Re: [gentoo-user] OT - which X terminal do you use?
On Saturday 13 May 2006 14:02, Nagatoro wrote:
What is the output of:
# echo $PS1
\[\033[38;5;[EMAIL PROTECTED];5;39m\]\h \[\033[38;5;25m\]\w
\[$(ps_retc_f $?)\]$?
\[\033[38;5;70m\]$(ps_job_f)\[\033[38;5;52m\]$(ps_dir_f)\n\[\]\D{%a
%T} \[\033[38;5;77m\]$ \[\033[0;0m\]
Holy crap! How on earth did you come up with that? ;) Could you post the
output of
# env | grep ps
too?
--
Bo Andresen
pgpriSptyz5sv.pgp
Description: PGP signature
Re: [gentoo-user] -unicode USE flag
On Saturday 13 May 2006 18:59, Mark Knecht wrote: What is unicode and why might an end user want it or need it? http://www.gentoo.org/doc/en/utf-8.xml -- Bo Andresen pgpDwUrn2UPqp.pgp Description: PGP signature
Re: [gentoo-user] Warning: Color name black is not defined
On Wednesday 10 May 2006 10:19, Alexander Skwar wrote: Hi! Since a recent update, I always get error messages like the following, when I start certain applications (eg. xterm): Warning: Color name black is not defined xterm: Cannot allocate color red xterm: Cannot allocate color magenta On bgo, I found http://bugs.gentoo.org/show_bug.cgi?id=78385 which suggests to make sure that RgbPath is correct in /etc/X11/xorg.conf. It is correct, I think: RgbPath /usr/share/X11/rgb [10:09:19 [EMAIL PROTECTED]:~] $ ls -la /usr/share/X11/rgb.txt -rw-r--r-- 1 root root 17371 10. Mai 10:07 /usr/share/X11/rgb.txt [snip] Any ideas about what might be broken? # ls -l /usr/share/X11/rgb.txt -rw-r--r-- 1 root root 17371 2006-04-07 03:03 /usr/share/X11/rgb.txt # grep black /usr/share/X11/rgb.txt 0 0 0 black # equery check rgb [ Checking x11-apps/rgb-1.0.0 ] * 9 out of 9 files good -- Bo Andresen pgpUU4FzZpJhz.pgp Description: PGP signature
Re: [gentoo-user] OT - which X terminal do you use?
On Friday 12 May 2006 21:18, Nagatoro wrote: Note that the prompt for konsole is blinking ie invisible every other second. What is the output of: # echo $PS1 -- Bo Andresen pgpoZ6pVk0Gru.pgp Description: PGP signature
Re: [gentoo-user] KDE and OpenOffice keyboard problems [solved]
On Thursday 11 May 2006 12:53, Dave Jones wrote: $ LC_ALL=en_GB.ISO-8859-15 oowriter2 The LC_ALL command above did not work, giving the following error messages: I18N: X Window System doesn't support locale en_GB.ISO-8859-15 Qt: Locales not supported on X server Checked out my defined locales with locale -a [SNIP] LC_ALL=en_GB.iso885915 oowriter2works fine. In fact, all the above locales except for C and POSIX work fine in OpenOffice. It seems that OOo does not depend on a utf8 locale, but does needs a local other the C or POSIX. Which is what I stated on the 4th of May. :) But en_GB.ISO-8859-15 works for me too even though locale -a report en_GB.iso885915 on my computer too. I guess that depends on the version of glibc.. Anyway nice to know it works for you now. :) -- Bo Andresen pgp5oUi7LKkyn.pgp Description: PGP signature
Re: [gentoo-user] reiserfs troubles
On Friday 14 April 2006 14:30, John Jolet wrote: as the others said, start digging a grave for this drive. Had a drive doing the same thing on a production box for a few months, phbs waited till it just died to order the replacement. these kinds of errors are warning signs of impending doom. Just to follow up on this one. It did last a couple of weeks before I really started digging into this. It has turned out that you and Volker Armin were right. I never was able to backup that particular partition (containing /, /usr, /tmp, /opt, /var, /etc) since even dd if=/dev/hda6 gave up after a while reporting an I/O error. All other partitions, however, are backed up properly, and nothing that could not be replaced was on that partition. At present I have tried reiserfsck --badblocks bad.txt with both --fix-fixable and --rebuild-tree. --fix-fixable told me to run --rebuild-tree and --rebuild-tree failed leaving the partition unmountable. I also tried dd if=/dev/zero of=/dev/hda6 which failed after a while. I finally tried formatting the partition with NTFS just to see if Windoze handles this better. Formatting it worked fine, scandisk returned no errors and I could copy 1,5 GB of data to the partition before it failed with a hardware error... In other words it doesn't. ;) Also during all those attempts the Current_Pending_Sectors reported by S.M.A.R.T. has grown from 39 to now 102.. So the result is that I have used another partition for my system since last friday and that the drive will be replaced next monday. Since that other partition is way smaller than the one I used to use and because it is such a short time, I thought I'd take the opportunity to try another distribution i.e. Kubuntu Dabber (beta) with KDE 3.5.2. I must say that I am impressed by the present state of Kubuntu, but I also really miss the control that Gentoo provides so next monday I'll be installing Gentoo on a brand new and more partitioned harddrive. :) I also really miss something equivalent to eix/diff-eix/eix-sync for Debian based systems. Hopefully (and probably) an insurance I have will pay for the replacement (except I'm going to purchase a bigger drive and will pay the difference). -- Bo Andresen pgpFzRVSGynAs.pgp Description: PGP signature
Re: [gentoo-user] KDE and OpenOffice keyboard problems [solved]
On Thursday 04 May 2006 15:29, Dave Jones wrote: I didn't get these errors when I ran the oowriter2 command above, so I guess my localdef for the en_GB.utf8 must have been OK. Case closed, an irritating problem fixed. From another thread: On Tuesday 09 May 2006 20:31, Dave Jones wrote: For some strange reason, on my system OpenOffice seems to need the locale set to utf8 to work properly with international keyboard layouts. Without it, the ' and keys are dead, working only with AltGr pressed. I don't understand why, but since I changed my locale to en_US.utf8, the quote keys and Open Office work perfectly. Does this not work for you (# means run as root, $ means run as user)? # localedef -i en_GB -f ISO-8859-15 en_GB.ISO-8859-15 $ LC_ALL=en_GB.ISO-8859-15 oowriter2 -- Bo Andresen pgp7jTMR4TFw9.pgp Description: PGP signature
Re: [gentoo-user] New USE flags???
On Tuesday 09 May 2006 04:58, Jerry McBride wrote: The suggested equery wasn't much help # equery belongs equery [ Searching for file(s) equery in *... ] app-portage/gentoolkit-0.2.2_pre4 (/usr/bin/equery) app-portage/gentoolkit-0.2.2_pre4 (/usr/share/doc/gentoolkit-0.2.2_pre4/equery) -- Bo Andresen pgpnYRgF8ELHF.pgp Description: PGP signature
Re: [gentoo-user] national keyboard in open office
On Tuesday 09 May 2006 17:28, Pawel K wrote: Unfortunately I can't type Polish charecters under mozilla as well (tried both pl_PL and pl_PL.utf8). I expect to use right ALT + character e.g. right ALT + o should result with ó and now it results with just o both in Mozilla and OpenOffice. Left ALT + o opens F'o'rmat menu under OpenOffice. How did you start OOo? Have you tried starting it from a cli (like xterm, konsole, ..)? Does it give any errors messages if you do that? E.g. if I set my locale to pl_PL I get the following output (which is expected since I didn't build a polish locale and hence it is not showing up in locale -a): # LC_ALL=pl_PL oowriter2 perl: warning: Setting locale failed. perl: warning: Please check that your locale settings: LC_ALL = pl_PL, LANG = pl_PL are supported and installed on your system. perl: warning: Falling back to the standard locale (C). I18N: Operating system doesn't support locale I18N: Operating system doesn't support locale en_US -- Bo Andresen pgpJOMOrUyPEu.pgp Description: PGP signature
Re: [gentoo-user] System.map not found - unable to check symbols
Reordered to make it more readable. On Tuesday 09 May 2006 15:56, de Almeida, Valmor F. wrote: System.map not found - unable to check symbols. which doesn't seem to cause problems during/after booting (??). I did a manual kernel compilation To do this, I always do: make all modules_install install This will do all the necessary steps. I tried the make all option and it added a /boot - . Inside /boot. Also, a menu.lst file was created inside /boot/grub that points to grub.conf. Other than that there no changes/additions we made. 'make all' is supposed to compile the kernel, 'make modules_install' will compile the kernel modules, 'make install' will install the kernel and 'make all modules_install install' will do all three of those things. I rebooted and had the same problem occurring: System.map not found -- unable to check symbols Could you provide the output of: # df -h | grep boot # ls -l /boot # uname -r And please no top-posting i.e. post your replies below whatever you are replying to. -- Bo Andresen pgp9vodvzSuoT.pgp Description: PGP signature
Re: [gentoo-user] national keyboard in open office
On Friday 05 May 2006 17:30, Pawel K wrote: Hello How to force open office to type national (Polish) fonts. I can see them on the screen when opening the document containing them but when I press the combination right alt-char it doesn't work. I've installed open office as english(USA) version and itshould stay like that. I want to be able to type Polish characters only. What is the output of # locale and # locale -a Not all locales will support those characters. You can read more on locales at http://www.gentoo.org/doc/en/guide-localization.xml . -- Bo Andresen pgptUL4kwfBlX.pgp Description: PGP signature
Re: [gentoo-user] why firefox is so slow?
On Thursday 04 May 2006 05:19, [EMAIL PROTECTED] wrote: On Monday 01 May 2006 11:03 pm, Alexander Skwar wrote: [EMAIL PROTECTED] wrote: Outlook and Outlook Express are the two worst mail clients in the universe. [SNIP] They're too slow Not really. You must not have used them. I have. They are. End of story. Please stop saying that everyone else is wrong and you are right when you obviously don't know what you are talking about! and bloated to do anything, and they don't accept ANYTHING other than MS Exchange server Wrt. Outlook Express, that's plain wrong. And wrt. Outlook: It's wrong as well. Outlook CAN interface IMAP servers. But you'll lose all the benefits of Outlook in this case. I said that they can't interface with non-MS Exchange servers withour throwing a Royal Temper Tantrum. They do not like POP. They like IMAP even less. You are wrong! With a statement like this it's really hard to believe that you ever actually tried those programs! Outlook has handled POP3 and IMAP since version 8 at least (that's just the first version I ever tried so I don't know anything about earlier versions). Besides that this has no relevance on this list whatsoever. So why discuss it?? Stop making statements of things of which you have no knowledge about. -- Bo Andresen pgpovli38bktY.pgp Description: PGP signature
Re: [gentoo-user] KDE and OpenOffice keyboard problems [solved]
On Thursday 04 May 2006 11:17, Dave Jones wrote: Bo, thank you *very* much, starting oowriter2 as you suggested using LC_All=en_US.utf8 oowriter2 worked, the us_intl keyboard now works as it should. Glad to hear that. :) I have no /etc/env.d/02locale file; did you create that file yourself? Yes, just create a file with the contents: LC_ALL=whatever Look at [1] and [2] for more information on locales on Gentoo. [1] http://www.gentoo.org/doc/en/guide-localization.xml#doc_chap3 [2] http://www.gentoo.org/doc/en/utf-8.xml -- Bo Andresen pgphsoHobO0Et.pgp Description: PGP signature
Re: [gentoo-user] KDE and OpenOffice keyboard problems [solved]
On Thursday 04 May 2006 12:08, Dave Jones wrote: As a matter of curiosity, what errors did you get when you tried en_GB.utf8 on your system? $ LC_ALL=en_GB.utf8 oowriter2 perl: warning: Setting locale failed. perl: warning: Please check that your locale settings: LANGUAGE = (unset), LC_ALL = en_GB.utf8, LANG = (unset) are supported and installed on your system. perl: warning: Falling back to the standard locale (C). I18N: Operating system doesn't support locale IIRC back when I followed the UTF-8 guide localedef complained that some data for creating the en_GB.UTF-8 locale was missing. It seems that the issue has been resolved since then. The following command fixed the problem so en_GB.utf8 is working for me too now: # localedef -i en_GB -f UTF-8 en_GB.UTF-8 -- Bo Andresen pgp3bNdR1D0qE.pgp Description: PGP signature
Re: [gentoo-user] KDE and OpenOffice keyboard problems
On Wednesday 03 May 2006 00:04, Jerônimo Backes wrote: Oh, I forgot to mention: DON'T use UTF-8 with KDE if you need to use some special keyboard layouts. Stay with ISO-8859-1 (the default encoding). KDE has issues with UTF-8 and certain keboard layouts. What kind of issues? Admittedly I have not tested KDE throughly but I have been using KDE with UTF-8 for a few months now and I have experienced no issues. There are programs that don't understand UTF-8 on my system (e.g. kompare) but I believe most og my KDE programs do understand UTF-8 (e.g. kontact, konquerer, konsole..). -- Bo Andresen pgpsr03XMKDtt.pgp Description: PGP signature
Re: [gentoo-user] KDE and OpenOffice keyboard problems
