Re: [gentoo-user] Quick emerge -u world questions

2004-01-04 Thread Oliver Lange
Spider wrote:
in this case, sig11 is a common error on compilations for systems where
there are hardware failures. i wanted to rule out that you're one more
of those who got an athlon for christmas and overclocked it at once,
only to have it fail in strange ways when doing large and demanding
compiles.
I didn't follow the thread (didn't read the original question), but i had
serious trouble with the 2.4.22 kernel and distcc and had to change back
to an older kernel (2.4.19). With 2.4.22, distcc threw tons of errors
at any attempt to compile distributed... but it seems that noone else
faced this phenomena..
--
[EMAIL PROTECTED] mailing list


Re: [gentoo-user] Quick emerge -u world questions

2004-01-03 Thread Spider
begin  quote
On Sat, 3 Jan 2004 21:46:45 -0500
Robert G. Waycott <[EMAIL PROTECTED]> wrote:

>  Spider, I
> don't follow your suggestion to 'check the RAM fex.' What's 'fex.'?

"fex", short for "for example"   
in this case, sig11 is a common error on compilations for systems where
there are hardware failures. i wanted to rule out that you're one more
of those who got an athlon for christmas and overclocked it at once,
only to have it fail in strange ways when doing large and demanding
compiles.



> Regarding xfree and Mozilla, I do not have Mozilla on my system. I use
> Opera for the most part. But xfree is running the latest updated
> version. It failed to compile once, but upon retrying, I succesfully
> built it. 


hmm, that sounds good, since if it can do xfree (another demanding task)
it should be able to do glibc/gcc.. 



>  However, now I cannot succesfully emerge such packages as glibc, gcc,
>  kdelibs, and other nefarious behemoths. The only addition to my
>  system that was absent from the 2.4.23 incarnation
> is bootsplash and framebuffer. Would these be in the background 
> gobbling up memory or other resources, effecting compilations? Thank
> you for reading. Send me to Siberia now.


The problem here is that with large things breaking, I'm led to suspect
hardware.   If you want to rule that out, there are known-good binary
copies of gcc + glibc in the archive over at
http://chinstrap.alternating.net ,   if you dare grab one from there
(i686 compile)  those "should" work. built on stable x86 , no fuzz or
weirdness...



However, thats about all I can suggest, the fact that it fails under
heavy load, and not always at the same place, suggests that its a memory
issue though :/


//Spider


-- 
begin  .signature
This is a .signature virus! Please copy me into your .signature!
See Microsoft KB Article Q265230 for more information.
end


pgp0.pgp
Description: PGP signature


Re: Re: [gentoo-user] Quick emerge -u world questions

2004-01-03 Thread Robert G . Waycott

From: Spider <[EMAIL PROTECTED]>
Date: 2004/01/03 Sat PM 07:36:39 EST
To: [EMAIL PROTECTED]
Subject: Re: [gentoo-user] Quick emerge -u world questions

lots of weird segfault /ice..  Are you overclocking?  If so this could
be a show of memorystrain (check the RAM fex.)  but if things like
mozilla and xfree and so on build, its not likely to be that either.


report to bugzilla, thats all the further advice I can offer if the two
above return false.

//Spider


As far as I can recall at the moment, my system is not overclocked. Now, before anyone 
loses their temper with me on that remark, it has been so blasted long since I have 
messed with a CPU setting, that I really have no idea what I may have done long ago. I 
have dual PIII-933s, and on bootup and in dmesg, they register as 933 ... no less, no 
more. So, I'll have to check my BIOS settings. Spider, I don't follow your suggestion 
to 'check the RAM fex.' What's 'fex.'? Regarding xfree and Mozilla, I do not have 
Mozilla on my system. I use Opera for the most part. But xfree is running the latest 
updated version. It failed to compile once, but upon retrying, I succesfully built it. 

Now, to add a further oddity to all of this. Right now I'm running 2.6.0-love sources. 
While running 2.4.23, I successfully emerged everything on the system to ~x86 without 
any errors. I think I was using -j4 for MAKEOPTS, and everything right through KDE 
compiled on my system in just about twelve hours. I had no errors at all. After 
getting 2.6 running--except that my USB is royally screwed up now--I decided to be 
daring and add 'nptl' to my USE flags and 'emerge -e world.' Without going too much 
into it, I successfully borked my box beyond recognition and had to reinstall 
everything. I did not compile a new kernel, but kept my old /boot partition with all 
my kernels on it. Everything boots just fine. I have re-emerged the proper kernel 
sources, and even recompiled the kernel and overwrote what was on /boot just to be 
safe. However, now I cannot succesfully emerge such packages as glibc, gcc, kdelibs, 
and other nefarious behemoths. The only addition to my system that was absent from the 
2.4.23 incarnation is bootsplash and framebuffer. Would these be in the background 
gobbling up memory or other resources, effecting compilations? Thank you for reading. 
Send me to Siberia now.

