Gentoo Websites Logo
Go to: Gentoo Home Documentation Forums Lists Bugs Planet Store Wiki Get Gentoo!
Bug 160597 - for -march=pentium4 in CFLAGS: "waiting for [...].o.lock to be removed" displayed by libtool in infinite loop (e.g. app-crypt/gpgme-0.3.14-r1)
Summary: for -march=pentium4 in CFLAGS: "waiting for [...].o.lock to be removed" displ...
Status: RESOLVED OBSOLETE
Alias: None
Product: Gentoo Linux
Classification: Unclassified
Component: [OLD] Core system (show other bugs)
Hardware: x86 Linux
: High normal (vote)
Assignee: Gentoo's Team for Core System packages
URL: https://bugzilla.novell.com/show_bug....
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2007-01-06 23:57 UTC by Jan Kaliszewski
Modified: 2011-11-28 18:03 UTC (History)
2 users (show)

See Also:
Package list:
Runtime testing required: ---


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description Jan Kaliszewski 2007-01-06 23:57:50 UTC
When emerging (in compilation phase) media-sound/glame (media-sound/glame-2.0.1  USE="nls -alsa -debug -gnome -vorbis") such a text displays:

[...]
mv -f glsignal.o .libs/glsignal.o
 i686-pc-linux-gnu-gcc -DHAVE_CONFIG_H -I. -I. -I../.. -I../../src/include -pipe -O2 -fomit-frame-pointer -mcpu=pentium4 -march=pentium4 -mfpmath=sse -Wall -Wno-sign-compare -D_REENTRANT -MT glsignal.lo -MD -MP -MF .deps/glsignal.Tpo -c glsignal.c >/dev/null 2>&1
/bin/sh ../../libtool --tag=CC --mode=compile i686-pc-linux-gnu-gcc -DHAVE_CONFIG_H -I. -I. -I../.. -I../../src/include    -pipe -O2 -fomit-frame-pointer -mcpu=pentium4 -march=pentium4 -mfpmath=sse -Wall -Wno-sign-compare  -D_REENTRANT -c -o glsimd.lo ./glsimd.c -funroll-loops
Waiting for -funroll-loops.o.lock to be removed
Waiting for -funroll-loops.o.lock to be removed
Waiting for -funroll-loops.o.lock to be removed
Waiting for -funroll-loops.o.lock to be removed

...and so on -- in *infinite* loop.

Reproducible: Always

Steps to Reproduce:
emerge media-sound/glame
Actual Results:  
See the description.

Expected Results:  
To emerge media-sound/glame.

I tried to find the bug in the Gentoo Bugzilla -- no results.
I tried to find the bug in Google and found a very similar bug in openSUSE Bugzilla:
https://bugzilla.novell.com/show_bug.cgi?id=224732

Probably it's the same problem with /usr/bin/libtool: that it "will not do the right thing if $lockfile starts with a hyphen" (see the link...).

----

Below -- my /etc/make.conf: (rather not important but may be useful when reproduction the case?)

[...]
CFLAGS="-pipe -O2 -fomit-frame-pointer -mcpu=pentium4 -march=pentium4 -mfpmath=sse"
CHOST="i686-pc-linux-gnu"
[...]
USE="7zip X aalib acl acpi apache2 audiofile bash-completion bindist bluetooth
     bootsplash bzip2 cairo calendar caps cdparanoia cdr css curl curlwrappers
     dba dvd dvdr dvdread dvi ethereal exif ffmpeg flac fltk fluidsynth
     font-server foomaticdb gd gg gimp gimpprint glut gmp gnutls gpgme gphoto2
     gstreamer icq idn imagemagick imap inkjar iproute2 irc irmc jabber jack
     java jpeg jpeg2k kdepim kipi ladcca ladspa lame lash lcms libgda mbox
     mikmod mmx mng modplug moznocompose moznoirc moznomail mozsvg mp3 mp4
     mpeg mpeg2 mplayer multiuser musepack musicbrainz net noamazon nodrm
     nsplugin objc openal openexr opengl operanom2 overlays pdf plugin png pnp
     real realmedia reiserfs ruby samba sasl seamonkey slang slp smime sndfile
     sounds soundtouch sox spamassassin speex spell sse sse2 svg tcltk tetex
     theora threads tidy tiff timidity tlen usb userlocales vim vim-syntax
     visualization wifi win32codecs wmf xattr xine xml xml2 xprint xrandr
     xscreensaver xvid zip -fortran -ipv6"
[...]
Comment 1 Jan Kaliszewski 2007-01-07 00:12:06 UTC
PS. Full emerge --info (maybe useful?):

