Gentoo Websites Logo
Go to: Gentoo Home Documentation Forums Lists Bugs Planet Store Wiki Get Gentoo!
Bug 107637 - emerging qt with -k downgrades it but without -k everything goes fine.
Summary: emerging qt with -k downgrades it but without -k everything goes fine.
Status: RESOLVED INVALID
Alias: None
Product: Portage Development
Classification: Unclassified
Component: Core - Interface (emerge) (show other bugs)
Hardware: All Linux
: High major (vote)
Assignee: Portage team
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2005-09-29 12:41 UTC by Rodrigo Severo
Modified: 2005-09-30 12:17 UTC (History)
0 users

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


Attachments
"emerge --tree --debug -uDpvk world" log (emerge.with.k.log,372.56 KB, text/plain)
2005-09-30 06:13 UTC, Rodrigo Severo
Details
"emerge --tree --debug -uDpv world" log (emerge.without.k.log.tar.bz2,31.14 KB, application/octet-stream)
2005-09-30 06:15 UTC, Rodrigo Severo
Details

Note You need to log in before you can comment on or make changes to this bug.
Description Rodrigo Severo 2005-09-29 12:41:15 UTC
I have two machines where I am taking advantage of the -k parameter on emerge.
But right now, when I am trying "emerge -uDpvk world" with the following results:

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

Calculating world dependencies ...done!
[ebuild     U ] dev-php/PEAR-XML_RPC-1.4.3 [1.4.2] 0 kB
[binary     UD] x11-libs/qt-3.3.4-r3 [3.3.4-r8] +cups -debug -doc -examples
-firebird +gif -immqt -immqt-bc -ipv6 -mysql* -nas -odbc +opengl -postgres*
-sqlite +xinerama +zlib

Total size of downloads: 0 kB

See that qt is being downgraded to 3.3.4-r3? Why?

If I try "emerge -uDpv world" I get the expected results:

katsuhiro rodrigo # emerge -uDpv world

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

Calculating world dependencies ...done!
[ebuild     U ] dev-php/PEAR-XML_RPC-1.4.3 [1.4.2] 0 kB

Total size of downloads: 0 kB

No messing with qt as it's already upgraded.

The behaviour of "emerge -uDpvk world" is buggy, isn't it?

Reproducible: Always
Steps to Reproduce:





katsuhiro rodrigo # emerge info
Portage 2.0.51.22-r2 (default-linux/x86/2005.1, gcc-3.3.6, glibc-2.3.5-r1,
2.6.11-win4lin i686)
=================================================================
System uname: 2.6.11-win4lin i686 AMD Athlon(tm) XP 2200+
Gentoo Base System version 1.6.13
distcc 2.18.3 i686-pc-linux-gnu (protocols 1 and 2) (default port 3632) [enabled]
dev-lang/python:     2.3.5-r2
sys-apps/sandbox:    1.2.12
sys-devel/autoconf:  2.13, 2.59-r6
sys-devel/automake:  1.4_p6, 1.5, 1.6.3, 1.7.9-r1, 1.8.5-r3, 1.9.6
sys-devel/binutils:  2.15.92.0.2-r10
sys-devel/libtool:   1.5.18-r1
virtual/os-headers:  2.6.11-r2
ACCEPT_KEYWORDS="x86"
AUTOCLEAN="yes"
CBUILD="i686-pc-linux-gnu"
CFLAGS="-O3 -march=i686 -pipe -fomit-frame-pointer"
CHOST="i686-pc-linux-gnu"
CONFIG_PROTECT="/etc /usr/kde/2/share/config /usr/kde/3.4/env
/usr/kde/3.4/share/config /usr/kde/3.4/shutdown /usr/kde/3/share/config
/usr/lib/X11/xkb /usr/share/config /var/qmail/control"
CONFIG_PROTECT_MASK="/etc/gconf /etc/terminfo /etc/env.d"
CXXFLAGS="-O3 -march=i686 -pipe -fomit-frame-pointer"
DISTDIR="/usr/portage/distfiles"
FEATURES="autoconfig buildpkg distcc distlocks fixpackages sandbox sfperms strict"
GENTOO_MIRRORS="http://gentoo.ccccom.com http://gentoo.mirror.sdv.fr
http://distro.ibiblio.org/pub/Linux/distributions/gentoo/
http://gentoo.mirrors.pair.com/ http://gentoo.osuosl.org/"
LANG="pt_BR"
MAKEOPTS="-j9"
PKGDIR="/usr/portage/packages"
PORTAGE_TMPDIR="/var/tmp"
PORTDIR="/usr/portage"
PORTDIR_OVERLAY="/usr/local/portage /usr/local/portage-fabrica"
SYNC="rsync://rsync.gentoo.org/gentoo-portage"
USE="x86 X a52 aac aalib alsa apm arts audiofile avi bitmap-fonts bzip2 bzlib
cddb cdparanoia cdr crypt cups curl dga dts dvd edl eds emboss encode faac faad
fam fame ffmpeg flac foomaticdb fortran gd gdbm gif gimpprint gmp gpm gstreamer
gtk2 imagemagick imlib jbig jpeg jpeg2k kde kdeenablefinal lcms ldap libcaca
libg++ libwww live lzo mad matroska md5sum mikmod mjpeg mng motif mp3 mpeg mysql
ncurses nls ogg oggvorbis opengl oss pam pdflib perl png postgres ppds python qt
quicktime readline real ruby samba scanner sdl speex spell ssl svg svga tcpd tga
theora tiff truetype truetype-fonts type1-fonts unicode vcd vorbis win32codecs
wmf xine xinerama xml2 xmms xv xvid xvmc zlib userland_GNU kernel_linux elibc_glibc"
Unset:  ASFLAGS, CTARGET, LC_ALL, LDFLAGS, LINGUAS
Comment 1 Jason Stubbs (RETIRED) gentoo-dev 2005-09-29 16:45:43 UTC
I'm going to mark this one as INVALID in advance, because I'm sure it's 
behaving correctly. If emerge --tree -uDpvk doesn't show you why, add --debug 
and add that as an attachment and I'll tell you why. 
Comment 2 Rodrigo Severo 2005-09-30 06:11:48 UTC
Man, I have already bugs marked INVALID but definitely none as fast as this one :)