On Wednesday 03 May 2006 15:13, Dave Jones wrote: Bo Andresen wrote on 05/03/06 14:31: On Wednesday 03 May 2006 00:04, Jerônimo Backes wrote: Oh, I forgot to mention: DON'T use UTF-8 with KDE if you need to use some special keyboard layouts. Stay with ISO-8859-1 (the default encoding). KDE has issues with UTF-8 and certain keboard layouts. What kind of issues? Admittedly I have not tested KDE throughly but I have been using KDE with UTF-8 for a few months now and I have experienced no issues. There are programs that don't understand UTF-8 on my system (e.g. kompare) but I believe most og my KDE programs do understand UTF-8 (e.g. kontact, konquerer, konsole..). This came out of my original question about having problems with typing accented characters into OpenOffice documents under KDE 3.4.3. The problem is that I need to press AltGr and the quote character to obtain a quote. This makes it impossible to enter accented characters into an OOo document. Section InputDevice Identifier Keyboard0 Driver kbd Option AutoRepeat500 30 Option CoreKeyboard Option XkbModel logicdp Option XkbLayout us_intl EndSection Using this xorg.conf set-up, I can enter the accented characters into all other applications apart from OOo under KDE. OOo works fine under IceWM, so I'd guess that it's a KDE issue. I still don't see that it has got anything to do with UFT-8. My X configuration looks like this: Section InputDevice Identifier Keyboard1 Driver kbd Option AutoRepeat 500 30 Option CoreKeyboard Option XkbModel pc105 Option XkbLayout dk EndSection In kcontrol - Regional Accessibility - Keyboard Layout there is a box called Enable keyboards layouts. This overrides the X settings. So if it works in IceWM just disable it. That's what I did. And I have no problems with any of those characters (áàéèíìóòöuúù€') in Openoffice either. -- Bo Andresen pgpKCsyU59I5R.pgp Description: PGP signature
Re: [gentoo-user] KDE and OpenOffice keyboard problems
On Wednesday 03 May 2006 18:04, Jerônimo Backes wrote: I mean issues with some keyboard layouts. Try to type some cedillas with a US-keyboard (combining ' and c ) in KDE with UTF-8 to see what you'll get. Do you mean like this: ç? I use a danish keyboard layout and have no idea how to produce that character. This one was created in OOo by Insert Special Character (and copy'n'paste). I did try naming a file with that character and putting it in its contents and found no issues. But I don't know if that was what you meant. Anyway I'm not saying that there are no issues. I just haven't experienced any. -- Bo Andresen pgpMsR5keXsDu.pgp Description: PGP signature
Re: [gentoo-user] KDE and OpenOffice keyboard problems
On Wednesday 03 May 2006 17:26, Dave Jones wrote: I've tried with the kcontrol Regional Keyboard settings enabled and disabled. No difference. The quote keys work only with AltGr pressed in OOo, but work normally in all other applications. I tried using XkbModel to pc105 to see if the Logitech Cordless Desktop Pro XkbModel was causing the odd effect. However, changing that did not change the weird quote keys kb behaviour either. If anything, the Regional Keyboard setting make the problem worse, affecting other applications too, so I disabled it again. The us_intl keyboard layout is shown in [1]. Your problem related to the red (i.e. dead) keys to the left of Enter, right? I have changed to that layout now (with no KDE override) and it seems to work the way it should here now. I'm not sure it did before I changed the contents of /etc/conf.d/keymaps. What does this show? # grep -vr '^#\|^$' /etc/conf.d/keymaps [1] http://en.wikipedia.org/wiki/Keyboard_layout#US-International Since the keys are dead you do have to press them twice in order to just get a single or double quote. If followed by a letter are able to produce accents (ó) or umlauts (ö) instead. And just for the record. I can produce a ç now. I don't seem to have any issues. But still a haven't tested with a broad range of programs. ;) As far as I know, I'm not using UTF-8. You are not. [SNIP] Maybe it's not an UTF-8 issue as Jerônimo suggested it might be, but I'm baffled by this odd kb behaviour showing only in OOo under KDE. Could you please show the output of locale and locale -a on your system? Sure. # locale LANG= LC_CTYPE=en_US.utf8 LC_NUMERIC=en_US.utf8 LC_TIME=en_US.utf8 LC_COLLATE=en_US.utf8 LC_MONETARY=en_US.utf8 LC_MESSAGES=en_US.utf8 LC_PAPER=en_US.utf8 LC_NAME=en_US.utf8 LC_ADDRESS=en_US.utf8 LC_TELEPHONE=en_US.utf8 LC_MEASUREMENT=en_US.utf8 LC_IDENTIFICATION=en_US.utf8 LC_ALL=en_US.utf8 # locale -a C da_DK en_DK en_DK.iso88591 en_DK.utf8 en_GB en_GB.ansix341968 en_GB.iso88591 en_US en_US.utf8 POSIX But I don't really think this is the problem. -- Bo Andresen pgpu9kjxGQbdX.pgp Description: PGP signature
Re: [gentoo-user] GnuPG Trouble - Found Problem, Need Solution
On Tuesday 25 April 2006 12:51, Mick wrote: I only have a couple of minutes so I apologise for not writing a complete answer. FWIW the gpg-agent was hard masked (yesterday?) and just adding it to your /etc/portage/package.keywords won't work. You need to use package.mask instead. However, I'm not sure why the hard masking was put in place - there may be a good reason to stop using gpg-agent and instead switch to another package. What should we use instead guys? # emerge -vp gpg-agent These are the packages that would be merged, in order: Calculating dependencies !!! All ebuilds that could satisfy gpg-agent have been masked. !!! One of the following masked packages is required to complete your request: - app-crypt/gpg-agent-1.9.19 (masked by: package.mask, ~x86 keyword) # Stefan Schweizer [EMAIL PROTECTED] (23 Apr 2005) # Please use app-crypt/gnupg-1.9.* now - app-crypt/gpg-agent-1.9.18 (masked by: package.mask, ~x86 keyword) So: # emerge -Cv gpg-agent # emerge -va gnupg -- Bo Andresen pgpsmtetCytuu.pgp Description: PGP signature
Re: [gentoo-user] reiserfs troubles
On Friday 14 April 2006 03:54, Hemmann, Volker Armin wrote: If you have smart enabled (and you should have) and smartmontools installed (you should have done that too), smart can tell you, if it is a hardware problem (which is pretty probably the cause of the problem). Thank you very much for the pointer. I had never heard of this before but it is definitely going on it the future.. It confirms that the disc has 39 bad blocks so now I am reading the faq to figure out how to deal with this. :) Back up everything... As stated in the original post I am trying to. But when I try to copy everything to an external harddrive after some time the external harddrive is disconnected. I am not sure if this is a separate issue or if it happens as a result of too many errors on my system disk. Anyway I have backed up everything that cannot be replaced so first I will deal with the bad blocks and then if the problem remains I will probably tar everything up to my server via ssh.. PS. This mail is signed. Can anyone verify the signature? [...] Thanks to you and Boyd for your replies on this too. From Boyd's post I am guessing that it works as well as it can work and from yours I am guessing that it needs some more configuration than you have done. Not sure though.. -- Bo Andresen pgp7xKx1qDZGV.pgp Description: PGP signature
Re: [gentoo-user] kde-base/kdebugdialog-3.5.2 emerge error
On Thursday 13 April 2006 07:35, Richard Fish wrote: cd /var/db/pkg for pkg in kde-base/*-3.4*; do emerge --unmerge $pkg done Why not just: cd /var/db/pkg emerge --unmerge --verbose --ask kde-base/*-3.4* -- Bo Andresen pgpO9B6TK90qK.pgp Description: PGP signature
[gentoo-user] reiserfs troubles
I am currently trying to backup everything on my laptop in order to
repartition it. My root partition, /dev/hda6 is a reiserfs file system. /opt
is currently on the root partition.
# cd /opt/whatever
# chmod +r helpindex.xml
chmod: cannot access `helpindex.xml': Permission denied
# dmesg | tail -n 5
hda: dma_intr: status=0x51 { DriveReady SeekComplete Error }
hda: dma_intr: error=0x40 { UncorrectableError }, LBAsect=19603129, high=1,
low=2825913, sector=19603129
ide: failed opcode was: unknown
end_request: I/O error, dev hda, sector 19603129
ReiserFS: hda6: warning: vs-13070: reiserfs_read_locked_inode: i/o failure
occurred trying to find stat data of [5806433 5806438 0x0 SD]
This file isn't really important to me but I seem to have a lot of
them in /opt. I am quite curious about what can be done with this..
Thanks in advance.
PS. This mail is signed. Can anyone verify the signature?
--
Bo Andresen
pgp4TwNAgMpxW.pgp
Description: PGP signature
Re: [gentoo-user] Help : mc and ln mixup
On Friday 14 April 2006 02:06, Rohit Sharma wrote: Hi there, Please see the following. Both /usr/bin/ln and /usr/bin/mc on my system seem to launch mc. if I delete one, the other disappears as well. if I emerge mc, /usr/bin/ln also appears. Even if /usr/bin/ln is deleted, which ln returns /usr/bin/ln What is happening. I am completely baffled. Sounds weird.. # cd /usr/bin/ # which ln /usr/bin/ln # mv ln .. # which ln /bin/ln # mv ../ln . # which ln /usr/bin/ln If you run # echo $PATH you will see that /usr/bin is listed before /bin. which searches the directories in $PATH and stops at the first match so since /usr/bin comes before /bin it returns /usr/bin/ln. If you delete /usr/bin/ln it should return /bin/ln which should be unaffected. If you delete /bin/ln then /usr/bin/ln will become a dangling symbolic link pointing at a file which doesn't exist.. Which version of coreutils do you have? # equery b ln [ Searching for file(s) ln in *... ] [...] sys-apps/coreutils-5.94-r1 (/bin/ln) sys-apps/coreutils-5.94-r1 (/usr/bin/ln - /bin/ln) If you are still having problems with this then show us the commands that you are running and the output that leads you to your conclusions.. HtH -- Bo Andresen pgpXwqrbekMu4.pgp Description: PGP signature
Re: [gentoo-user] Switching to Gentoo
On Thursday 06 April 2006 07:41, Alexander Skwar wrote:
Bo Andresen wrote:
Heh.. this one is better ;) :
# cat /var/lib/portage/world | awk -F\/ '{system(eix -F -C $1 -e $2
-c | head -n 1)}'
Nah, it's NOT, because of the Useless Use Of Cat. Even
better:
awk -F\/ '{system(eix -F -C $1 -e $2 -c | head -n 1)}'
/var/lib/portage/world
Useless use of cat is always... well... useless. But even despite the useless
use of cat it is better than the original command. Here is why:
# eix ^sys-libs/libstdc++-v3$
Found 0 matches
$ eix -C sys-libs -e libstdc++-v3 -c
[I] sys-libs/libstdc++-v3 (3.3.6): Compatibility package for running binaries
linked against a pre gcc 3.4 libstdc++
Found 1 matches
Here the original command resulted in an empty line whereas the new command
results in the package being shown. Still I really should remember to avoid
useless use of cat especially when posting to a mailing. :)
--
Bo Andresen
--
gentoo-user@gentoo.org mailing list
Re: [gentoo-user] I don't want to upgrade to xorg-x11-7.0
On Thursday 06 April 2006 23:59, Alexander H. Faeroey wrote: server portage # cat /etc/portage/package.mask =x11-base/xorg-x11-6.8.2-r6 I think you need to add =x11-base/xorg-x11-6.8.2-r6 Bigger or equal to NO! That would force him to downgrade. He only wanted to mask version 7 and up. -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] I don't want to upgrade to xorg-x11-7.0
On Thursday 06 April 2006 23:45, Daevid Vincent wrote: How do I prevent my server from upgrading to Xorg 7.0 modular (although ideally, I would like to upgrade any 6.x regular Xorg versions that come along). [SNIP] server portage # cat /etc/portage/package.mask =x11-base/xorg-x11-6.8.2-r6 But when I do an emerge -Davu world it still tries to pull xorg dependencies: !!!(dependency required by x11-base/xorg-x11-7.0-r1 [ebuild]) I *DO* have this in my package.keywords file, along with a bunch of other ones for KDE/Gnome/etc: x11-base/xorg-x11 ~x86 But shouldn't the 'mask' file have priority? You really seem to have misunderstood the concept of masked packages. Richard and Neil have told you what to do but I would advise you to read the following: http://www.gentoo.org/doc/en/handbook/handbook-x86.xml?part=2chap=1#doc_chap4 http://www.gentoo.org/doc/en/handbook/handbook-x86.xml?part=3chap=3 -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Why does Xaw3d depend on indirectly on x11-misc/imake and x11-misc/gccmakedep?
On Wednesday 05 April 2006 11:19, Erik wrote:
It seems like Xaw3d depends indirectly on x11-misc/imake and
[SNIP]
[ebuild R ] x11-libs/Xaw3d-1.5-r1 0 kB
[SNIP]
But I can not find out why because nothing seems to depend on
x11-misc/imake and x11-misc/gccmakedep:
# cat /usr/portage/x11-libs/Xaw3d/Xaw3d-1.5-r1.ebuild
[SNIP]
DEPEND=${RDEPEND}
=sys-apps/sed-4
|| ( ( x11-proto/xextproto
x11-misc/imake
x11-misc/gccmakedep )
virtual/x11 )
[SNIP]
# equery d x11-misc/imake
[ Searching for packages depending on x11-misc/imake... ]
# equery d x11-misc/xorg-cf-files
[ Searching for packages depending on x11-misc/xorg-cf-files... ]
# equery d x11-misc/gccmakedep
[ Searching for packages depending on x11-misc/gccmakedep... ]
What command should I execute to find out why x11-misc/imake and
x11-misc/gccmakedep are needed?
# equery d imake
[ Searching for packages depending on imake... ]
x11-libs/Xaw3d-1.5-r1
virtual/x11-7.0-r1
# equery d gccmakedep
[ Searching for packages depending on gccmakedep... ]
x11-libs/Xaw3d-1.5-r1
virtual/x11-7.0-r1
I think you are running an old version of gentoolkit. equery has improved
quite a bit in the newest unstable version.
# eix -e gentoolkit -c
[I] app-portage/gentoolkit (0.2.2_pre4): Collection of administration scripts
for Gentoo
--
Bo Andresen
--
gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Why does Xaw3d depend on indirectly on x11-misc/imake and x11-misc/gccmakedep?
On Wednesday 05 April 2006 12:57, Erik wrote:
No Xaw3d there as far as I can see.
If you look in the ebuild of Xaw3d do you then see this?
$ cat /usr/portage/x11-libs/Xaw3d/Xaw3d-1.5-r1.ebuild | grep -A 5 DEPEND
RDEPEND=|| ( ( x11-libs/libXt
x11-libs/libX11
x11-libs/libXmu
x11-libs/libXpm
x11-libs/libXp )
virtual/x11 )
--
DEPEND=${RDEPEND}
=sys-apps/sed-4
|| ( ( x11-proto/xextproto
x11-misc/imake
x11-misc/gccmakedep )
virtual/x11 )
If you don't then perhaps emerge --sync ?
--
Bo Andresen
--
gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Why does Xaw3d depend on indirectly on x11-misc/imake and x11-misc/gccmakedep?
