/srv/irclogs.ubuntu.com/2004/12/31/#ubuntu-devel.txt

=== sivang [~sivang@80.179.82.41.forward.012.net.il] has joined #ubuntu-devel
wasabi_Kamion, ping (re: debian-installer preseed)12:13
=== AndyFitz [~andy@220-245-97-227-qld-pppoe.tpgi.com.au] has left #ubuntu-devel []
=== lexhider_ [~lexhider@dip-220-235-84-168.vic.westnet.com.au] has joined #ubuntu-devel
wasabi_Anybody aware where the debian-installer gets the name of the Release to install?12:55
wasabi_Trying to get Hoary's d-i to install Warty.12:55
wasabi_Got it building using Warty udebs, but the net-retriever part is still trying to grab hoary Release file.12:55
=== BradB [~bradb@modemcable046.50-70-69.mc.videotron.ca] has joined #ubuntu-devel
zulbluefoxicy, im getting the same01:13
=== justdave [~dave@66.227.241.236.gha.mi.chartermi.net] has joined #ubuntu-devel
=== TD [~mike@foghorn.codeweavers.com] has left #ubuntu-devel ["No]
=== bitserf [~ljb@222-152-17-146.jetstream.xtra.co.nz] has joined #ubuntu-devel
jon1012is there some geeks having geek codes here ? ;)01:21
jon1012(just to know if people continue to use geek codes lol...)01:22
=== doko [doko@dsl-082-082-067-235.arcor-ip.net] has joined #ubuntu-devel
=== lexhider_ [~lexhider@dip-220-235-84-168.vic.westnet.com.au] has joined #ubuntu-devel
=== johanbr [~j@d207-81-98-60.bchsia.telus.net] has joined #ubuntu-devel
=== srbaker [~srbaker@blk-224-161-8.eastlink.ca] has joined #ubuntu-devel
=== BradB [~bradb@modemcable046.50-70-69.mc.videotron.ca] has joined #ubuntu-devel
=== robtaylor_ [~robtaylor@nat2.camtrial.co.uk] has joined #ubuntu-devel
=== mdke [~Matt|@81-178-85-6.dsl.pipex.com] has joined #ubuntu-devel
=== lexhider [~lexhider@dip-220-235-84-168.vic.westnet.com.au] has joined #ubuntu-devel
=== johanbr [~j@d207-81-98-60.bchsia.telus.net] has left #ubuntu-devel []
wasabi_Hmm.04:04
wasabi_Wonder if I can use the Sarge installer to install ubuntu.04:05
=== mdke [~Matt|@mdke.user] has joined #ubuntu-devel
calclong double on hoary is hosed04:26
calcit states it has 18 digits of precision but only has the same as double04:26
calcat least on amd64 port04:26
calcf - 6 d - 15 l - 1804:27
calcf - 0.33333334 0.33333334326744080 0.3333333432674407959004:27
calcd - 0.33333333 0.33333333333333331 0.3333333333333333148304:27
calcl - 0.33333333 0.33333333333333331 0.3333333333333333148304:27
=== calc kicks himself
calcnm i used the wrong specifier04:28
ajmitch_aha04:28
=== calc wonders how using the wrong specifier gives you the same result as the double, wtf?
calcshouldn't that just not work at all04:28
=== calc dislikes getting weird correct looking output ;)
=== usual [~colin@alb-69-202-40-142.nycap.rr.com] has joined #ubuntu-devel
wasabi_HAHA got it.06:23
=== wasabi_ has debian-installer SVN installing Warty.
