 07:24 am, Francesco Riosa wrote:
>  
>
>>gcc: installation problem, cannot exec `as': No such file or directory
>>    
>>
>
>run `binutils-config 1`
>
>there's a bug open atm where portage doesnt seem to run pkg_postinst() for 
>some reason and thus when you upgrade your binutils, the proper symlinks 
>arent installed automatically
>-mike
>  
>
Thank you for the answer, since I still didn't had time to fix it the 
time for search is saved.
Anyway the question is a bit different:
There are "often"  problem of various kind that regard the toolchain, 
varying from linkage, paths, environmental variables ...
Everytime one of this happen users (and probably devs ;) have a Gentoo 
in a quite unusable state.
My question is:
can we fight against this in any way ?
if yes what is needed (dev or portage or user side) ?

francesco





--
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] gcc-3.3.5.20050110-r1 for stable

[Stefan Sperling]

On Mon, Apr 11, 2005 at 03:22:38PM +0200, Francesco Riosa wrote:
> Anyway the question is a bit different:
> There are "often"  problem of various kind that regard the toolchain, 
> varying from linkage, paths, environmental variables ...
> Everytime one of this happen users (and probably devs ;) have a Gentoo 
> in a quite unusable state.
> My question is:
> can we fight against this in any way ?
> if yes what is needed (dev or portage or user side) ?

More testing. I for one did not have a problem at all
upgrading to gcc-3.3.5.20050130-r1 on a stable system when mike
requested testing.
When it actually went stable, problems suddenly showed up on
other peoples' machines. Had they been testing as well, these
errors would have been caught before going stable.

One advantage (or, in this case, problem) of gentoo is that the
system is highly customizable. A lot of errors only show up in
certain configurations. So the more people test things and report
errors, the more stable the system (and the toolchain) becomes.
-- 
stefan                                               PGP Key: 0xF59D25F0
--
gentoo-dev@gentoo.org mailing list