On Wednesday 05 April 2006 16:00, Erik wrote: Yes I see this. Do you have imake installed? If not then perhaps install it and see if the problem with equery persists. Another thing you could try is run emerge --metadata or --sync. I don't know if this could be a bug. When I run equery d -a x11-misc/imake I get about twice as much output (including Xaw3d) as you did. So if the problem persists and noone else has a suggestion you might consider filing a bug on equery. -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Switching to Gentoo
On Thursday 09 March 2006 23:50, Bo Andresen wrote:
In /var/lib/portage/world there is a list of all of all software that you
have explicitly installed. The rest of what is installed should be
depencies of packages in the world file. So to get a list of packages that
I installed I would use (this is only one line):
#while read pkg; do eix --force-color --compact ^${pkg}$ | head -n 1; done
/var/lib/portage/world
Heh.. this one is better ;) :
# cat /var/lib/portage/world | awk -F\/ '{system(eix -F -C $1 -e $2 -c |
head -n 1)}'
--
Bo Andresen
--
gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Re: usenet and web mail
On Thursday 06 April 2006 01:06, JimD wrote: If that doesn't do it, I can VNC in to home from work or I could try to see how fast X11 forwarding would be. And what would you use then which you cannot run locally? I mean if the only issue is that port 119 is closed then it seems you can do what I do. Create a ton of ssh tunnels and point the news/mail/whatever client towards localhost (and whatever port you forwarded it to). I currently have 8 ssh tunnels running to access external mail, news and irc servers locally. I even created an init script for this... ;) -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Re: usenet and web mail
On Thursday 06 April 2006 02:27, JimD wrote: From work I seem to only be able to get to thes ports: [SNIP] 443 is what I connect to ssh over [SNIP] I could never figure out how to get ssh tunneling working. So I just have my firewall forward port 443- 22, 21-5900, 8080-119... The with just forwarding the ports is that I don't get any compression. For example, what would be the ssh command to connect from WORK to HOME and tunnel port 5900 from HOME to WORK so that I can run vnc at WORK to localhost:: and really be connecting to HOME. ssh is listening on port 443 on HOME. If you can connect to home with a command similar to: # ssh -p 443 [EMAIL PROTECTED] And you wish to connect to newsserver.com at port 119 (news) and mailserver.com at port 143 (imap) which can be reached from home then you can create tunnels to those by: # ssh -p 443 [EMAIL PROTECTED] -L 119:newsserver.com:119 -L 143:mailserver.com:143 After that you just direct your mail and news clients towards localhost as the servername. One way of doing that is to add: 127.0.0.1 newsserver.com mailserver.com to your /etc/hosts. You can add as many tunnels as you want to the same ssh connection by just specifying more -L localport:servername_or_ip:port to the ssh command. Read man ssh for more info on ssh local port forwarding. -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Beautification - Splash
On Tuesday 04 April 2006 20:48, Lord Sauron wrote: [SNIP] and then I modified /boot/grub/menu.lst: localhost ~ # cat /boot/grub/menu.lst default 0 timeout 7 splashimage=(hd0,0)/grub/splash.xpm.gz title Current Kernel root (hd0,0) kernel /vmlinuz root=/dev/hda3 initrd /fbsplash there is the [SNIP] So now you have told it where to find the theme files. Now you just need to alter your kernel line to specify what theme to use and in which mode. An example is given in [1]. But to view full documentation of kernel parameters for splash type in a terminal (this assumes that you have splashutils installed): # zcat -c /usr/share/doc/splashutils*/kernel_parameters.gz | less Also I have just added two sections to the howto i.e. [2] and [3]. [1] http://gentoo-wiki.com/HOWTO_gensplash#GRUB_Example [2] http://gentoo-wiki.com/HOWTO_gensplash#View_contents_of_initramfs [3] http://gentoo-wiki.com/HOWTO_gensplash#Choosing_a_theme -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Beautification - Splash
On Wednesday 05 April 2006 02:33, Lord Sauron wrote:
[SNIP]
So I now use the theme livecd-2006.0 Just my $0.02, but it's sorta
difficult to change themes. Soon as I'm more bash literate, I'll try
to make a script to change themes in a easier manner
(lsauron_cngsplshthm [theme], perhaps? I dunno... Mind of a
programmer, once more). I counted 2 places I have to change to change
themes:
/etc/splash splashutils_geninitramfs... // generate new initramfs file
/boot/grub/menu.lst // point to it, and make sure kernel command line
args are nice and happy
[SNIP]
I guarantee nothing... ;) :
#!/bin/bash
RES=1400x1050
THEME=${1}
splash_geninitramfs ${THEME} -r ${RES} -g /boot/fbsplash -v
sed -i -e 's/theme:[a-z0-9.-]*/theme:'${THEME}'/' /boot/grub/menu.lst
PS: I'm playing with this GPG stuff... I want to see if any of you
can use [verify] this. I'm just curious... I had some nut case
ghosting off of my domain and email and then emailing some friends. I
was not happy. Someone suggested signing my messages. So, I'm
trying. Tell me if I'm doing it right if it's conveinient.
Message was signed on 1/1-1970 00:59 with unknown key 0x8F22D39E.
The validity of the signature cannot be verified.
[SNIP]
End of signed message
I guess not...
--
Bo Andresen
--
gentoo-user@gentoo.org mailing list
Re: [gentoo-user] OT: Application for video cutting ?
On Wednesday 05 April 2006 04:06, Meino Christian Cramer wrote: How can I manage to install qt-4 without cripple the rest of my system (qt-3 based) ? Is there a way of parallel installation/usage ? And if so -- how can I manage it gentoo-like ? # emerge -va =qt-4.1.2 These are the packages that would be merged, in order: Calculating dependencies... done! [ebuild NS ] x11-libs/qt-4.1.2 USE= 27,269 kB From man emerge: N = new, (not yet installed) S = new, SLOT installation (side-by-side versions) Of course you can just pick any version. Qt IS slotted so it will not harm your system at all. As long as you don't emerge --prune it (which would remove the old version). -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] courier-imap won't start
On Sunday 02 April 2006 05:15, JimD wrote:
I have it up and running with no issues. Maybe try to go through the
HOWTO again? What I usually do is look at the startup script and try
to make a test script to get it running. Here is an example for
starting courier-imap. Just copy it to a file and try to execute it as
root:
#!/bin/bash
ADDRESS=0
MAXDAEMONS=10
MAXPERIP=10
PIDFILE=/var/run/imapd.pid
TCPDOPTS=-nodnslookup -noidentlookup
PORT=143
MAILDIR=.maildir
exec_prefix=/usr
/usr/lib/courier-imap/couriertcpd -address=$ADDRESS \
-stderrlogger=/usr/lib/courier-imap/courierlogger \
-stderrloggername=imapd \
-maxprocs=$MAXDAEMONS -maxperip=$MAXPERIP \
-pid=$PIDFILE $TCPDOPTS \
$PORT ${exec_prefix}/sbin/imaplogin \
/usr/lib/courier-imap/courier-imapd.indirect ${MAILDIR}
This script does start it up successfully. Still the initscript doesn't. I
have tested that all the variables are set to the same things in the
initscript simply by printing them with echo and they are.
/usr/lib/courier-imap/courier-imapd.indirect doesn't exist btw...
I trying to use version 4.0.1
# eix courier -I -c
[I] net-libs/courier-authlib (0.58): courier authentication library
[I] net-mail/courier-imap (4.0.1): An IMAP daemon designed specifically for
maildirs
--
Bo Andresen
--
gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Beautification - Splash
http://gentoo-wiki.com/HOWTO_fbsplash -- Bo Andresen -- gentoo-user@gentoo.org mailing list
[gentoo-user] courier-imap won't start
I have followed http://www.gentoo.org/doc/en/virt-mail-howto.xml#doc_chap3 till chapter 3. But when I get to code listing 3.3 i.e.: # /etc/init.d/courier-imapd start # /etc/init.d/courier-imapd-ssl start # /etc/init.d/courier-pop3d start # /etc/init.d/courier-pop3d-ssl start the daemons all fail to start. There are no error messages of any kind. Nothing written to dmesg and I don't have a clue about how to figure out a reason for this. I also don't have a clue about what info might be relevant so please ask for it. Postfix does work like a charm and the certificates has been successfully created. Only the imap and pop3 daemons won't start. # /etc/init.d/courier-imapd start * Starting courier-authlib: authdaemond ...[ ok ] * Starting courier-imapd ... [ !! ] # /etc/init.d/courier-pop3d start * Starting courier-pop3d ... [ !! ] -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Issues with Dependencies
On Saturday 01 April 2006 01:52, Lord Sauron wrote: Hi, I'm having a bit more trouble. I'm trying to re-compile KDE so that maybe some functionality which didn't compile right the first time will work. However, it says I've got some broken dependencies. localhost ~ # emerge --pretend kde These are the packages that I would merge, in order: Calculating dependencies ...done! [blocks B ] =kde-base/kbounce-3.4* (is blocking kde-base/kdegames-3.4.3) [blocks B ] =kde-base/libkdegames-3.4* (is [SNIP] I don't know what to do. I don't want to loose Quanta+ or KLaptopDaemon, and I frankly didn't know Kommander was there... Please help a old Debian person who isn't used to doing this stuff by himself... Oh, I already tried emerge --depclean, but that didn't fix all the problems apparently. In the mean time I'll try remerging KLaptopDaemon, but I would like to resolve these issues to both learn how and to make life easier in the future. http://www.gentoo.org/doc/en/kde-config.xml#doc_chap2 You are trying to install the monolithic packages while what you have installed are the split packages. I would recommend that you stick with the split packages and emerge kde-meta instead of kde. That way you will have the option to uninstall packages that you don't use later on without having to remerge anything. -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Issues with Dependencies
On Saturday 01 April 2006 02:25, Lord Sauron wrote: localhost ~ # emerge --pretend kde-meta These are the packages that I would merge, in order: Calculating dependencies ...done! [blocks B ] =kde-base/kdebase-3.4* (is blocking kde-base/kdialog-3.4.1) [SNIP] [blocks B ] =kde-base/kdegraphics-3.4* (is blocking kde-base/kdegraphics-kfile-plugins-3.4.3) [SNIP] No. Still doesn't like me. I never know about the meta packages, so I could be on Gnome for a little while until I get this sorted out, huh? Seems you have a mixture of monolithic and split. Like I said before I recommend the split packages. To get those you have to unmerge the monolithic packages. So: # emerge --unmerge --ask --verbose kdebase kdegraphics # emerge --ask --verbose kde-meta It is all explained at http://www.gentoo.org/doc/en/handbook/handbook-x86.xml?part=2chap=1#doc_chap4 And please learn to cut out anything you don't reply to and reply below that which you do reply to. -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] New To Gentoo and Emerge, No ACPI in Kernel
On Wednesday 29 March 2006 21:58, Lord Sauron wrote:
Okay, here's where I've isolated the problem to. This might be a
rather lengthy explanation, so make sure you have about 15 minutes on
your hands before diving in. However, the explanation shouldn't take
long - I've never actually compiled/installed/used a kernel before.
Okay, enough apologising in advance: down to business.
I know that it correctly compiles the kernel. I put a new name for
the new kernel (test1) to try and ID it as it floats about all the
other kernels I'm too scared to delete.
# make install
Sticks it into /boot. /boot now reads
System.mapconfig.old
System.map-2.6.15-gentoo-r1 grub
System.map-2.6.15-gentoo-r1.old
initramfs-genkernel-x86-2.6.15-gentoo-r5 System.map-2.6.15-gentoo-r1test1
kernel-genkernel-x86-2.6.15-gentoo-r5
System.map-2.6.15-gentoo-r1test1.old lost+found
System.map.oldvmlinuz
boot vmlinuz-2.6.15-gentoo-r1
configvmlinuz-2.6.15-gentoo-r1.old
config-2.6.15-gentoo-r1 vmlinuz-2.6.15-gentoo-r1test1
config-2.6.15-gentoo-r1.old vmlinuz-2.6.15-gentoo-r1test1.old
config-2.6.15-gentoo-r1test1 vmlinuz.old
config-2.6.15-gentoo-r1test1.old
Not terribly exciting. However, I went to /boot/grub/menu.lst and it
reads as such:
localhost boot # cat ./grub/menu.lst
default 0
timeout 7
splashimage=(hd0,0)/grub/splash.xpm.gz
title=Gentoo Linux
root (hd0,0)
kernel /kernel-genkernel-x86-2.6.15-gentoo-r5 root=/dev/ram0
init=/linuxrc ramdisk=8192 real_root=/dev/hda3
initrd /initramfs-genkernel-x86-2.6.15-gentoo-r5
The most concerning part is the last three lines. For any kernel, it
appears to demand the kernel itself. If you'll refer back to # ls
/boot then you'll notice that kernel-2.6.15-gentoo-r1test1 isn't
there.
Sure it is. It's vmlinuz-2.6.15-gentoo-r1test1 and make install even made a
symlink to it: vmlinuz.
If you type ls -l /boot/vmlinuz it should give something that end on:
/boot/vmlinuz - vmlinuz-2.6.15-gentoo-r1test1
Nor is the initrd.
If you don't use genkernel you don't actually need an initrd. If you want one
you have to enable it. I can't tell you how since I don't use it myself.
I don't know where they might be, or if
they're not there then how to generate them.
Try adding the following to /boot/grub/menu.lst:
title Gentoo Linux test1
root (hd0,0)
kernel /vmlinuz root=/dev/hda3
This will create a new menu item in Grub during start up. If you want this
kernel to be selected by default you either add above the genkernel menu item
shown above or change the default to 1. After default you can add a fallback
line and set it to another kernel than the default.