emerge -uDavkt world gives me (just the relevant portion):

[nomerge      ] kde-base/kdeaddons-meta-3.4.1  +arts
[nomerge      ]  kde-base/noatun-plugins-3.4.1  +arts -berkdb -debug
+kdeenablefinal -kdexdeltas +sdl +xinerama
[nomerge      ]   kde-base/noatun-3.4.1  +arts +audiofile -debug +kdeenablefinal
-kdexdeltas +xine +xinerama
[nomerge      ]    kde-base/kdelibs-3.4.1-r1  +alsa +arts +cups -debug -doc
+jpeg2k +kdeenablefinal -kerberos -openexr +spell +ssl +tiff +xinerama -zeroconf
[binary     UD]     x11-libs/qt-3.3.4-r3 [3.3.4-r8] +cups -debug -doc -examples
-firebird +gif -immqt -immqt-bc -ipv6 -mysql* -nas -odbc +opengl -postgres*
-sqlite +xinerama +zlib

emerge -uDavt world gives no qt messing at all.

I looked at the debug output and I believe I understood why is this happening
but it still looks as an unwanted feature (= bug): why with -k I get a downgrade
and without -k I don't.

Just to be clear, I would understand if this kind of thing happened with -K but
with -k?

Anyway I am attaching the debug output of "emerge --tree --debug -uDpvk world"
and of "emerge --tree --debug -uDpv world". Is this really the intended behaviour?
Comment 3 Rodrigo Severo 2005-09-30 06:13:15 UTC
Created attachment 69553 [details]
"emerge --tree --debug -uDpvk world" log
Comment 4 Rodrigo Severo 2005-09-30 06:15:13 UTC
Created attachment 69554 [details]
"emerge --tree --debug -uDpv world" log

I compressed this one as the file is too big.
Comment 5 Jason Stubbs (RETIRED) gentoo-dev 2005-09-30 09:31:16 UTC
Yep, as you've probably already noted the binary package you have of 
kdelibs-3.4.1-r1 effectively DEPENDs on >=qt-3.3.3 <=qt-3.3.4-r7 and the 
closest binary you have to that is qt-3.3.4-r3. Thus portage chooses that one 
and continues on its way. 
 
Whether ignoring what the installed package DEPENDs on is preferable is 
debateable (and pretty one sided at that ;) but the deficiencies of the 
resolver in portage-2.0.x have been accepted. This bug is in the domain of what 
is considered an unfixable deficiency. This and other fundamental flaws will be 
resolved in an upcoming rewrite. 
 
Perhaps I should have marked it CANTFIX or WONTFIX? 
Comment 6 Rodrigo Severo 2005-09-30 12:17:04 UTC
Please don't change it from RESOLVED INVALID as you would ruin a new personal
record :) Seriously INVALID X [WONTFIX|CANTFIX] isn't much of an issue here but
thanks for asking anyway.

I don't think ignoring what the packages DEPENDs on is a good solution at all. 

I see that the dependency is defined in the ebuild as "$(qt_min_version 3.3.3)"
and in the "emerge --debug" file as a fixed list of allowed versions.

Just to be sure I understood the issue correctly, the depencencies are recorded
after intallation in a different file, i.e., the .ebuild file isn't used anymore
AND the dependencies aren't recorded in the same format they are defined in the
.ebuild file: in a less adaptative format.

Please consider this a request to extend the more adaptative format of
dependency recording to the relevant files after installation in future portage
versions. This seems as a perfect solution for issues like these, besides
eventual implementaions problems I might be unaware of.

Thanks for your help, time and attention explaining me the finer issues of
portage management. And please leave this bug as RESOLVED INVALID!