Neuros


--***--
'Esse quam videri.'
To be, rather than to appear.


--
[EMAIL PROTECTED] mailing list



Re: [gentoo-user] Quick emerge -u world questions

2004-01-03 Thread Spider
begin  quote
On Sat, 3 Jan 2004 17:40:19 -0500
"Robert G. Waycott" <[EMAIL PROTECTED]> wrote:

> On Saturday 03 January 2004 11:03 am, Spider wrote:
> 
> > Now, whats the exact error when compiling gcc?  Not just the ERROR:
> > foo/bar-baz at line...  put some 30-50 lines more than that so we
> > get a context.
> >
> >
> > //Spider
> 
> I'll give it a shot here. I'm still trying to learn just what is 
> relevant/how to read these error spitouts. I just tried it again.
> Also, glibc is failing to compile to the latest ~x86 version. Here is
> the readout of gcc:
> 

lots of weird segfault /ice..  Are you overclocking?  If so this could
be a show of memorystrain (check the RAM fex.)  but if things like
mozilla and xfree and so on build, its not likely to be that either.


report to bugzilla, thats all the further advice I can offer if the two
above return false.

//Spider

-- 
begin  .signature
This is a .signature virus! Please copy me into your .signature!
See Microsoft KB Article Q265230 for more information.
end


pgp0.pgp
Description: PGP signature


Re: [gentoo-user] Quick emerge -u world questions

2004-01-03 Thread Robert G. Waycott
On Saturday 03 January 2004 11:03 am, Spider wrote:

> Now, whats the exact error when compiling gcc?  Not just the ERROR:
> foo/bar-baz at line...  put some 30-50 lines more than that so we get
> a context.
>
>
> //Spider

I'll give it a shot here. I'm still trying to learn just what is 
relevant/how to read these error spitouts. I just tried it again. Also, 
glibc is failing to compile to the latest ~x86 version. Here is the 
readout of gcc:

Immediately at end of readout

cc-3.3.2/gcc/. -I/var/tmp/portage/gcc-3.3.2-r5/work/gcc-3.3.2/gcc/config 
-I/var/tmp/portage/gcc-3.3.2-r5/work/gcc-3.
3.2/gcc/../include /var/tmp/portage/gcc-3.3.2-r5/work/gcc-3.3.2/gcc/c-objc-common.c 
-o c-objc-common.o
rm -rf libcpp.a
ar rc libcpp.a cpplib.o cpplex.o cppmacro.o cppexp.o cppfiles.o 
cpptrad.o cpphash.o cpperror.o cppinit.o cppdefault.
o cppmain.o hashtable.o line-map.o mkdeps.o prefix.o mbchar.o
ranlib libcpp.a
gcc -c   -O -DIN_GCC   -W -Wall -Wwrite-strings -Wstrict-prototypes 
-Wmissing-prototypes -pedantic -Wno-long-long   
-DHAVE_CONFIG_H-I. -I. 
-I/var/tmp/portage/gcc-3.3.2-r5/work/gcc-3.3.2/gcc 
-I/var/tmp/portage/gcc-3.3.2-r5/work/g
cc-3.3.2/gcc/. -I/var/tmp/portage/gcc-3.3.2-r5/work/gcc-3.3.2/gcc/config 
-I/var/tmp/portage/gcc-3.3.2-r5/work/gcc-3.
3.2/gcc/../include /var/tmp/portage/gcc-3.3.2-r5/work/gcc-3.3.2/gcc/builtins.c 
-o builtins.o
gcc -c   -O -DIN_GCC   -W -Wall -Wwrite-strings -Wstrict-prototypes 
-Wmissing-prototypes -pedantic -Wno-long-long   
-DHAVE_CONFIG_H-I. -I. 
-I/var/tmp/portage/gcc-3.3.2-r5/work/gcc-3.3.2/gcc 
-I/var/tmp/portage/gcc-3.3.2-r5/work/g
cc-3.3.2/gcc/. -I/var/tmp/portage/gcc-3.3.2-r5/work/gcc-3.3.2/gcc/config 
-I/var/tmp/portage/gcc-3.3.2-r5/work/gcc-3.
3.2/gcc/../include /var/tmp/portage/gcc-3.3.2-r5/work/gcc-3.3.2/gcc/combine.c 
-o combine.o
/var/tmp/portage/gcc-3.3.2-r5/work/gcc-3.3.2/gcc/combine.c: In function 
`can_combine_p':
/var/tmp/portage/gcc-3.3.2-r5/work/gcc-3.3.2/gcc/combine.c:1236: 
unrecognizable insn:
(insn 1210 669 670 (set (reg:SI 1 edx)
(UnKnown Unknown)) -1 (nil)
(nil))
/var/tmp/portage/gcc-3.3.2-r5/work/gcc-3.3.2/gcc/combine.c:1236: 
Internal compiler error in extract_insn, at recog.c
:2149
Please submit a full bug report,
with preprocessed source if appropriate.
See http://bugs.gentoo.org/> for instructions.
make[2]: *** [combine.o] Error 1
make[2]: *** Waiting for unfinished jobs
make[2]: Leaving directory 
`/var/tmp/portage/gcc-3.3.2-r5/work/build/gcc'
make[1]: *** [stage1_build] Error 2
make[1]: Leaving directory 
`/var/tmp/portage/gcc-3.3.2-r5/work/build/gcc'
make: *** [bootstrap-lean] Error 2