This is a part of my menu.lst:
# Boot automatically after 30 secs.
timeout 5
# By default, boot the first entry.
default 0
# Fallback to the second entry.
fallback 1
# Reboot 5 seconds after a kernel panic
panic=5
# Nice splash image for grub :)
splashimage=(hd0,1)/grub/splash.xpm.gz
title Gentoo Linux
root (hd0,1)
kernel /vmlinuz root=/dev/hda6 video=vesafb-tng:[EMAIL PROTECTED],mtrr,ywrap
splash=silent,theme:livecd-2006.0 quiet
CONSOLE=/dev/tty1
initrd /fbsplash-livecd-2006.0-1400x1050
title Gentoo Linux (Old)
root (hd0,1)
kernel /vmlinuz.old root=/dev/hda6 video=vesafb-tng:[EMAIL PROTECTED],mtrr,ywrap
splash=verbose,theme:livecd-2006.0 quiet
CONSOLE=/dev/tty1
initrd /boot/fbsplash-livecd-2006.0-1024x768
The indented lines are on the end of the kernel line above them. My initrd's
are create by splashutils and have nothing to do with compiling the kernel.
If they are removed it boots just as well just without the livecd-2006 theme.
If I can find out those two things then I should be able to test my
new kernel and see if it actually worked.
On 3/28/06, Neil Bothwick [EMAIL PROTECTED] wrote:
On Tue, 28 Mar 2006 11:51:11 +0200, Bo Andresen wrote:
Sorry, what does YMMV mean?
Those are the resources that I use for that kind of questions:
http://en.wikipedia.org/wiki/YMMV
http://www.urbandictionary.com/define.php?term=ymmv
You can also emerge wtf.
I assume wtf will tell me what wtf stands for... is the last letter
representative of a forbidden word, by any chance?
Of course ;) :
$ wtf wtf
WTF: {what,when,where,who,why} the fuck
$ wtf ymmv
YMMV: your mileage may vary
Yeah, it's just that the wiki and sometimes the urban dictionary are often
more detailed. :)
--
Bo Andresen
--
gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Why is portage trying to pull in xorg when I don't use it...
On Thursday 30 March 2006 00:09, Daevid Vincent wrote: I have a VMWare that I use for LAMP development. I have never put Xorg on it, nor do I ever want X windows on it. Recently, when I do an 'emerge -Davut world', I see this: - vmware ~ # emerge -Davtu world These are the packages that I would merge, in reverse order: Calculating world dependencies !!! Packages for the following atoms are either all !!! masked or don't exist: sys-apps/fileutils sys-apps/textutils sys-apps/sh-utils Seems you have fileutils, textutils and sh-utils in your world file though they are not part of portage. Are you using them? If you don't then perhaps unmerge them.. That is, however, uncorrelated to the problem below. ...done! [ebuild U ] net-fs/samba-3.0.21b [3.0.14a-r2] +acl -async -automount -cups -doc -examples -kerberos -ldap -ldapsam -libclamav +mysql -oav +pam -postgres +python -quotas +readline (-selinux) -swat -syslog -winbind -xml +xml2 17,143 kB [ebuild U ] www-client/links-2.1_pre20 [2.1_pre19] -X -directfb -fbcon +gpm -javascript +jpeg -livecd +png +sdl +ssl -svga +tiff -unicode 3,768 kB [ebuild N] media-libs/libsdl-1.2.8-r1 -X -aalib -alsa -arts -dga -directfb -esd -fbcon -ggi -libcaca -nas -noaudio -noflagstrip -nojoystick -novideo +opengl +oss -pic -svga -xinerama +xv 2,541 kB ^^^ When opengl is requested links depends on virtual/opengl which is provided by x11-base/xorg-x11 given that the opengl use flag is enabled for that too... So just add -opengl to your use flags. [ebuild N] media-libs/audiofile-0.2.6-r1 365 kB [ebuild N] x11-terms/xterm-207 -Xaw3d -doc -toolbar +truetype -unicode 727 kB [nomerge ] sys-apps/utempter-0.5.5.6 [ebuild N] x11-base/xorg-x11-6.8.2-r6 -3dfx -3dnow +bitmap-fonts -cjk -debug -dlloader -dmx -doc -font-server -insecure-drivers -ipv6 -minimal -mmx -nls -nocxx +opengl +pam -sdk -sse -static +truetype-fonts +type1-fonts (-uclibc) -xprint +xv 44,705 kB -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] New To Gentoo and Emerge, No ACPI in Kernel
On Thursday 30 March 2006 02:53, Lord Sauron wrote: Later I hope to reinsert my Live CD and get the pretty stuff off of it to beautify my Gentoo. For what kind of beautifying do you need the Live CD? If you are referring to the splash theme that it uses then it is in portage. http://gentoo-wiki.com/HOWTO_fbsplash However, even before beautification I have to figure out how to mount my USB memory stick (/dev/sda1), however, that's for another thread. What you want is CONFIG_USB_STORAGE in the kernel configuration. The following is from make menuconfig: Symbol: USB_STORAGE [=m] Prompt: USB Mass Storage support Defined at drivers/usb/storage/Kconfig:9 Depends on: USB Location: - Device Drivers - USB support -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] New To Gentoo and Emerge, No ACPI in Kernel
On Tuesday 28 March 2006 04:51, Lord Sauron wrote: I'm currently running 2.6.15-r8 of suspend2_sources, so if you're using a different kernel YMMV. Sorry, what does YMMV mean? Those are the resources that I use for that kind of questions: http://en.wikipedia.org/wiki/YMMV http://www.urbandictionary.com/define.php?term=ymmv -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Card games in portage
On Tuesday 28 March 2006 08:37, Teresa and Dale wrote: Hi folks, My girlfriends mom comes over sometimes and she likes to play card games on the computer. Anybody have a list of the games in portage that are card games? You know, like Solitaire or something like that. I have the ones that come with KDE and Pysol too. Just looking for some more for her. She's used to windoze by the way. Also, is there a place to go to find out which ones compare to what ever that is on windoze? I know they have to change the names sonce they may be copyrighted or something. I don't know anything about pysol but these games are in Windows: # eix -C games -r 'hearts|freecell' * games-board/hearts Available versions: 1.98 Installed: none Homepage:http://hearts.luispedro.org/index.php Description: clone of the hearts game for KDE that comes with Windows * games-board/xfreecell Available versions: 1.0.5b Installed: 1.0.5b Homepage:http://www2.giganet.net/~nakayama/ Description: A freecell game for X -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] What's bringing in Mozzilla?
On Sunday 26 March 2006 20:50, Ernie Schroder wrote: from mozilla-firefox-bin ebuild: Of course you mean from mplayerplug-in ebuild. ;) DEPEND==media-video/mplayer-1.0_pre5 gecko-sdk? ( net-libs/gecko-sdk ) !gecko-sdk? ( || ( =www-client/mozilla-1.6 www-client/mozilla-firefox ) [SNIP] Could I edit the ebuild like so? DEPEND==media-video/mplayer-1.0_pre5 gecko-sdk? ( net-libs/gecko-sdk ) !gecko-sdk? ( || ( =www-client/mozilla-1.6 www-client/mozilla-firefox www-client/mozilla-firefox-bin ) ) [SNIP] As you have surely realized by now this suggests that mplayerplug-in and mozilla-firefox-bin are incompatible. If this is not correct and they are compatible then you should file a bug. To test it copy the ebuild to an overlay and modify it as you have suggested. If you do not copy it to an overlay your changes will be wiped the next you emerge --sync. -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] What's bringing in Mozzilla?
On Sunday 26 March 2006 20:57, Ernie Schroder wrote: I follow the logic, but, why won't mozilla-firefox-bin satisfy portage? moz launcher has been on the machine for a LONG time as has firefox-bin and has never wanted mozilla or mozilla-firefox before If you compare mplayerplug-in-3.21 with the previous version then you see that this is a new dependency of mplayerplug-in when the gecko-sdk use flag is disabled. So probably they added mozilla or mozilla-firefox and forgot about mozilla-firefox-bin or mozilla-firefox-bin is in fact incompatible with this version of mplayerplug-in. -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] What's bringing in Mozzilla?
On Sunday 26 March 2006 21:16, Manuel McLure wrote: As other have said, mplayer-plugin wants to pull in mozilla-firefox. The reason for this is that it needs the Mozilla source libraries to build against. One workaround for this is to do So there is a good reason for mozilla-firefox-bin not satisfying it's depency after all. :) -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Splash livecd-2006.0 not working
On Friday 24 March 2006 04:59, Richard Fish wrote: On 3/23/06, Bo Andresen [EMAIL PROTECTED] wrote: root (hd0,1) Filesystem type is reiserfs, partition type 0x83 kernel /vmlinuz root=/dev/hda6 video=vesafb-tng:[EMAIL PROTECTED],mtrr,ywrap spla sh=verbose,theme:live-cd-2006.0 quiet CONSOLE=/dev/tty1 Did you notice you have a typo here? That should be splash=verbose,theme:livecd-2006.0 ^^ Oops, thanks... But still: Booting 'Gentoo Linux' root (hd0,1) Filesystem type is reiserfs, partition type 0x83 kernel /vmlinuz root=/dev/hda6 video=vesafb-tng:[EMAIL PROTECTED],mtrr,ywrap splash=verbose,theme:livecd-2006.0 quiet CONSOLE=/dev/tty1 [Linux-bzImage, setup=0x1c00, size=0x19d4cd] initrd /fbsplash-livecd-2006.0-1400x1050 [Linux-initrd @ 0x1ff68000, 0x87c57 bytes] Uncompressing Linux... Ok, booting the kernel. Can't open config file /etc/splash/livecd-2006.0/1400x1050.cfg. Failed to load image (null). Failed to get verbose splash image. The theme does, however, get loaded when the boot service splash is started. I think it lacks some activity though. -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] How to tar?
On Friday 24 March 2006 15:36, Boyd Stephen Smith Jr. wrote: Michael Kintzios wrote: what I think is needed here is untarring of the archive, while untarred data is dynamically deleted immediately after untarred to make space for more data to be untarred . . . do I make sense? Yes, but GNU tar cannot do that, it can only do one command at a time, either --extract or --delete or ... Yes, that's why I was hoping that some clever bash-ery may be able to pipe the lot together. Perhaps: tar xvf gentoo_usr.tar | while read file; do tar --delete f gentoo_usr.tar $file; done That might just screw up your tar file and/or extract junk; I didn't test it at all. ROFL. No that won't work. ;) You cannot delete while extracting and when extraction is completed there is no point. This, however, does work: tar tf gentoo_usr.tar | sort -r | while read file; do tar -xf gentoo_usr.tar $file tar --delete -f gentoo_usr.tar $file; done First of all the dash before f when deleting is necessary. That's just syntax. Secondly the sort -r is VERY important to make sure it extracts the deepest files (in terms of path) first then deletes them. Both -x and --delete or recursive by default. The problem with this, however, is that it only works with a tar file. Apparently it is not possible to delete a file from a compressed tar file. -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Splash livecd-2006.0 not working [SOLVED]
On Friday 24 March 2006 17:19, Richard Fish wrote: On 3/24/06, Bo Andresen [EMAIL PROTECTED] wrote: root (hd0,1) Filesystem type is reiserfs, partition type 0x83 kernel /vmlinuz root=/dev/hda6 video=vesafb-tng:[EMAIL PROTECTED],mtrr,ywrap splash=verbose,theme:livecd-2006.0 quiet CONSOLE=/dev/tty1 [Linux-bzImage, setup=0x1c00, size=0x19d4cd] initrd /fbsplash-livecd-2006.0-1400x1050 [Linux-initrd @ 0x1ff68000, 0x87c57 bytes] Can you check the initramfs to ensure that it contains the necessary files? zcat /boot/fbsplash-livecd-2006.0-1400x1050 | cpio --list | grep splash Thank you very much. This was exactly what I was looking for. # zcat /boot/fbsplash-livecd-2005.1-1400x1050 | cpio --list | grep splash 1527 blocks etc/splash etc/splash/suspend2 etc/splash/suspend2/Vera.ttf etc/splash/suspend2/1400x1050.cfg etc/splash/suspend2/images etc/splash/suspend2/images/text.png etc/splash/suspend2/images/background-1400x1050.png etc/splash/suspend2/images/verbose-1400x1050.png sbin/splash_helper Obviously I made a mistake two days ago when I made this. I have tried this a LOT of times so I do not think it is possible that I have made this mistake every time. I am more uncertain about the dash that you pointed out previously. :( I also updated baselayout to 1.12.0_pre a few days ago. I don't know if that has any relevance.. Anyhow.. Thank you. It now works like a charm. I think I'll add this command to the wiki since I could have used it earlier... :) -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Have portage lost its memory?
On Thursday 23 March 2006 15:07, Jules Colding wrote:
I used to give the shell prompts different colours on different
machines to help avoid this. Or rather, the local one would always be
the same colour, but shells under ssh sessions were colour-coded by
machine.
I've lost the script I wrote for this somewhere in the mists of time
(if I remember right, it was copied and hacked from a bash prompt
example that colour-coded according to the login type: ssh, telnet,
local, etc.)
Someday I might get round to recreating it...
That would be helpful.
Here is an example that you could put in your .bashrc:
# Is this an ssh connection?
if [[ ! -z ${SSH_TTY} ]]; then
# Set prompt to \green([EMAIL PROTECTED]) \blue($PWD \$) green(..
PS1='\[\033[01;[EMAIL PROTECTED] \[\033[01;34m\]\w \$ \[\033[01;32m\]'
# Not an ssh connection
else
# Set prompt to \green([EMAIL PROTECTED]) \blue($PWD \$) black(..
PS1='\[\033[01;[EMAIL PROTECTED] \[\033[01;34m\]\w \$ \[\033[00m\]'
fi
If you want other colors or whatever refer to man console_codes.
--
Bo Andresen
--
gentoo-user@gentoo.org mailing list
Re: [gentoo-user] How to tar?
On Thursday 23 March 2006 21:24, Michael Kintzios wrote: What should I run to untar the rest of /usr (excluding /usr/portage) into /dev/hda3 and at the same time delete it from within the gentoo_usr.tgz archive, so that I get some space in /dev/hda2 to untar /usr/portage? Really, what I think is needed here is untarring of the archive, while untarred data is dynamically deleted immediately after untarred to make space for more data to be untarred . . . do I make sense? You don't have to scp the archieve to the machine before unpacking it. http://gentoo-wiki.com/HOWTO_Backup#Securely_backing_up_a_filesystem_on_a_remote_machine Also if you look at man tar you'll find tar --exclude PATTERN HtH -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] How to tar?