Portage 2.1.1-r2 (default-linux/x86/2006.1, gcc-4.1.1, glibc-2.4-r4, 2.6.15-27-386 i686)
=================================================================
System uname: 2.6.15-27-386 i686 Intel(R) Pentium(R) 4 CPU 2.40GHz
Gentoo Base System version 1.12.6
Last Sync: Tue, 02 Jan 2007 02:01:02 +0000
app-admin/eselect-compiler: [Not Present]
dev-java/java-config: 1.3.7, 2.0.30
dev-lang/python:     2.4.3-r4
dev-python/pycrypto: 2.0.1-r5
dev-util/ccache:     [Not Present]
dev-util/confcache:  [Not Present]
sys-apps/sandbox:    1.2.17
sys-devel/autoconf:  2.13, 2.61
sys-devel/automake:  1.4_p6, 1.5, 1.6.3, 1.7.9-r1, 1.8.5-r3, 1.9.6-r2, 1.10
sys-devel/binutils:  2.16.1-r3
sys-devel/gcc-config: 1.3.14
sys-devel/libtool:   1.5.22
virtual/os-headers:  2.6.17-r2
ACCEPT_KEYWORDS="x86"
AUTOCLEAN="yes"
CBUILD="i686-pc-linux-gnu"
CFLAGS="-pipe -O2 -fomit-frame-pointer -mcpu=pentium4 -march=pentium4 -mfpmath=sse"
CHOST="i686-pc-linux-gnu"
CONFIG_PROTECT="/etc /usr/kde/3.5/env /usr/kde/3.5/share/config /usr/kde/3.5/shutdown /usr/share/config"
CONFIG_PROTECT_MASK="/etc/env.d /etc/env.d/java/ /etc/gconf /etc/java-config/vms/ /etc/revdep-rebuild /etc/terminfo"
CXXFLAGS="-pipe -O2 -fomit-frame-pointer -mcpu=pentium4 -march=pentium4 -mfpmath=sse"
DISTDIR="/usr/portage/distfiles"
FEATURES="autoconfig distlocks metadata-transfer parallel-fetch sandbox sfperms strict"
GENTOO_MIRRORS="http://distfiles.gentoo.org http://distro.ibiblio.org/pub/linux/distributions/gentoo"
LANG="pl_PL.UTF-8"
LINGUAS="pl"
PKGDIR="/usr/portage/packages"
PORTAGE_RSYNC_OPTS="--recursive --links --safe-links --perms --times --compress --force --whole-file --delete --delete-after --stats --timeout=180 --exclude='/distfiles' --exclude='/local' --exclude='/packages'"
PORTAGE_TMPDIR="/var/tmp"
PORTDIR="/usr/portage"
PORTDIR_OVERLAY="/usr/local/portage"
SYNC="rsync://rsync.gentoo.org/gentoo-portage"
USE="x86 7zip X aalib acl acpi alsa_cards_ali5451 alsa_cards_als4000 alsa_cards_atiixp alsa_cards_atiixp-modem alsa_cards_bt87x alsa_cards_ca0106 alsa_cards_cmipci alsa_cards_emu10k1x alsa_cards_ens1370 alsa_cards_ens1371 alsa_cards_es1938 alsa_cards_es1968 alsa_cards_fm801 alsa_cards_hda-intel alsa_cards_intel8x0 alsa_cards_intel8x0m alsa_cards_maestro3 alsa_cards_trident alsa_cards_usb-audio alsa_cards_via82xx alsa_cards_via82xx-modem alsa_cards_ymfpci alsa_pcm_plugins_adpcm alsa_pcm_plugins_alaw alsa_pcm_plugins_asym alsa_pcm_plugins_copy alsa_pcm_plugins_dmix alsa_pcm_plugins_dshare alsa_pcm_plugins_dsnoop alsa_pcm_plugins_empty alsa_pcm_plugins_extplug alsa_pcm_plugins_file alsa_pcm_plugins_hooks alsa_pcm_plugins_iec958 alsa_pcm_plugins_ioplug alsa_pcm_plugins_ladspa alsa_pcm_plugins_lfloat alsa_pcm_plugins_linear alsa_pcm_plugins_meter alsa_pcm_plugins_mulaw alsa_pcm_plugins_multi alsa_pcm_plugins_null alsa_pcm_plugins_plug alsa_pcm_plugins_rate alsa_pcm_plugins_route alsa_pcm_plugins_share alsa_pcm_plugins_shm alsa_pcm_plugins_softvol apache2 audiofile bash-completion berkdb bindist bitmap-fonts bluetooth bootsplash bzip2 cairo calendar caps cdparanoia cdr cli cracklib crypt css cups curl curlwrappers dba dlloader dri dvd dvdr dvdread dvi elibc_glibc ethereal exif ffmpeg flac fltk fluidsynth font-server foomaticdb gd gdbm gg gimp gimpprint glut gmp gnutls gpgme gphoto2 gpm gstreamer iconv icq idn imagemagick imap inkjar input_devices_keyboard input_devices_mouse iproute2 irc irmc isdnlog jabber jack java jpeg jpeg2k kdepim kernel_linux kipi ladcca ladspa lame lash lcms libg++ libgda linguas_pl mbox mikmod mmx mng modplug moznocompose moznoirc moznomail mozsvg mp3 mp4 mpeg mpeg2 mplayer multiuser musepack musicbrainz ncurses net nls noamazon nodrm nptl nptlonly nsplugin objc openal openexr opengl operanom2 overlays pam pcre pdf perl plugin png pnp ppds pppd python readline real realmedia reflection reiserfs ruby samba sasl seamonkey session slang slp smime sndfile sounds soundtouch sox spamassassin speex spell spl sse sse2 ssl svg tcltk tcpd tetex theora threads tidy tiff timidity tlen truetype-fonts type1-fonts udev unicode usb userland_GNU userlocales video_cards_i810 video_cards_vesa video_cards_vga vim vim-syntax visualization wifi win32codecs wmf xattr xine xml xml2 xorg xprint xrandr xscreensaver xvid zip zlib"
Unset:  CTARGET, EMERGE_DEFAULT_OPTS, INSTALL_MASK, LC_ALL, LDFLAGS, MAKEOPTS, PORTAGE_RSYNC_EXTRA_OPTS
Comment 2 Jan Kaliszewski 2007-01-07 00:19:04 UTC
PS2. It' probably not important but maybe: I was trying to emerge it under chroot envoronment (ran in Ubuntu).
Comment 3 Jan Kaliszewski 2007-01-07 22:56:26 UTC
Today I've the same problem when emerging [ebuild  NS   ]   app-crypt/gpgme-0.3.14-r1  USE="-doc"