!!! ERROR: sys-devel/gcc-3.3.2-r5 failed.
!!! Function src_compile, Line 411, Exitcode 2
!!! (no error message)


Some time before that

gcc -c   -O -DIN_GCC   -W -Wall -Wwrite-strings -Wstrict-prototypes 
-Wmissing-prototypes -pedantic -Wno-long-long   
-DHAVE_CONFIG_H-DSTANDARD_STARTFILE_PREFIX=\"../../../\" 
-DSTANDARD_EXEC_PREFIX=\"/usr/lib/gcc-lib/\" -DDEFAULT_
TARGET_VERSION=\"3.3.2\" -DDEFAULT_TARGET_MACHINE=\"i686-pc-linux-gnu\" 
-DSTANDARD_BINDIR_PREFIX=\"/usr/i686-pc-linu
x-gnu/gcc-bin/3.3/\" -DTOOLDIR_BASE_PREFIX=\"../../../../\"  `test 
"X${SHLIB_LINK}" = "X" || test "yes" != "yes" || 
echo "-DENABLE_SHARED_LIBGCC"` `test "X${SHLIB_MULTILIB}" = "X" || echo 
"-DNO_SHARED_LIBGCC_MULTILIB"` \
-I. -I. -I/var/tmp/portage/gcc-3.3.2-r5/work/gcc-3.3.2/gcc 
-I/var/tmp/portage/gcc-3.3.2-r5/work/gcc-3.3.2/gc
c/. -I/var/tmp/portage/gcc-3.3.2-r5/work/gcc-3.3.2/gcc/config 
-I/var/tmp/portage/gcc-3.3.2-r5/work/gcc-3.3.2/gcc/../
include /var/tmp/portage/gcc-3.3.2-r5/work/gcc-3.3.2/gcc/java/jvspec.c 
-o jvspec.o)
(cd intl && make all)
make[3]: Entering directory 
`/var/tmp/portage/gcc-3.3.2-r5/work/build/gcc/intl'
make[3]: Nothing to be done for `all'.
make[3]: Leaving directory 
`/var/tmp/portage/gcc-3.3.2-r5/work/build/gcc/intl'
gcc   -O -DIN_GCC   -W -Wall -Wwrite-strings -Wstrict-prototypes 
-Wmissing-prototypes -pedantic -Wno-long-long   -DH
AVE_CONFIG_H-I. -I. 
-I/var/tmp/portage/gcc-3.3.2-r5/work/gcc-3.3.2/gcc 
-I/var/tmp/portage/gcc-3.3.2-r5/work/gcc-
3.3.2/gcc/. -I/var/tmp/portage/gcc-3.3.2-r5/work/gcc-3.3.2/gcc/config 
-I/var/tmp/portage/gcc-3.3.2-r5/work/gcc-3.3.2
/gcc/../include \
   -c /var/tmp/portage/gcc-3.3.2-r5/work/gcc-3.3.2/gcc/c-parse.c -o 
c-parse.o
/var/tmp/portage/gcc-3.3.2-r5/work/gcc-3.3.2/gcc/java/jvspec.c:58: 
warning: string length `969' is greater than the 
length `509' ISO C89 compilers are required to support


some time before that

In file included 
from /var/tmp/portage/gcc-3.3.2-r5/work/gcc-3.3.2/gcc/gcc.c:148:
configargs.h:2: warning: string length `932' is greater than the length 
`509' ISO C89 compilers are required to supp
ort
/var/tmp/portage/gcc-3.3.2-r5/work/gcc-3.3.2/gcc/gcc.c:718: warning: 
string length `524' is greater than the length 
`509' ISO C89 compilers are required to support
/var/tmp/portage/gcc-3.3.2-r5/work/gcc-3.3.2/gcc/gcc.c:865: warning: 

Re: [gentoo-user] Quick emerge -u world questions

2004-01-03 Thread KamaolaKid


On another note, I think the Gentoo Devs are doing a damn spiffy job. 
I'd like to figure out how to run my car on Gentoo. Anyone care to 
maintain?
   