On Thursday 23 March 2006 22:52, Bo Andresen wrote: On Thursday 23 March 2006 21:24, Michael Kintzios wrote: What should I run to untar the rest of /usr (excluding /usr/portage) into /dev/hda3 and at the same time delete it from within the gentoo_usr.tgz archive, so that I get some space in /dev/hda2 to untar /usr/portage? Really, what I think is needed here is untarring of the archive, while untarred data is dynamically deleted immediately after untarred to make space for more data to be untarred . . . do I make sense? You don't have to scp the archieve to the machine before unpacking it. http://gentoo-wiki.com/HOWTO_Backup#Securely_backing_up_a_filesystem_on_a_r emote_machine Perhaps that link wasn't as useful to you as I thought when I transmitted it. Here are a couple of other examples. I think it requires GNU tar. This compacts data recursively from /from/path and using gzip, pipes it through ssh and extracts it into /to/path: # tar -zcf - /from/path | ssh desktop.homelinux.com tar -C /to/path -xzf - And this just pipes through ssh and extracts using bunzip2 to /to/path on remote machine # cat file.tar.bz2 | ssh desktop.homelinux.com tar -C /to/path -xjf - -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] sudo echo
On Thursday 23 March 2006 23:38, Renat Golubchyk wrote: On Thu, 23 Mar 2006 18:27:46 -0300 Daniel da Veiga [EMAIL PROTECTED] wrote: Sudo takes a command as parameter, enclose the whole command in quotes and try again, like this: sudo echo app-portage/porthole ~* /etc/portage/package.keywords ^ ^ ^ ^ Careful with those quotation marks - you might want to escape them ;-) I would use single quotes on the outside to avoid the confusion: sudo 'echo app-portage/porthole ~* /etc/portage/package.keywords' Yeah, and the neat thing ... it still doesn't work... ;) As Daniel admitted in reply to Hollys mail in this thread he had an alias for sudo. -- Bo Andresen -- gentoo-user@gentoo.org mailing list
[gentoo-user] Splash livecd-2006.0 not working
I did follow this guide a long time ago and again yesterday. http://gentoo-wiki.com/HOWTO_fbsplash When I use emergence as theme everything is working just like it is supposed to. But when I change to livecd-2005.1 or livecd-2006.0 (I think any theme that has activity instead of just a picture) I get this error during bootup: = Booting 'Gentoo Linux' root (hd0,1) Filesystem type is reiserfs, partition type 0x83 kernel /vmlinuz root=/dev/hda6 video=vesafb-tng:[EMAIL PROTECTED],mtrr,ywrap spla sh=verbose,theme:live-cd-2006.0 quiet CONSOLE=/dev/tty1 [Linux-bzImage, setup=0x1c00, size=0x19d4cd] initrd /boot/fbsplash-livecd-2006.0-1400x1050 [Linux-initrd @ 0x1ff68000, 0x87c57 bytes] Uncompressing Linux... Ok, booting the kernel. Can't open config file /etc/splash/live-cd-2006.0/1400x1050.cfg. Failed to load image (null). Failed to get verbose splash image. = From grub: = title Gentoo Linux root (hd0,1) kernel /vmlinuz root=/dev/hda6 video=vesafb-tng:[EMAIL PROTECTED],mtrr,ywrap splash=verbose,theme:live-cd-2006.0 quiet CONSOLE=/dev/tty1 initrd /boot/fbsplash-livecd-2006.0-1400x1050 = = # ls -l /boot [...] -rw-r--r-- 1 root root 556119 Mar 23 18:42 fbsplash-livecd-2006.0-1400x1050 lrwxrwxrwx 1 root root 26 Mar 23 00:03 vmlinuz - vmlinuz-2.6.15-suspend2-r8 -rw-r--r-- 1 root root 1700557 Mar 23 00:03 vmlinuz-2.6.15-suspend2-r8 [...] = Until yesterday I was using gentoo-sources but that has not changed anything. I have had this problem for a long time. Obviously something is wrong with the initramfs. It was created by this command: # splash_geninitramfs -g /boot/fbsplash-livecd-2006.0-1400x1050 -r 1400x1050 -v livecd-2006.0 If while booted a type: # splash_manager --theme livecd-2006.0 -c set the splash theme is loaded successfully on tty1. Any ideas? -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] sudo echo
On Thursday 23 March 2006 23:48, JimD wrote:
addkey()
{
sudo sh -c echo $* /etc/portage/package.keywords
}
For keywording I prefer to use this script:
http://users.cybercity.dk/~dsl89966/keix
It allows me to do:
$ eix porth
* app-portage/porthole
Available versions: ~0.4.1 [M]0.5.0
Installed: none
Homepage:http://porthole.sourceforge.net
Description: A GTK+-based frontend to Portage
Found 1 matches
$ sudo keix porth
Do you wish to add '=app-portage/porthole-0.4* ~x86' to package.keywords?
(Yes/no)
Adding '=app-portage/porthole-0.4* ~x86' to package.keywords
$
Of course it requires that app-portage/eix is installed and updated.
--
Bo Andresen
--
gentoo-user@gentoo.org mailing list
Re: [gentoo-user] bastard.sh - mass unmasker/keywords utility
On Tuesday 21 March 2006 17:20, Joshua Schmidlkofer wrote: Yeah, so? *crickets* Thanks, the d.g.o. was pretty clear, but I can imagine that confusing people, the referenced document was next to useless, did you read it? Well, I did have a look at it and obviously it is intended for developers only. What I wanted people to read before installing Gnome 2.14* is this: On Tuesday 21 March 2006 15:39, Bo Andresen wrote: # emerge -vp =gnome-2.14* [...] #Don't unmask these and don't file bugs for them [...] But maybe that's just me... ;) -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Init sequence
On Wednesday 22 March 2006 20:26, Sergio Polini wrote: I'ld like to know how the sequence of init scripts is set up. Did you look in the Gentoo handbook? http://www.gentoo.org/doc/en/handbook/handbook-x86.xml?part=2chap=4 If after reading that you still have unanswered questions then ask again. Moreover, I'd like to know if it's possible to verify the actual sequence of the init scripts. I can look at the commands that a make command would execute but do not execute them, or at the problems reported by fsck avoiding to repair them, by the -n option. Could I check an init sequence in a similar way? I don't think so. But it will check syntax of the entire script if you run any command on it. Only syntax though. -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] bastard.sh - mass unmasker/keywords utility
On Tuesday 21 March 2006 10:47, Joshua Schmidlkofer wrote: So, WFM [works for me], hope it's useful to others. In any case, this is a resend of the script, since I got some ambiguous 'blocked message' errors, I put it up on my website, and left it for all: Secure: https://embassy.asylumware.com/projects/asylumware/wiki/bastard Plain: http://embassy.asylumware.com/projects/asylumware/wiki/bastard # emerge -vp =gnome-2.14* These are the packages that would be merged, in order: Calculating dependencies !!! All ebuilds that could satisfy =gnome-2.14* have been masked. !!! One of the following masked packages is required to complete your request: - gnome-base/gnome-2.14.0 (masked by: package.mask, ~x86 keyword) # John N. Laliberte [EMAIL PROTECTED] (12 Mar 2006) # GNOME 2.14 mask. You must follow instructions here: # http://d.g.o/~allanonjl/gnome/2.13/adding.from.overlay.txt # for adding files from our overlay. #Most of these packages will break/not compile because of eclass #changes that won't be made until every package is in the tree. #Don't unmask these and don't file bugs for them # Start GNOME 2.14 mask Did you the read comments here (d.g.o refers to dev.gentoo.org in case anyone is wondering)? I most certainly wouldn't go ahead and unmask any package with that kind of explanation why it was masked in the first place. What bothers me about this, however, is not the fact that you did unmask it but rather the fact the you leave a script here which is supposed to be able to unmask and unkeyword any package without giving any kind of warnings about this. IMHO any script that is made public and which does what your script is doing should print out the reason why each package it unmasks was originally masked and perhaps even ask for confirmation. Also I think it is a bit amusing that running your script without any arguments tells me that I should include the versioned name of any package that I want unmasked/unkeyworded. And then giving any kind of argument(s) makes it unmask/unkeyword gnome-2.14*. Without checking the argument(s) that I gave. I know.. it's just a minor bug. I also think (without knowing it) that it will in fact work for most packages when that minor bug is corrected. ;) It does, however, work for gnome-2.14.0 and it does add a LOT of lines to package.keyword and package.mask. E.g. media-libs/gst-plugins-base adds five lines to package.keyword i.e. version 0.10.0 to 0.10.4 on a line each. I don't think the results will ever differ on the first two version numbers so I think the optimal solution for this would be to just add version 0.10*. I'm not sure that it will never differ though. Just for the record I am not trying to offend you here. This is supposed to be constructive critisism. ;) -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Lilypond version
On Tuesday 21 March 2006 23:44, Allan Spagnol Comar wrote: Hi, I don't know where should I post this kind of message, but I need to use lilypond as one of my working tools. I got stunned when I saw today that lilypond version available at portage tree is 2.0.3 the latest stable version is 2.6.5 and 2.0.3 is extreme old. Is there a reason that lilypond version is not up to date and should I build myself ( without portage ) if I want to used it ? -- An application asked: Requeires Windows 9x, NT4 or better, so I´ve installed Linux The newest version of lilypond in portage is 2.5.2. It is still in testing i.e. ~ARCH. $ eix lilypond * media-sound/lilypond Available versions: 2.0.3 ~2.2.4 ~2.2.6 ~2.4.2 ~2.5.2 Installed: none Homepage:http://lilypond.org/ Description: GNU Music Typesetter Also you should search bugs.gentoo.org for the package before asking this kind of question. That would have brought you to this bug: http://bugs.gentoo.org/show_bug.cgi?id=97574 If you wish to use an ebuild that is not in portage have a look at this howto: http://gentoo-wiki.com/HOWTO_Installing_3rd_Party_Ebuilds -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] amd64 Installer LiveCD?
On Sunday 19 March 2006 22:27, JimD wrote: Is there a liveCD for amd64? The amd64 handbook part 1, chapter3 under the Do I need Networking? section states: The stage3 file built by the amd64 Installer LiveCD is optimized for generic amd64 usage and uses NPTL. However looking on the download page I only see the LiveCD for x86. I really liked the LiveCD for x86 because I was able to browse the web/email while the installer did its thing : ) I am getting an amd64 system tomorrow and will be putting Gentoo on it and it would nice to have a GUI desktop during the install. http://www.gentoo.org/doc/en/handbook/2006.0/handbook-amd64.xml?part=1chap=2#doc_chap3 You can download the Universal Installation CD (and, if you want to, the Packages CD as well) from one of our mirrors. The Installation CD is located in the releases/amd64/2006.0/installcd directory; the Package CD is located in the releases/amd64/2006.0/packagecd directory. -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] amd64 Installer LiveCD?
On Monday 20 March 2006 01:32, JimD wrote: Bo Andresen wrote: http://www.gentoo.org/doc/en/handbook/2006.0/handbook-amd64.xml?part=1ch ap=2#doc_chap3 Is that the same as the *live* CD? The x86 live CD for 2006.0 had a GUI that I could use why the installer did its thing. I thought the regular universal installer was console only? Yes. The versioned 2006.0 Universal Installation CD aka the Installer LiveCD is the one with a GUI. Admittedly they haven't made that very clear but you will see it if you continue to chapter 3... -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] amd64 Installer LiveCD?
On Monday 20 March 2006 02:19, Boyd Stephen Smith Jr. wrote: On Sunday 19 March 2006 18:49, Bo Andresen [EMAIL PROTECTED] wrote Yes. The versioned 2006.0 Universal Installation CD aka the Installer LiveCD is the one with a GUI. Admittedly they haven't made that very clear but you will see it if you continue to chapter 3... No. They are not the same. However, there are 3 different gentoo LiveCDs, each appropriate for a different type of install. On the tracker we have (and I'm seeding), among other things: livecd-amd64-installer-2006.0 (LiveCD with Gentoo Installer) install-amd64-universal-2006.0 (Universal LiveCD [no installer]) packages-amd64-2006.0 (Packages CD) and install-amd64-minimal-2006.0 (Minimal LiveCD [no installer]) The livecd-installer is new for this release. The install-universal and install-minimal are the traditional live CDs, with and without stages and snapshots (and maybe some packages?). The package CD is really only useful if you are installing the GRP (so, you can't tweak USE flags) but will get you a system with Gnome or KDE and a number of useful applications as fast as a standard binary distribution. Yes you are right. Thanks for clearing that out. I misread the first section of Chapter 3. The in livecd-amd64-installer-2006.0 is located in releases/amd64/2006.0/livecd on the mirrors in opposition to my statement in my earlier replies. Actually I just downloaded the x86 version yesterday because I wanted to see it. It is neat that you can actually do some real work on the computer while installing with the old methods using the new installer cd. :) Seems to me that the have picked the greatest possible approach to this. -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Re: A mater of style in gentoo booting proccess...
On 3/18/06, Remy Blank [EMAIL PROTECTED] wrote:
echo $COLUMNS
On Saturday 18 March 2006 18:28, apix kernel wrote:
Thanks, that was i was looking for.. :)
Also have a look at
# man console_codes
And just an example (one line):
echo -e '\x1B[01;32m * \x1B[0m Starting xdm ...\x1B['
`echo ${COLUMNS}-12|bc`'G\x1B[01;34m[ \x1B[01;32mok \x1B[01;34m]\x1B[0m'
;)
--
Bo Andresen
--
gentoo-user@gentoo.org mailing list
Re: [gentoo-user] What on Earth is Portage doing for so long?