[...]
checking for unsigned int... yes
checking size of unsigned int... 4
checking for stpcpy... yes
checking for vasprintf... yes
checking for fopencookie... yes
checking for gpgsm... /usr/bin/gpgsm
configure: creating ./config.status
config.status: creating Makefile
config.status: WARNING:  Makefile.in seems to ignore the --datarootdir setting
config.status: creating assuan/Makefile
config.status: WARNING:  assuan/Makefile.in seems to ignore the --datarootdir setting
config.status: creating gpgme/Makefile
config.status: WARNING:  gpgme/Makefile.in seems to ignore the --datarootdir setting
config.status: creating tests/Makefile
config.status: WARNING:  tests/Makefile.in seems to ignore the --datarootdir setting
config.status: creating tests/gpg/Makefile
config.status: WARNING:  tests/gpg/Makefile.in seems to ignore the --datarootdir setting
config.status: creating tests/gpgsm/Makefile
config.status: WARNING:  tests/gpgsm/Makefile.in seems to ignore the --datarootdir setting
config.status: creating doc/Makefile
config.status: WARNING:  doc/Makefile.in seems to ignore the --datarootdir setting
config.status: creating bonobo/Makefile
config.status: WARNING:  bonobo/Makefile.in seems to ignore the --datarootdir setting
config.status: creating complus/Makefile
config.status: WARNING:  complus/Makefile.in seems to ignore the --datarootdir setting
config.status: creating gpgme/gpgme3-config
config.status: creating config.h
config.status: executing depfiles commands

        GPGME v0.3.14 has been configured as follows:

        GnuPG version: min. 1.2.0
        GnuPG path:    /usr/bin/gpg

        GpgSM version: min. 0.9.2
        GpgSM path:    /usr/bin/gpgsm

make  all-recursive
make[1]: Wejście do katalogu `/var/tmp/portage/gpgme-0.3.14-r1/work/gpgme-0.3.14'
Making all in assuan
make[2]: Wejście do katalogu `/var/tmp/portage/gpgme-0.3.14-r1/work/gpgme-0.3.14/assuan'
make  all-am
make[3]: Wejście do katalogu `/var/tmp/portage/gpgme-0.3.14-r1/work/gpgme-0.3.14/assuan'
source='assuan-util.c' object='assuan-util.lo' libtool=yes \
        depfile='.deps/assuan-util.Plo' tmpdepfile='.deps/assuan-util.TPlo' \
        depmode=gcc3 /bin/sh ../depcomp \
        /bin/sh ../libtool --mode=compile i686-pc-linux-gnu-gcc -DHAVE_CONFIG_H -I. -I. -I.. -I.. -I../include    -pipe -O2 -fomit-frame-pointer -mcpu=pentium4 -march=pentium4 -mfpmath=sse -I/usr/include -Wall -Wcast-align -Wshadow -Wstrict-prototypes -c -o assuan-util.lo `test -f 'assuan-util.c' || echo './'`assuan-util.c
Waiting for assuan-util.o.lock to be removed
Waiting for assuan-util.o.lock to be removed
Waiting for assuan-util.o.lock to be removed
Waiting for assuan-util.o.lock to be removed
Waiting for assuan-util.o.lock to be removed

...and so on -- displayed in *infinite* loop (by /usr/bin/libtool)

# grep -r 'Waiting for $lockfile to be removed' /usr/bin/*
/usr/bin/libtool:       $show "Waiting for $lockfile to be removed"

So it must be (for 99.9%) the same bug.
Note that this time the name doesn't start with a hyphen -- so probably name starting with a hyphen is not *the* problem.

That Gentoo is very young (installed a few days ago).
All that problems caused under chroot environment (ran in Ubuntu, because not all the packages important for me are emerged). So when I try doing it under real Gentoo -- I'll write whether the bug still causes.
Comment 4 SpanKY gentoo-dev 2007-01-08 04:57:17 UTC
it isnt package specific at all

also, nothing is running /usr/bin/libtool on your system ... each package should have its own version bundled in

generally this issue comes up when you have a squirrelly mount/tmp setup
Comment 5 Jan Kaliszewski 2007-01-08 20:05:03 UTC
1.
1.
Hm. I have no /tmp on a separate partition and the directory's permissions seems to be ok:

# ls -dl /tmp
drwxrwxrwt 12 root root 480 sty  8 16:41 /tmp

So what can be wrong with /tmp in this case? :-O

2.
These bugs cause still under real Gentoo (not only under chroot from Ubuntu).

3.
It *is* probably important (sorry that hadn't noticed):

>>> Unpacking glame-2.0.1.tar.gz to /var/tmp/portage/glame-2.0.1/work                
 * Requested autoconf 2.5
 * Using autoconf (GNU Autoconf) 2.61
 * Using autoheader (GNU Autoconf) 2.61
 * QA Notice: ${WANT_AUTOMAKE} variable unset. Please report on http://bugs.gentoo.org/

>>> Unpacking gpgme-0.3.14.tar.gz to /var/tmp/portage/gpgme-0.3.14-r1/work                                
 * Applying gpgme3.m4.patch ... [ ok ]                                          
 * Running eautoreconf in '/var/tmp/portage/gpgme-0.3.14-r1/work/gpgme-0.3.14'
 * QA Notice: ${WANT_AUTOCONF} variable unset. Please report on http://bugs.gentoo.org/           
Comment 6 Robin Johnson archtester Gentoo Infrastructure gentoo-dev Security 2007-01-08 22:39:39 UTC
The old kernel on your Ubuntu system is affected by the tmpfs atimes bug that I personally found and submitted patch to LKML for:
http://www2.kernel.org/git/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=cfd95a9cf58cd9e92d4c23b5ee20b07a3d121477

The QA notices already have seperate bugs filed, so you can leave those for now.

I think covers all of this bug now.
Comment 7 Antek Grzymała (antoszka) 2007-01-08 22:46:49 UTC
(In reply to comment #6)

> The QA notices already have seperate bugs filed, so you can leave those for
> now.
> 
> I think covers all of this bug now.