I *am* running my car on gentoo.  Well, my car PC not my car...  It has 
a 7" touch screen in the dash and a dual P3 1GHz system mounted in a 
custom case in the trunk.  Of course it's running gentoo :-).  I have 
pictures of the project.  I'll post a link when I find them.

-- Kyle S.

--
[EMAIL PROTECTED] mailing list


Re: [gentoo-user] Quick emerge -u world questions

2004-01-03 Thread Spider
begin  quote
On Sat, 3 Jan 2004 08:37:58 -0500
"Robert G. Waycott" <[EMAIL PROTECTED]> wrote:



> A] Are there certain CFLAGS that should NEVER be used with gcc?

Anything above -O2 , basically.  This is done in the ebuild already
("strip-flags" )

 
> 2] Am I actually going to run into any problems with the updated 
> programs if I don't install gcc-3.3.2-r5?

you may, some programs might have miscompiled earlier, so an upgrade
would be necessary for them to work properly / even compile.  Case to
case basis, generally its ok not to, but compiler upgrades are always a
good idea inside the same version.



> D] Why is gcc-3.3.2-r5 at the top of the list for 'emerge -u world'
> but 'emerge -u gcc' says it must update pam, pam-login, hashalot, 
> util-linux, baselayout, and glibc first--all of which actually follow 
> gcc in the list of packages to be updated by 'emerge -u world'?

ordering,  depending on the target the dependency tree is calculated
differently.  (there are multiple targets inside "world" which will be
incorporated into a best-fit way, when doing a single target  you will
get a less branched path. )

> 
> 4] Will I have to rebuild my packages or run revdep-rebuild after 
> finally getting gcc-3.3.2-r5 to emerge one day?
Nope.


> 
> On another note, I think the Gentoo Devs are doing a damn spiffy job. 
> I'd like to figure out how to run my car on Gentoo. Anyone care to 
> maintain?

Hmm,  Thats a lot of cd's youd have to throw out on the ground to make
it run all the way on it,  I'd start by parking it on Gentoo. Much
simpler and less cd's required.

*G*



Now, whats the exact error when compiling gcc?  Not just the ERROR:
foo/bar-baz at line...  put some 30-50 lines more than that so we get a
context.


//Spider


-- 
begin  .signature
This is a .signature virus! Please copy me into your .signature!
See Microsoft KB Article Q265230 for more information.
end


pgp0.pgp
Description: PGP signature


Re: [gentoo-user] Quick emerge -u world questions

2004-01-03 Thread Robert G. Waycott
On Saturday 03 January 2004 08:37 am, Robertt G. Waycott wrote:
> Gentooers:
>
>   For some reason or another--either of which are unbeknownst to
> me--while running 'emerge -u world,' gcc-3.3.2-r5 just will not
> compile. I've done a little bit of messing with CFLAGS, to no avail.
> So, I'm using the handy dandy --skipfirst option with emerge to
> *hopefully* plow through the rest of the list. And, of course, this
> now brings up some questions.
>
> A] Are there certain CFLAGS that should NEVER be used with gcc?
>
> 2] Am I actually going to run into any problems with the updated
> programs if I don't install gcc-3.3.2-r5?
>
> D] Why is gcc-3.3.2-r5 at the top of the list for 'emerge -u world'
> but 'emerge -u gcc' says it must update pam, pam-login, hashalot,
> util-linux, baselayout, and glibc first--all of which actually follow
> gcc in the list of packages to be updated by 'emerge -u world'?
>
> 4] Will I have to rebuild my packages or run revdep-rebuild after
> finally getting gcc-3.3.2-r5 to emerge one day?
>
>
> On another note, I think the Gentoo Devs are doing a damn spiffy job.
> I'd like to figure out how to run my car on Gentoo. Anyone care to
> maintain?
>
> Bob
> [Neuros]
>
>
> --
> [EMAIL PROTECTED] mailing list

Gee golly! How embarassing! I spelled my own first name wrong in my 
email setup. One 't'.

Bob
[Neuros]


--
[EMAIL PROTECTED] mailing list