On Friday 17 March 2006 15:02, Bruno Lustosa wrote: Good question :) But one way to speed this up is to use CDB, so that instead of keeping cache in separate files, it's all in one, resulting in a *much* faster cache update. There are directions here: http://gentoo-wiki.com/TIP_speed_up_portage_with_cdb And if you go back one step you'll find other ways to speed up Portage. http://gentoo-wiki.com/Index:TIP#Portage_Speed What I do is keep $PORTDIR and $DISTDIR on seperate partitions. Having a small partition for $PORTDIR ensures that all of Portage will be in the same place on the harddrive. $DISTDIR is on a separate partition because otherwise it wouldn't be a small partition. ;) $PKGDIR is not on a separate partition because I don't use it. Also Portage 2.1 (which is not supported by cdb) is nice. :) -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] 2006.0 minimal install howto
On Friday 17 March 2006 22:23, maxim wexler wrote: Have you read the install handbooks yet? Yes. To paraphrase what I found: if you want you can use the minimal install CD. If you got a link to something a trifle more verbose I'd sure like to follow it. http://www.gentoo.org/doc/en/index.xml?catid=install#doc_chap2 http://www.gentoo.org/doc/en/handbook/index.xml Pick your architecture and start reading. It's all covered quite throughly. If you have a specific problem with it then tell us exactly how far you have gotten and what the problem is. -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] cdb with portage 2.1 (was: CVSup vs Gentoo's Rsync)
On Friday 17 March 2006 19:24, Alexander Skwar wrote: Is anyone using CDB with Portage 2.1? # grep ewarn /usr/portage/sys-apps/portage/portage-2.1_pre6-r3.ebuild ewarn This series contains a completely rewritten caching framework. ewarn If you are using any cache modules (such as the CDB cache ewarn module) portage will not work until they have been disabled. -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] cdb with portage 2.1
On Saturday 18 March 2006 07:16, Alexander Skwar wrote: Yes, I know, but please read the thread. Oh sorry. I probably shouldn't write to a mailing list when that tired... ;) -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] overnetclc: error while loading shared libraries: libstdc++.so.5: cannot open
On Thursday 16 March 2006 06:57, Boyd Stephen Smith Jr. wrote: On Wednesday 15 March 2006 08:29, Bo Andresen [EMAIL PROTECTED] wrote about '[gentoo-user] overnetclc: error while loading shared libraries: libstdc++.so.5: cannot open': I am having a problem with overnet on an amd64 computer. For some reason it connat load libstdc++.so.5. $ overnetclc overnetclc: error while loading shared libraries: libstdc++.so.5: cannot open shared object file: No such file or directory I checked and realized that I did not have this file so I emerged libstdc++-v3. This did get me the file in /usr/lib64. Still no luck. I suppose the problem is that overnet is a binary package which requires a 32 bit library. Still I don't know how to solve this problem. Perhaps the problem is that I need multilib but I don't know how to select that since the multilib use flag is missing. Multilib is handled by profiles now. From the part of your post I've trimmed (DOH) you are using a multilib profile, so the use flag is treated as always on. But shouldn't libstdc++-v3 install both 64 bit and 32 bit libraries then? It most certainly doesn't. # equery files libstdc++-v3 [ Searching for packages matching libstdc++-v3... ] * Contents of sys-libs/libstdc++-v3-3.3.6: /etc /etc/env.d /etc/env.d/99libstdc++ /usr /usr/lib64 /usr/lib64/libstdc++-v3 /usr/lib64/libstdc++-v3/libstdc++.so.5 - libstdc++.so.5.0.7 /usr/lib64/libstdc++-v3/libstdc++.so.5.0.7 Also this leads me to believe that the use flag is actually off. This may have no importance (because of the profile) but I am quite confused on how this is supposed to work. # equery uses libstdc++-v3 [SNIP] [ Found these USE variables for sys-libs/libstdc++-v3-3.3.6 ] U I [SNIP] - - multilib : On 64bit systems, if you want to be able to compile 32bit and 64bit binaries [SNIP] Also, you might need one (or more) of the packages from app-emulation/emul-x86-*. # eix -s emul -S libstdc++ * app-emulation/emul-linux-x86-compat Available versions: 1.0 1.0-r1 Installed: none Homepage:http://www.gentoo.org/ Description: emul-linux-x86 version of lib-compat, with the addition of a 32bit libgcc_s and the libstdc++ versions provided by gcc 3.3 and 3.4 for non-multilib systems. The description leads me to believe that this is for non-multilib systems. But I want multilib. Am I missing anything here? If anyone can direct me towards some proper documentation of how multilib is supposed to work I would appreciate that. And as always thanks for your replies. :) -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] overnetclc: error while loading shared libraries: libstdc++.so.5: cannot open [SOLVED]
On Thursday 16 March 2006 14:32, Boyd Stephen Smith Jr. wrote: # eix -s emul -S libstdc++ * app-emulation/emul-linux-x86-compat Available versions: 1.0 1.0-r1 Installed: none Homepage:http://www.gentoo.org/ Description: emul-linux-x86 version of lib-compat, with the addition of a 32bit libgcc_s and the libstdc++ versions provided by gcc 3.3 and 3.4 for non-multilib systems. The description leads me to believe that this is for non-multilib systems. But I want multilib. Am I missing anything here? The description must be out of date with current practice; from my system (a multilib system): # equery b libstdc++.so.5 [ Searching for file(s) libstdc++.so.5 in *... ] app-emulation/emul-linux-x86-compat-1.0-r1 (/emul/linux/x86/usr/lib/libstdc++.so.5 - libstdc++.so.5.0.5) games-fps/doom3-1.3.1302-r1 (/opt/doom3/libstdc++.so.5) As you can see from your own mail emul-linux-x86-compat does not install any libs into /usr/lib32. Therefore I still think that the description may be correct and that package may be unaware of multilib being enabled. Nonetheless it does solve the problem and that's good enough for me. :) On Thursday 16 March 2006 14:30, Boyd Stephen Smith Jr. wrote: From http://www.gentoo.ro/doc/en/gentoo-amd64-faq.xml#multilib : What is multilib and how can I use it? Every AMD64 processor is able to run 32-bit code as well as 64-bit code. However, when you have a 32-bit application, you are unable to mix it with 64-bit libraries or vice versa. You can, however, natively run 32-bit applications if all shared libraries it needs are available as 32-bit objects. You can choose whether you want multilib support or not by selecting the according profile. The default is a multilib-enabled profile. While this does explain that multilib is in fact enabled it tells me nothing about how it is supposed to work. Just above it [1] there is a section about the emul-linux-x86 packages which apparently explains that emul-linux-x86-compat should have been a dependency of overnet. But it does not state anything about multilib being required for those emul-linux-x86 packages to work. And if multilib isn't required for it then I am uncertain about what good it actually does. [1] http://www.gentoo.ro/doc/en/gentoo-amd64-faq.xml#emul32 -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Re: QT 4
On Thursday 16 March 2006 23:43, Shawn Haggett wrote: Although I don't know why portage doesn't also want to install the 4.1.1 version of qt if they are slotted... That's because qt is neither in the world file or a depency of any package which is in the world file. It wants to upgrade to newest qt-3* because that is a depency of one or more packages in the world file. -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Re: QT 4
On Thursday 16 March 2006 23:03, James wrote: Dmitry S. Makovey dmitry at athabascau.ca writes: since packages you use (I assume KDE etc.) are not using qt4 (i.e. require specifically qt3 branch) portage doesn't find any reasons to bump version of qt. AFAIR Qt is a slotted package and you can safely go ahead and do emerge =x11-libs/qt-4.1.1 but you packages wouldn't use it. Ok this kinda makes sense. But what exactly is a 'slotted package', Slotted means that several versions of a package can coexist at the same system. In the case of qt version 3 goes into /usr/qt/3 and version 4 goes into /usr/qt/4. how do I determine when a packages is slotted, and where do I read more about 'slotted'? Have a look at 'man emerge' # emerge -vp qt These are the packages that would be merged, in order: Calculating dependencies... done! [ebuild NS ] x11-libs/qt-4.1.1 USE=... 27,110 kB ^^^ The 'S' means that it is slotted. On Thursday 16 March 2006 20:11, James wrote: I really confused as this is not very clear what the hardmasked, testing and stable versions of QT? I've been told that QT 4 is 'very young' which I interpret as unstable and buggy. How do I tell if qt-4.1.1 is hard masked or testing as I've never seen this before. I would recommend app-portage/eix as it gives a clear overview of the available versions of any package. # eix -e qt * x11-libs/qt Available versions: 3.3.4-r8 ~3.3.4-r9 [M]3.3.5 [M]3.3.5-r1 4.1.0-r1 4.1.0-r2 4.1.1 Installed: 3.3.4-r8 Homepage:http://www.trolltech.com/ Description: The Qt toolkit is a comprehensive C++ application development framework. Versions that are prefixed by [M] are hard masked usually because they are broken. They should not be unmasked without knowing why they were originally masked. Versions prefixed by a ~ are masked by ~ARCH i.e. testing and can be used by adding them to package.keyword. eix respects package.* an hence 4.1.0-r1 is shown as stable because of this: # grep x11-libs/qt /etc/portage/package.keywords =x11-libs/qt-4.1* ~x86 As you can see you can specify versions to unmask if you want a testing version of qt-4.1* but stable for any other version. This is explained in 'man portage'. To see why a package is hard masked you have to look in the package.mask files. Packages are masked on several levels. Both general and profile specific. On my system I am using the default-linux/x86/2006.0 profile. # ls -ld /etc/make.profile lrwxrwxrwx 1 root root 48 Mar 10 02:02 /etc/make.profile - ../usr/portage/profiles/default-linux/x86/2006.0 Therefore packages on my system might be masked in the following locations (note that not all of those files actually exist because no packages are hard masked on some of those levels. Still these are the locations where portage will look.): /usr/portage/profiles/default-linux/x86/2006.0/package.mask /usr/portage/profiles/default-linux/x86/package.mask /usr/portage/profiles/default-linux/package.mask /usr/portage/profiles/package.mask And the reason why qt-3.3.5 is hard masked: # grep -A 2 Qt /usr/portage/profiles/package.mask # Qt-3.3.5 causes a lot of compilation failures. # See bug #106402. ~x11-libs/qt-3.3.5 But qt-4 is not hard masked so it should be working or at least it won't break anything. ;) HtH -- Bo Andresen -- gentoo-user@gentoo.org mailing list
[gentoo-user] overnetclc: error while loading shared libraries: libstdc++.so.5: cannot open
I am having a problem with overnet on an amd64 computer. For some reason it connat load libstdc++.so.5. $ overnetclc overnetclc: error while loading shared libraries: libstdc++.so.5: cannot open shared object file: No such file or directory I checked and realized that I did not have this file so I emerged libstdc++-v3. This did get me the file in /usr/lib64. Still no luck. I suppose the problem is that overnet is a binary package which requires a 32 bit library. Still I don't know how to solve this problem. Perhaps the problem is that I need multilib but I don't know how to select that since the multilib use flag is missing. # equery hasuse multilib [ Searching for USE flag multilib in all categories among: ] * installed packages [I--] [ ] sys-libs/libstdc++-v3-3.3.4 (5) [I--] [ -] sys-libs/glibc-2.3.5-r2 (2.2) [I--] [ -] sys-devel/gcc-3.4.4-r1 (3.4) # emerge -vp `equery hasuse multilib | sed -e 's/^/=/'` These are the packages that would be merged, in order: Calculating dependencies... done! [ebuild R ] sys-libs/libstdc++-v3-3.3.4 USE=nls nptl -build 0 kB [ebuild R ] sys-libs/glibc-2.3.5-r2 USE=nls nptl nptlonly -build -erandom -glibc-compat20 -glibc-omitfp -hardened -linuxthreads-tls -pic -profile -userlocales 0 kB [ebuild R ] sys-devel/gcc-3.4.4-r1 USE=fortran gtk nls -bootstrap -boundschecking -build -gcj -hardened -ip28 -multislot -nocxx -nopie -nossp -objc -vanilla 0 kB Total size of downloads: 0 kB # ls -ld /etc/make.profile lrwxrwxrwx 1 root root 50 Feb 24 14:25 /etc/make.profile - ../usr/portage/profiles/default-linux/amd64/2006.0 Any hints on this would be appreciated. If you need any additional information please ask. -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] how to enable glx
On Friday 10 March 2006 03:16, [EMAIL PROTECTED] wrote: I enabled the glx, but it seems the speed is still slow. $ glxinfo | grep rendering direct rendering: Yes The output of glxgears is about 130 FPS. I use intel_agp. As I have stated before I get around 230 FPS when [EMAIL PROTECTED] is running and it jumps to around 1300 FPS when I stop [EMAIL PROTECTED] Nothing else changed. This is an Radeon 9000 Mobility. Make sure you have nothing else running and see if that makes a difference. ~ $ glxgears 1001 frames in 5.0 seconds = 200.200 FPS 1136 frames in 5.0 seconds = 227.200 FPS 1186 frames in 5.0 seconds = 237.200 FPS 1149 frames in 5.0 seconds = 229.800 FPS 1185 frames in 5.0 seconds = 237.000 FPS ~ $ sudo /etc/init.d/foldingathome stop * Stopping [EMAIL PROTECTED] on CPU 1 ... [ ok ] ~ $ glxgears 6299 frames in 5.0 seconds = 1259.800 FPS 6744 frames in 5.0 seconds = 1348.800 FPS 6941 frames in 5.0 seconds = 1388.200 FPS 6897 frames in 5.0 seconds = 1379.400 FPS 6801 frames in 5.0 seconds = 1360.200 FPS -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Moving to xorg-x11-7.0 (Modular X)
On Thursday 09 March 2006 07:49, Boyd Stephen Smith Jr. wrote: Also, it's really confusing trying to figure out /why/ they are bringing in virtual-x11. emerge -pvt is giving confusing results; basically saying that virtual-x11 is coming in because of packages which don't even have an optional dependency on X. :( I tried using equery g, but that actually says the packages I'm trying to bring in /won't/ need virtual/x11-6.8. AFAICT equery detects compile time dependencies (DEPEND) and run time depencies (RDEPEND) but not post dependencies (PDEPEND). So I think you need to look for PDEPEND in the ebuilds manually. -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Re: Switching to Gentoo
On Thursday 09 March 2006 21:39, Harry Putnam wrote: # equery belongs `which dd` [ Searching for file(s) /usr/bin/dd in *... ] sys-apps/coreutils-5.94-r1 (/usr/bin/dd - /bin/dd) That gives you the address of the package this tool resides in on the portage tree (/usr/portage/ is dropped off the front end) No reason to use which for this. # equery b dd [ Searching for file(s) dd in *... ] sys-apps/coreutils-5.2.1-r7 (/bin/dd) sys-apps/coreutils-5.2.1-r7 (/usr/bin/dd - /bin/dd) -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Switching to Gentoo
On Thursday 09 March 2006 20:16, Jim wrote:
I have now been using Gentoo for a few days and I am running into some
issues I hope this group can help with. At first emerge was trying to
merge some really old packages. For example I wanted mysql 5.x, php 5.x
and Apache 2.x and Gentoo was only going to install mysql 4.x. I did
find out about ACCEPT_KEYWORDS=~x86 which has helped with getting more
recent versions of software installed.