wasabi_It's only impressive because I have no idea what I'm doing.06:23
wasabi_Cool. Full preseed support.06:25
=== wasabi_ unf
fabbionemorning06:57
fabbionebob2: i can't find the bug about the ppc "go to sleep & die" patch07:20
fabbionedid you assign it to me?07:20
=== ultrafunk [~pd@eth779.vic.adsl.internode.on.net] has joined #ubuntu-devel
=== pitti [~martin@195.227.105.180] has joined #ubuntu-devel
pittiMorning guys07:51
fabbionehey pitti 07:51
pittifabbione: had a nice trip home?07:52
fabbionepitti: we need to take a look at the new CAN's07:52
fabbionepitti: no, it sucked as hell07:52
fabbionehad 3 hours delay because the landing gear broke07:52
pittiugh07:52
bob2fabbione: oh, I suck, sorry08:01
bob2fabbione@c.c? 08:01
fabbionebob2: done already08:05
=== ultrafunk [~pd@eth779.vic.adsl.internode.on.net] has joined #ubuntu-devel
=== Simira [~rpGirl@ti511220a080-0846.bb.online.no] has joined #ubuntu-devel
=== seb128 [~seb128@ANancy-151-1-49-175.w83-196.abo.wanadoo.fr] has joined #ubuntu-devel
pittiHi seb128!09:27
seb128hey !09:27
seb128pitti: how was the travel ?09:27
pittiseb128: 45 minutes delay, but otherwise fine09:27
pittiseb128: and yours?09:27
=== mvo [~Michael@ip181.135.1511I-CUD12K-01.ish.de] has joined #ubuntu-devel
seb12830min delayed, out of this fine :)09:28
fabbionedamn i was the only one with 3 hours delay?09:29
seb128yep, you win for the moment apparently :)09:31
pittisjoerd: ping09:41
sjoerdpitti: pong09:41
pittisjoerd: I assume that Debian won't get python 2.3 for sarge?09:42
pittisjoerd: because our hal now uses python 2.409:42
Treenakspitti: 2.4 you mean?09:42
pittisjoerd, Treenaks: yes, of course. 2.409:42
sjoerdi guess it stays with 2.3 as default yes09:42
=== pitti should wake up now
sjoerdsaw the rebuild against python 2.4 in the ubuntu changelog09:44
pittisjoerd: I have to merge the changes again, and the current autofoo changes are a real mess09:44
pittisjoerd: the current package contains autofoo changes in both the package diff.gz and the patch09:45
sjoerdhrm oh, not so good09:45
pittidoko: do the python2.4 hal changes require an autoreconf run or can I revert these changes?09:45
sjoerdpython 2.4 is default in ubuntu now ?09:48
Mithrandiryes09:49
sjoerdah09:49
cenerentolahi ya09:49
=== enrico [~enrico@enrico.developer.debian] has joined #ubuntu-devel
cenerentolaenrico: buon giorno09:49
enricoCiao!09:49
Mithrandirhi enrico 09:49
sjoerdpitti: doko should know what's planned for python in debian.. but my guess would be python 2.4 default wil be postsarge09:52
sjoerdanyway time for a shower and getting myself to the uni :)09:53
Mithrandirswitching to p2.4 in debian now would be madness and cause for doko to be killed by Kamion next time they meet. ;)09:53
pittiright09:53
seb128Mithrandir: any idea on #4785 ?09:54
TreenaksMithrandir: will there be pictures?09:54
MithrandirTreenaks: video! :P09:55
=== mvo [~Michael@ip181.135.1511I-CUD12K-01.ish.de] has joined #ubuntu-devel
TreenaksMithrandir: 8)09:56
Mithrandirseb128: not really, my first thought was "wrong compiler version", but it doesn't look like that.09:57
seb128Mithrandir: ok, I'll ping Scott :)09:59
seb128thanks anyway09:59
MithrandirI wonder how libtool can fuck that up, I can try to reproduce09:59
seb128I've no problem on my i386 box, so if anyone with an amd64 install can try this that would be nice10:00
Mithrandirreproduced here10:01
seb128ok10:02
seb128thanks10:02
Mithrandiryup, removing rpath fixes it10:03
pittiHi ogra10:06
ogramorning everybody10:06
pittidoko: ping10:06
Mithrandirmozilla-browser ships with /usr/lib/libxpcom.so with an unversioned SONAME.  They should so be shot.10:07
=== ogra just crawls out of his programmers cave.....