Hey, the same thing happens on his Gentoo system proper (as he wrote in comment #5).
Comment 8 Jan Kaliszewski 2007-01-08 23:06:34 UTC
(In reply to comment #6)
> The QA notices already have seperate bugs filed, so you can leave those for
> now.

The problem reported here is not a problem of QA notice and so on... I wrote
about it for additional information (because maybe it has sth to do with
that...).

> I think covers all of this bug now.

Unfortunately -- it doesn't.
Comment 9 Robin Johnson archtester Gentoo Infrastructure gentoo-dev Security 2007-01-08 23:15:11 UTC
the libtool locking apparent bug is caused by the tmpfs on the old kernel.
either stop using tmpfs, or upgrade your kernel.

you list nothing else specific under comment #5.
Just 1. info about your /tmp, 2. "These bugs cause still under real Gentoo (not only under chroot from Ubuntu)." with no problems following it. 3. the QA Notices. 
Comment 10 Jan Kaliszewski 2007-01-08 23:28:58 UTC
(In reply to comment #9)
> the libtool locking apparent bug is caused by the tmpfs on the old kernel.
> either stop using tmpfs, or upgrade your kernel.

I don't use tmpfs for /tmp (I'd written that).
I have kernel 2.6.19.1
Comment 11 Jan Kaliszewski 2007-01-08 23:30:32 UTC
PS.
As an experiment, I tried turning on tmpfs for /tmp -- the same effect.
Comment 12 Robin Johnson archtester Gentoo Infrastructure gentoo-dev Security 2007-01-08 23:31:23 UTC
vapier: your bug now. I call outdated libtool somewhere.
Comment 13 Antek Grzymała (antoszka) 2007-01-08 23:31:50 UTC
(In reply to comment #9)
> the libtool locking apparent bug is caused by the tmpfs on the old kernel.
> either stop using tmpfs, or upgrade your kernel.

kernel is 2.6.19.1 and he's NOT using tmpfs (as noted in #5, this is *plain*
*old* tmp with listed permissions).

[a]
Comment 14 Antek Grzymała (antoszka) 2007-01-08 23:32:30 UTC
(In reply to comment #12)
> vapier: your bug now. I call outdated libtool somewhere.

This is a fresh Gentoo install with stage3 and portage-current tarball.
Comment 15 SpanKY gentoo-dev 2007-01-09 01:10:08 UTC
QA Notice for $WANT variables has nothing to do with this bug
Comment 16 Martin Väth 2007-03-16 09:28:49 UTC
(In reply to comment #3)
> app-crypt/gpgme-0.3.14-r1  USE="-doc"

The bug with =app-crypt/gpgme-0.3.14-r1 (AFAIK no other package is concerned, also not other gpgme-versions) is that it goes into an endless loop when it should instead die with an error or give a warning. A way to produce this bug is to add -Wl,-O1 to the CFLAGS (not only to LDFLAGS).
Comment 17 Jan Kaliszewski 2007-06-07 13:30:01 UTC
Important supplement!

The bug occurs if CFLAGS contains "-march=pentium4".

And for example commands:
# CFLAGS="-march=i686" emerge -av glame
# CFLAGS="-march=i586" emerge -av =gpgme-0.3.14-r1
-- emerge the packages with no problems.
Comment 18 Robin Johnson archtester Gentoo Infrastructure gentoo-dev Security 2007-06-07 15:12:46 UTC
Ok, I traced it slightly further, and it's possibly GCC4.1 interacting with libtool.

GCC4.1 emits this with your CFLAGS...
`-mcpu=' is deprecated. Use `-mtune=' or '-march=' instead.

If you change to using -mtune instead of -mcpu (keep -march), does the problem go away?
Comment 19 Jan Kaliszewski 2007-06-07 20:16:12 UTC
(In reply to comment #18)
> If you change to using -mtune instead of -mcpu (keep -march), does the problem
> go away?

Indeed, it works!

So -- not really a bug??
Comment 20 Robin Johnson archtester Gentoo Infrastructure gentoo-dev Security 2007-06-07 20:44:23 UTC
spanky: there you go, libtool is reacting badly to GCC warnings.
valid case for NOT changing the make.conf file to use -mtune is that older GCC versions don't support it on x86.
Comment 21 Samuli Suominen (RETIRED) gentoo-dev 2009-05-30 17:40:05 UTC
Clearly sound@ has nothing to do here, in fact, glame is in the punted soon list for using ancient guile.