In Debian/Ubuntu etc. you can choose between stable and testing. The
equivalent of testing in Gentoo is enabled by putting ACCEPT_KEYWORDS=~ARCH
in /etc/make.conf (in your case ARCH=x86). If like me you choose to run
generally stable and only choose testing (~ARCH) for some packages, then you
should never use ACCEPT_KEYWORDS when emerging something. ACCEPT_KEYWORDS may
on rare occasions be interesting with emerge --pretend to see what a testing
package requires. Instead when installing a testing package you should put it
in /etc/portage/package.keywords. If you do emerge something with
ACCEPT_KEYWORDS on the command line then portage will downgrade the package
next time you try to upgrade world (emerge --update --verbose --deep world).
Also never emerge something without first trying with --pretend or --ask to
see what it will do.
Some tasks I could do in Fedora/Ubuntu that I want to know how to do on
Gentoo:
See list of all *installed* software.
I would use (it's a capital i):
#eix --installed --compact
This provides a list of all installed packages including all dependencies.
In /var/lib/portage/world there is a list of all of all software that you
have explicitly installed. The rest of what is installed should be depencies
of packages in the world file. So to get a list of packages that I installed
I would use (this is only one line):
#while read pkg; do eix --force-color --compact ^${pkg}$ | head -n 1; done
/var/lib/portage/world
Browse available software that can be installed.
Others have mentioned kuroo which is by far the best gui that I have ever seen
for portage (not that I ever use a gui ;) ). The most recent version of kuroo
in portage i.e. kuroo 0.7* does not support portage 2.1* (~ARCH) so if you
use that you need kuroo 0.8.0_rc1. There is an ebuild available from the
project page [1] that you can easily use through a local overlay [2].
[1] http://tux.myftp.org/installation.html
[2] http://gentoo-wiki.com/HOWTO_Installing_3rd_Party_Ebuilds
See what version of a particular software package is installed.
#eix package
See if any new versions of *installed* software are available.
To do a sync I always use eix-sync rather than emerge --sync. This provides a
much better overview of what was changed during that particular sync. To see
all versions of a particular package
#eix package
To see if there are newer versions of anything in world:
#emerge --update --verbose --deep --pretend world
Important packages to install for portage are as others have mentioned eix and
gentoolkit.
HtH
--
Bo Andresen
--
gentoo-user@gentoo.org mailing list
[gentoo-user] [OT] Terminal formatting and colors escape sequences
I know this is very off topic, but I have no idea how to find a place where it actually is on topic.. so I'm posting it here. I wish to be able to run a program (eix-sync/diff-eix) in cron that prints colors (with use of --force-color) and then send that colored output as a mail. In order to get colors in a mail a have to use html. If there exist a program that is capable of converting escape sequences used for formatting and coloring an xterm to html I would love to know about it. Otherwise I'll make it myself (with a very limited range of supported escape sequences). My problem is that I am unable to locate a reference that defines the escape sequences. Guessing by testing with xterm isn't really a optimal way to find out... -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] [OT] Terminal formatting and colors escape sequences
Just in case somebody wonders what I'm talking about here is an example (^[ is an escape character): ^[[32;01m*^[[0m Running emerge --sync ... ^[[A^[[73G ^[[34;01m[ ^[[32;01mok^[[34;01m ]^[[0m ^[[32;01m*^[[0m Running update-eix ... ^[[A^[[73G ^[[34;01m[ ^[[32;01mok^[[34;01m ]^[[0m ^[[32;01m = green ^[[0m = black ^[[73G = right justify ^[[34;01m = blue ^[[A = ? -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] [OT] Terminal formatting and colors escape sequences
On Thursday 02 March 2006 15:56, Hans-Werner Hilse wrote: I'd say, the Perl module HTML::FromANSI should do what you want (available from cpan). It brings a script, ansi2html, that provides access from the command line. Note that you might have to play with the TERM environment variable. Sounds interesting. Unfortunately I don't know Perl but I think I'll have a look at it anyway... Thanks. -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] [OT] Terminal formatting and colors escape sequences
Does anyone know of I way in which to force emerge to show colors when piping the output to a file? I really could use a tip. -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] No direct rendering with ATI Radeon 7000/VE
On Wednesday 01 March 2006 11:33, Izar Ilun wrote: Well, I've already posted this herehttp://forums.gentoo.org/viewtopic-t-438630.htmlbut as I get no answers I ask you for desperated help. Consider that I want/need to use Kernel's free ATI drivers and nothing more, is it possible to get direct rendering that way? How can I do it? Any idea little idea will be very appreciated. Thanx in advance! P.D: If you need more info, ask for it please! ;-) Could you post output of: # lsmod | grep 'drm\|agp\|rad' # zcat /proc/config.gz | grep MTRR # lspci | grep -i radeon You should be able to find som info about the problem in the log in /var/log/Xorg.0.log too. Perhaps y could post what you think is relevant. Also I should mention that there have been several threads on this on this mailing list. -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] 3d rendering with dri radeon
On Saturday 25 February 2006 18:50, Bo Andresen wrote: On Saturday 25 February 2006 00:36, Benno Schulenberg wrote: But it's working now. So tell us how many frames glxgears is doing now, with and without radeon. :) Actually the results using dri is worse than the results without dri in terms of frame rates. Without dri I get something like 250 FPS. With dri I get something like 228 FPS. The great difference though is when looking at the X cpu usage. glxgears uses around 0.5 % with or without dri. But without dri X uses 92% of the cpu resources. With dri X uses between 0.5 and 1% cpu. So obviously dri is preferable... ;) Either something is seriously wrong with my settings somehow or x11-drm is really crappy. A have no idea.. Fact is yesterday I was forces to boot back in to my previous kernel since the one I created for x11-drm (i.e. with dri disabled in the kernel configuration) failed me. The previous kernel had dri compiled as modules and I have never had direct rendering working with it... until now.. I am very surprised by this but it turns out direct rendering is working now with the kernel modules. And the performance is quite a bit better than before: ~ $ glxinfo | grep dir direct rendering: Yes ~ $ glxgears 6057 frames in 5.0 seconds = 1211.400 FPS 7273 frames in 5.0 seconds = 1454.600 FPS 7268 frames in 5.0 seconds = 1453.600 FPS 7275 frames in 5.0 seconds = 1455.000 FPS 7274 frames in 5.0 seconds = 1454.800 FPS 7275 frames in 5.0 seconds = 1455.000 FPS 7268 frames in 5.0 seconds = 1453.600 FPS 7275 frames in 5.0 seconds = 1455.000 FPS 7272 frames in 5.0 seconds = 1454.400 FPS 7201 frames in 5.0 seconds = 1440.200 FPS 7259 frames in 5.0 seconds = 1451.800 FPS 7270 frames in 5.0 seconds = 1454.000 FPS 7272 frames in 5.0 seconds = 1454.400 FPS I am clueless as to why this is working now when it wasn't before... -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] 3d rendering with dri radeon
On Tuesday 28 February 2006 13:13, Bo Andresen wrote: I am very surprised by this but it turns out direct rendering is working now with the kernel modules. And the performance is quite a bit better than before: [SNIP] Well, when I conducted the tests that yielded a performance of 228 FPS with dri I forgot that I had [EMAIL PROTECTED] running in the background. This makes the whole difference in performance. By starting that service I get down to the same lousy framerates with dri from the kernel and by stopping it again the performance goes up again. So all in all the performance without [EMAIL PROTECTED] and with dri is 5-6 times better without [EMAIL PROTECTED] and without dri. I am clueless as to why this is working now when it wasn't before... This still does not explain why this is working now when it wasn't before I compiled a kernel without dri and emerged x11-drm. But hey, it works... ;) -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] 3d rendering with dri radeon
On Tuesday 28 February 2006 21:31, Benno Schulenberg wrote: Bo Andresen wrote: The previous kernel had dri compiled as modules and I have never had direct rendering working with it... until now.. I am very surprised by this but it turns out direct rendering is working now with the kernel modules. Are you certain that the radeon module being used is the one from that old kernel and not the one from x11-drm? I am very certain... but then again, I've been so before in this thread and yet been terribly wrong. Do you think there is any way to check it? Other than by compiling a new kernel with the same kernel configuration. If not that, maybe you changed something in xorg.conf? Sad to say I don't remember anymore. What versions of kernel, BTW? Kernel version has not changed - 2.6.15-r1. -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] 3d rendering with dri radeon
While the two kernels are the same version from portage they have of course different EXTRAVERSION. For the kernel I compiled for dri I simply added '-dri' to EXTRAVERSION. So when running uname -r it yielded 2.6.15-r1-dri. The previous kernel which I am running now lacks the '-dri'. On Tuesday 28 February 2006 23:16, Benno Schulenberg wrote: Bo Andresen wrote: On Tuesday 28 February 2006 21:31, Benno Schulenberg wrote: Are you certain that the radeon module being used is the one from that old kernel and not the one from x11-drm? I am very certain... but then again, I've been so before in this thread and yet been terribly wrong. Do you think there is any way to check it? Yes. Unmerge x11-drm, reboot, and see if it still works. :) Unmerging x11-drm does not remove the kernel due to CONFIG_PROTECT. --- cfgpro obj /lib/modules/2.6.15-gentoo-r1-dri/x11-drm/radeon.ko --- cfgpro obj /lib/modules/2.6.15-gentoo-r1-dri/x11-drm/r128.ko --- cfgpro obj /lib/modules/2.6.15-gentoo-r1-dri/x11-drm/mach64.ko --- cfgpro obj /lib/modules/2.6.15-gentoo-r1-dri/x11-drm/drm.ko --- cfgpro dir /lib/modules/2.6.15-gentoo-r1-dri/x11-drm I do, however, believe that this output proves that I am not running the x11-drm modules since it does contain '-dri'. Kernel version has not changed - 2.6.15-r1. Yeah, then it's the x11-drm radeon module you're using, not the one from the kernel. If it had been a diffent version (2.6.14, say), the module would have been in a different subdir in /lib/modules and wouldn't get used. I'm not really sure what you by this. Why does that mean that I'm using the x11-drm modules? -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] What happens with masked packages?
On Sunday 26 February 2006 06:16, Boyd Stephen Smith Jr. wrote: Again, hard to do automatically. Wheras, if I could just set ACCEPT_UPSTREAM=BETA I'd get all the betas. Or I could use package.upstream and but in kde-extra/kaffeine ALPHA and get anything assigned more than a snapshot number for that package. (Instead of manually checking after each sync to see if there's a new, masked version.) How exactly is is you want this to work. I mean for example gaim-2.0.0_beta2-r1 is a beta and it's very unstable (well, it crashed occasionally for me). In order to get it you need to put it in package.unmask and package.keywords. Do you want to have to put it package.upstream too? Or don't you want it to be masked even though it's very unstable? Should package.upstream override package.mask? -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] What happens with masked packages?
On Sunday 26 February 2006 21:40, Boyd Stephen Smith Jr. wrote: How exactly is is you want this to work. My proposal at this point, would be for an additional restriction on packages based on a new UPSTREAM variable in the ebuild itself, ACCEPT_UPSTREAM variable in make.conf / the environment, and the package.upstream file in /etc/portage. I read your previous posts about this as that you wanted it to be easier to get beta versions but what you want is in fact the exact opposite - further restriction. Now I get it. -- Bo Andresen -- gentoo-user@gentoo.org mailing list
[gentoo-user] Where do these use flags come from?
I decided I wanted to remove the ipv6 use flag which I have had enabled in make.conf for quite a while but never really been on a ipv6 network and don't suspect I will in the near future. When upgrading firefox I noted it has that use flag and decided I want to know what it actually does. Only, I cannot find it anywhere in the ebuilds! So where does it come from and what *exactly* does it do? ~ # emerge -uvp mozilla-firefox These are the packages that would be merged, in order: Calculating dependencies... done! [ebuild U ] www-client/mozilla-firefox-1.5.0.1-r2 [1.5.0.1-r1] USE=java mozdevelop xprint -debug -gnome -ipv6* -xinerama 33 kB Total size of downloads: 33 kB ~ # grep USE /usr/portage/www-client/mozilla-firefox/*.ebuild /usr/portage/www-client/mozilla-firefox/mozilla-firefox-1.0.7-r4.ebuild:IUSE=gnome java mozdevelop mozsvg mozcalendar /usr/portage/www-client/mozilla-firefox/mozilla-firefox-1.5-r11.ebuild:IUSE=java mozdevelop /usr/portage/www-client/mozilla-firefox/mozilla-firefox-1.5-r9.ebuild:IUSE=java mozdevelop /usr/portage/www-client/mozilla-firefox/mozilla-firefox-1.5.0.1-r2.ebuild:IUSE=java mozdevelop ~ # grep ipv6 /usr/portage/www-client/mozilla-firefox/*.ebuild ~ # ~ # equery u mozilla-firefox [ Searching for packages matching mozilla-firefox... ] [ Colour Code : set unset ] [ Legend: Left column (U) - USE flags from make.conf ] [ : Right column (I) - USE flags packages was installed with ] [ Found these USE variables for www-client/mozilla-firefox-1.5.0.1-r1 ] U I - - debug : Tells configure and the makefiles to build for debugging. Effects vary across packages, but generally it will at least add -g to CFLAGS. Remember to set FEATURES=nostrip too - - gnome : Adds GNOME support - + ipv6 : Adds support for IP version 6 + + java : Adds support for Java + + mozdevelop : Enable features for web developers (e.g. Venkman) - - xinerama : Add support for the xinerama X11 extension, which allows you to stretch your display across multiple monitors + + xprint : Support for xprint, http://www.mozilla.org/projects/xprint/ -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Where do these use flags come from?