ograi ripped apart the lock.c code from xscreensaver last week.........glued it together again......10:08
TreenaksMithrandir: who's stopping you? :)10:08
ograthese are the leftovers ;)10:08
ograhttp://www.grawert.net/xss_shot_1.png10:08
pittilamont: still here?10:09
fabbionehey ogra10:11
ograhey fabio.....back in the cold again ?10:12
fabbioneyeah10:12
ograbrrr10:12
fabbioneit sucks10:12
seb128ok, anyone has an idea of why  multisync_0.82-3.2ubuntu2 doesn't try to build ? (no build log for it)10:14
=== pitti [~martin@195.227.105.180] has joined #ubuntu-devel
=== mvo [~Michael@ip181.135.1511I-CUD12K-01.ish.de] has joined #ubuntu-devel
ograseb128: ask michael banck ? i think he packaged it.....10:15
seb128ogra: I've spoken with him yesterday10:16
ograhmm10:16
seb128but the package is uploaded, his part is done10:16
seb128ogra: I know who to ask and what I'm doing, don't bother, but thanks anyway :)10:16
ogra:)10:17
fabbioneperhaps is in new?10:19
seb128fabbione: according to azeem should not10:19
seb128and the source is in the archive10:19
fabbioneis it in universe?10:22
seb128yep10:22
fabbioneweird10:23
fabbioneoh probably the override 10:23
fabbioneif it is unknown it gets scheduled as last10:24
fabbionewhen the buildds are really really idling10:24
seb128oh ok10:24
=== seb128_ [~seb128@ANancy-151-1-19-189.w83-194.abo.wanadoo.fr] has joined #ubuntu-devel
=== cenerentola [~cenerento@84.222.38.88] has joined #ubuntu-devel
seb128_grrrr10:28
seb128_dsl hangup ...10:28
seb128_<fabbione> when the buildds are really really idling10:28
seb128_<seb128> oh ok10:28
seb128_--- Disconnected ().10:28
fabbionei didn't write anything different10:28
fabbioneor more10:28
seb128_ok10:29
seb128_I'll wait for lamont and ask him to kick the build :)10:30
fabbionedoko: ping10:35
fabbionedebstriptranslations <- phear10:39
pittifabbione: do you have any better name?10:48
fabbioneno no.. sounds cool10:48
pittifabbione: to soothe the "phear" a bit, it is not activated by default :-)10:49
=== carlos [~carlos@69.Red-80-33-181.pooles.rima-tde.net] has joined #ubuntu-devel
=== ogra [~ogra@p508EA26B.dip.t-dialin.net] has joined #ubuntu-devel
=== Simira [~rpGirl@ti511220a080-0846.bb.online.no] has joined #ubuntu-devel
=== Astharot [~isager@82.52.100.18] has joined #ubuntu-devel
Simiraseb128: do you fix gdm bugs?11:38
=== rburton [~ross@82-133-111-179.dyn.gotadsl.co.uk] has joined #ubuntu-devel
TreenaksSimira: no, he creates them :P11:38
seb128Simira: depending. Most of the time I just forward them upstream11:39
SimiraTreenaks: that as well, I guess :)11:39
fabbionetime to kill dpatch in the kernel package11:40
=== martink [~martin@pD9EB28D6.dip0.t-ipconnect.de] has joined #ubuntu-devel
Simiraseb128: gnome panel crashes when I use ctrl + x for cutting text in web forms. I also had a problem with Gnome-panel running two times at logon.11:40
SimiraI didn't bother to report a bug, because there were a lot of Gnome crashes-bugs...11:41
seb128Simira: the second bug happens sometime11:41
seb128Simira: for the first one, which browser are you using ? Does it happen all the time ?11:41
Simiraseb128: sometime? Always, in my case. :(11:41
seb128killall gnome-panel11:41
seb128and save the session next time you log off11:42
seb128the session has probably something wrong11:42
Simiraseb128: various mozilla browsers (thunderbird and just mozilla, that is). Everytime I use keybord shortcuts for cutting text11:42