On Monday 27 February 2006 03:47, Luis Ortiz wrote: Only, I cannot find it anywhere in the ebuilds! So where does it come from and what *exactly* does it do? That ebuild inherits the mozconfig-2.eclass Look at /usr/portage/eclass/mozconfig-2.eclass and you'll find ipv6 defined in the IUSE variable. Thanks for quick and specific response. :) -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Where do these use flags come from?
On Monday 27 February 2006 04:03, Bo Andresen wrote: BTW stop top-posting, please. :) On Monday 27 February 2006 04:13, Brett I. Holcomb wrote: Evidently I didn't understand what you were asking the first time - sorry it didn't meet your needs. I learned something, too - that the eclasses can pass their flags on. On Sunday February 26 2006 22:03, Bo Andresen wrote: [SNIP] Do you never read to the end of mails to which you reply? Stop top-posting. Don't include quotes to which you do not reply. Reply below quotes to which you do reply. It is not that hard! http://www.catb.org/jargon/html/T/top-post.html -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] 3d rendering with dri radeon
On Saturday 25 February 2006 15:03, Benno Schulenberg wrote: Bo Andresen wrote: Is there a way to disable dri for this test without restarting X? Good question. I don't know. What I do is simply move the driver in /usr/lib/modules/dri/ out of the way, or renaming it temporarily (to say NOTradeon_dri.so). That doesn't work for me. But moving /usr/lib/modules/dri/r200_dri.so does. ;) Moving the entire directory would of course work for everyone... -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] 3d rendering with dri radeon
On Saturday 25 February 2006 00:36, Benno Schulenberg wrote: But it's working now. So tell us how many frames glxgears is doing now, with and without radeon. :) Actually the results using dri is worse than the results without dri in terms of frame rates. Without dri I get something like 250 FPS. With dri I get something like 228 FPS. The great difference though is when looking at the X cpu usage. glxgears uses around 0.5 % with or without dri. But without dri X uses 92% of the cpu resources. With dri X uses between 0.5 and 1% cpu. So obviously dri is preferable... ;) Listed the output below if anyone is interested. ~$ glxinfo | grep dir direct rendering: No OpenGL renderer string: Mesa GLX Indirect ~$ glxgears 792 frames in 5.0 seconds = 158.400 FPS 1945 frames in 5.0 seconds = 389.000 FPS 1847 frames in 5.0 seconds = 369.400 FPS 1265 frames in 5.0 seconds = 253.000 FPS 1138 frames in 5.0 seconds = 227.600 FPS 1139 frames in 5.0 seconds = 227.800 FPS 1265 frames in 5.0 seconds = 253.000 FPS 1138 frames in 5.0 seconds = 227.600 FPS 1139 frames in 5.0 seconds = 227.800 FPS 1265 frames in 5.0 seconds = 253.000 FPS 1265 frames in 5.0 seconds = 253.000 FPS 1265 frames in 5.0 seconds = 253.000 FPS 1265 frames in 5.0 seconds = 253.000 FPS 1265 frames in 5.0 seconds = 253.000 FPS 1265 frames in 5.0 seconds = 253.000 FPS 1138 frames in 5.0 seconds = 227.600 FPS X usage ~92% ~$ glxinfo | grep dir direct rendering: Yes ~$ glxgears 1035 frames in 5.0 seconds = 207.000 FPS 1162 frames in 5.0 seconds = 232.400 FPS 1155 frames in 5.0 seconds = 231.000 FPS 1143 frames in 5.0 seconds = 228.600 FPS 844 frames in 5.0 seconds = 168.800 FPS 1145 frames in 5.0 seconds = 229.000 FPS 1135 frames in 5.0 seconds = 227.000 FPS 1164 frames in 5.0 seconds = 232.800 FPS 1144 frames in 5.0 seconds = 228.800 FPS 627 frames in 5.0 seconds = 125.400 FPS 1048 frames in 5.0 seconds = 209.600 FPS 1148 frames in 5.0 seconds = 229.600 FPS 1141 frames in 5.0 seconds = 228.200 FPS 1057 frames in 5.0 seconds = 211.400 FPS 1077 frames in 5.0 seconds = 215.400 FPS 1137 frames in 5.0 seconds = 227.400 FPS X usage: 0.5-1% -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] 3d rendering with dri radeon
On Sunday 26 February 2006 05:02, Bruce Burden wrote: Okay, I have decided to use the xorg DRM module. However, I am still doing something wrong: First of all did you follow the guide at http://www.gentoolinux.org/doc/en/dri-howto.xml ? Or any other guide? Assuming you used the guided above: 1) Could you post the output of: zcat /proc/config.gz | grep 'MTRR\|AGP\|DRM' 2) What version of x11-drm are you using? Did you set VIDEO_CARDS=ati in make.conf (or on the command line while emerge x11-drm)? BTW what version of xorg-x11? 3) What does the Module section of your xorg.conf look like? I have: Section Device Identifier X600 Driver radeon VendorName ATI Technologies Inc BoardName RV350 [MOBILITY RADEON X600] BusID PCI:1:0:0 EndSection for the xorg.conf entry. Should it be ati? No. # lsmod Module Size Used by radeon 98464 0 drm61592 1 radeon intel_agp 18332 1 agpgart27216 2 drm,intel_agp [SNIP] Looks all right. -- Bo Andresen
Re: [gentoo-user] 3d rendering with dri radeon
On Friday 24 February 2006 20:14, Benno Schulenberg wrote: Bo Andresen wrote: On Friday 24 February 2006 00:04, Benno Schulenberg wrote: Maybe try again, taking extra care to avoid typos in VIDEO_CARDS=ati? I did do that. And as stated before it won't compile without the VIDEO_CARDS variable in it's environment. Tried several times with both ati and radeon.. I've emerged the latest version of portage (2.1_pre4-r1), put VIDEO_CARDS=ati in /etc/make.conf, have no video_cards_... set in the USE flags, and have emerged x11-drm. It does produce a radeon module (and a drm.ko, r128.ko, and mach64.ko besides). I was very sure that I had tested this twice with VIDEO_CARDS=ati. Both times the ati flag didn't get enabled i.e. emerge -vp x11-drm showed: [ebuild R ] x11-base/x11-drm-20051223 VIDEO_CARDS=-ati -i810 -mga -nv -savage -sis -tdfx -via 0 kB This of course didn't result in the radeon driver being built. It seems, however, that my memory must be failing (can't find any other explanation) since when I remove the use flag now and set VIDEO_CARDS=ati the flag does get enabled and the radeon module built. I really don't understand this - was so sure... So obviously there is no bug other than me.. ;) What version of portage are you using that this isn't working for you? I have been using portage 2.1_pre4-r1 all along. Having portage sending mails to me is invaluable. :) -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] 3d rendering with dri radeon
On Saturday 25 February 2006 00:36, Benno Schulenberg wrote: But it's working now. So tell us how many frames glxgears is doing now, with and without radeon. :) Is there a way to disable dri for this test without restarting X? -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] chroot: cannot run command `bin/bash': Exec format error
On Thursday 23 February 2006 00:42, Boyd Stephen Smith Jr. wrote: livecd gentoo # chroot /mnt/gentoo bin/bash chroot: cannot run command `bin/bash': Exec format error Your 32-but kernel can't run the 64-bit bash. You'll have to use a 64-bit kernel (or as 32-bit stage3, and then gradually recompile) Will a 64 bit kernel be able to run a 32 bit bash? In order to get a 64 bit kernel a have to set CFLAGS=-march=k8 and set the processor type to K8 in the kernel configuration, right? -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] chroot: cannot run command `bin/bash': Exec format error
On Thursday 23 February 2006 22:40, Boyd Stephen Smith Jr. wrote: On Thursday 23 February 2006 15:31, Bo Andresen [EMAIL PROTECTED] wrote about 'Re: [gentoo-user] chroot: cannot run command `bin/bash': Exec format error': On Thursday 23 February 2006 00:42, Boyd Stephen Smith Jr. wrote: livecd gentoo # chroot /mnt/gentoo bin/bash chroot: cannot run command `bin/bash': Exec format error [SNIP] Will a 64 bit kernel be able to run a 32 bit bash? A 64-bit kernel will run 32-bit binaries fine... Um, there may be a needed kernel option though... CONFIG_IA32_EMUL? Anyone? I cannot seem to find any such kernel config option. In order to get a 64 bit kernel a have to set CFLAGS=-march=k8 Your CFLAGS in make.conf don't affect your kernel, normally. Didn't really think so either. It's just that I still get the Exec format error when I try to chroot. Is there a way to very that I really am running a 64 kernel? I don't use genkernel maybe it does some crazy magic like that. I don't either. and set the processor type to K8 in the kernel configuration, right? Just setting the proper processor type should build your kernel as 64-bit. Did do that. Thanks for your replies.. -- Bo Andresen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] chroot: cannot run command `bin/bash': Exec format error
On Thursday 23 February 2006 23:33, Boyd Stephen Smith Jr. wrote: On Thursday 23 February 2006 15:56, Bo Andresen [EMAIL PROTECTED] A 64-bit kernel will run 32-bit binaries fine... Um, there may be a needed kernel option though... CONFIG_IA32_EMUL? Anyone? I cannot seem to find any such kernel config option. I think these are relevant: $ zgrep -i ia32 /proc/config.gz CONFIG_IA32_EMULATION=y CONFIG_IA32_AOUT=y I cannot find those options. ~ # zcat /proc/config.gz | grep -i '32\|k8' CONFIG_X86_32=y CONFIG_INIT_ENV_ARG_LIMIT=32 CONFIG_MK8=y # CONFIG_SCSI_QLA2322 is not set # CONFIG_SCSI_NSP32 is not set # CONFIG_PCNET32 is not set # CONFIG_USB_SERIAL_MCT_U232 is not set # CONFIG_NLS_CODEPAGE_932 is not set CONFIG_CRYPTO_CRC32C=m CONFIG_CRC32=y CONFIG_LIBCRC32C=m I notice CONFIG_X86_32=y but cannot figure out where to change that in make menuconfig. Typing /X86_32 just gives this as a search result: Symbol: X86_32 [=y] Is there a way to verify that I really am running a 64 kernel? I believe this tells you: $ uname -m x86_64 ~ # uname -m i686 So, you may want to configure, make, and install your kernel like: make ARCH=x86_64 menuconfig make ARCH=x86_64 make ARCH=x86_64 install (You don't need a CROSS_COMPILE prefix since gcc should work fine.) Should I ignore all these warnings? # make ARCH=x86_64 menuconfig scripts/kconfig/mconf arch/x86_64/Kconfig # # using defaults found in .config # .config:90:warning: trying to assign nonexistent symbol X86_ELAN .config:91:warning: trying to assign nonexistent symbol X86_VOYAGER .config:93:warning: trying to assign nonexistent symbol X86_SUMMIT .config:94:warning: trying to assign nonexistent symbol X86_BIGSMP .config:96:warning: trying to assign nonexistent symbol X86_GENERICARCH .config:97:warning: trying to assign nonexistent symbol X86_ES7000 .config:98:warning: trying to assign nonexistent symbol M386 .config:99:warning: trying to assign nonexistent symbol M486 .config:100:warning: trying to assign nonexistent symbol M586 .config:101:warning: trying to assign nonexistent symbol M586TSC .config:102:warning: trying to assign nonexistent symbol M586MMX .config:103:warning: trying to assign nonexistent symbol M686 .config:104:warning: trying to assign nonexistent symbol MPENTIUMII .config:105:warning: trying to assign nonexistent symbol MPENTIUMIII .config:106:warning: trying to assign nonexistent symbol MPENTIUMM .config:107:warning: trying to assign nonexistent symbol MPENTIUM4 .config:108:warning: trying to assign nonexistent symbol MK6 .config:109:warning: trying to assign nonexistent symbol MK7 .config:111:warning: trying to assign nonexistent symbol MCRUSOE .config:112:warning: trying to assign nonexistent symbol MEFFICEON .config:113:warning: trying to assign nonexistent symbol MWINCHIPC6 .config:114:warning: trying to assign nonexistent symbol MWINCHIP2 .config:115:warning: trying to assign nonexistent symbol MWINCHIP3D .config:116:warning: trying to assign nonexistent symbol MGEODEGX1 .config:117:warning: trying to assign nonexistent symbol MCYRIXIII .config:118:warning: trying to assign nonexistent symbol MVIAC3_2 .config:119:warning: trying to assign nonexistent symbol X86_GENERIC .config:121:warning: trying to assign nonexistent symbol X86_XADD .config:125:warning: trying to assign nonexistent symbol X86_WP_WORKS_OK .config:126:warning: trying to assign nonexistent symbol X86_INVLPG .config:127:warning: trying to assign nonexistent symbol X86_BSWAP .config:128:warning: trying to assign nonexistent symbol X86_POPAD_OK .config:129:warning: trying to assign nonexistent symbol X86_CMPXCHG64 .config:131:warning: trying to assign nonexistent symbol X86_INTEL_USERCOPY .config:132:warning: trying to assign nonexistent symbol X86_USE_PPRO_CHECKSUM .config:140:warning: trying to assign nonexistent symbol X86_UP_APIC .config:141:warning: trying to assign nonexistent symbol X86_UP_IOAPIC .config:145:warning: trying to assign nonexistent symbol X86_MCE_NONFATAL .config:146:warning: trying to assign nonexistent symbol X86_MCE_P4THERMAL .config:147:warning: trying to assign nonexistent symbol TOSHIBA .config:148:warning: trying to assign nonexistent symbol I8K .config:149:warning: trying to assign nonexistent symbol X86_REBOOTFIXUPS .config:160:warning: trying to assign nonexistent symbol NOHIGHMEM .config:161:warning: trying to assign nonexistent symbol HIGHMEM4G .config:174:warning: trying to assign nonexistent symbol REGPARM .config:220:warning: trying to assign nonexistent symbol APM .config:231:warning: trying to assign nonexistent symbol PCI_GOBIOS .config:232:warning: trying to assign nonexistent symbol PCI_GOMMCONFIG .config:233:warning: trying to assign nonexistent symbol PCI_GODIRECT .config:234:warning: trying to assign nonexistent symbol PCI_GOANY .config:1405:warning: trying to assign nonexistent symbol X86_FIND_SMP_CONFIG .config:1406:warning: trying to assign nonexistent symbol X86_MPPARSE .config:1459:warning: trying