seb128on all the pages ?11:42
Simirayes, I fixed the session, but I get the same error when gnome panel crashes on me :p11:43
seb128ie: google.com for example ?11:43
Simiraseb128: yup11:43
Simiraand then I get "Error: detected a panel already running and will now exit", when it restarts11:43
Simira*into session manager and kill some gnome panels*11:44
Simiraoops11:44
SimiraI think I killed gnome-panel totally... now it completely disappeared11:45
seb128it should respawn11:45
Simiraah, yes. I just had to click away the error once more11:45
Simirabut now, I use mozilla firefox (not thunderbird, of course), and it crashes whenever I try to cut text by ctrl-x11:46
seb128ok11:46
seb128could you get a backtrace ?11:47
Simirauh... of what?11:47
seb128of the crash when you ctrl-X11:47
seb128you don't get a bug-buddy window ?11:48
Simiranope11:48
seb128the panel crash, right ?11:48
SimiraGnome-panel just crashes, restarts, and then I get the "panel already running"-error11:48
seb128hum11:48
seb128hoary ?11:48
Simirakind of annoying11:48
Simirayes, last update11:49
seb128ok11:49
seb128dpkg -l bug-buddy ?11:49
Simiranot completely installed due to some config-failure, it says. I'm running the last update now, so I'll try again as soon as it's done11:50
seb128apt-get -f install11:51
seb128ok11:51
seb128install gnome-panel-dbg too11:51
seb128so you can get a full backtrace11:51
Simiraok11:51
=== moyogo [~moyogo@Toronto-HSE-ppp3718050.sympatico.ca] has joined #ubuntu-devel
Simiraseb128: I got a report bug-option now, so I put it there.12:08
seb128you have gnome-panel-dbg installed ?12:09
Simirayes12:09
seb128ok, cool12:09
seb128thanks12:10
Simiranp12:10
=== Gorth [~gorth@cpe.atm2-0-51110.0x50a4d38e.abnxx10.customer.tele.dk] has joined #ubuntu-devel
=== bitserf [~ljb@222-152-17-146.jetstream.xtra.co.nz] has joined #ubuntu-devel
dokopitti: pong12:48
dokofabbione: pong12:48
fabbionedoko: dbus is FTBFS with the recent python2.4 changes12:49
Kamionwasabi_: pong, but you seem to have figured stuff out for yourself12:50
=== sid77 [~sid77@151.11.187.66] has joined #ubuntu-devel
Kamionwasabi_: FYI mirror/suite is the debconf variable used to decide from which suite to retrieve udebs, generally12:50
=== mojo_ [~mojo@220-244-212-78-vic.tpgi.com.au] has joined #ubuntu-devel
pittidoko: did you already see http://people.no-name-yet.com/~lamont/buildLogs/d/dbus/0.22+cvs.200412122010-0ubuntu3/12:55
pittidoko: hal FTBFS because of missing python2.4-dbus build-dep12:56
pittidoko: and p2.4-dbus fails12:56
=== sid77 [~sid77@151.11.187.66] has joined #ubuntu-devel
=== sivang [~sivang@80.179.82.41.forward.012.net.il] has joined #ubuntu-devel
=== sid77 [~sid77@151.11.187.66] has joined #ubuntu-devel
sid77hi01:11
dokopitti, fabbione: looking. hal failing should be clear, dbus builds fine for me. investigating ...01:12
fabbionedoko: universe enabled?01:13
=== mvo [~egon@ip181.135.1511I-CUD12K-01.ish.de] has joined #ubuntu-devel
dokofabbione: yes01:13
fabbionedoko: that might be :-)01:14
fabbionepython2.4-pyrex isn't in main01:14
fabbionehal is dep-wait dbus01:14
dokook, yes, then python2.4-pyrex needs to move to main ... 01:16
fabbionedoko: did you use perl to fix dbus too?01:17
dokoelmo: ping? moving python2.4-pyrex to main?01:17
dokofabbione: ???01:17
fabbionedoko: nevermind :-)01:17
=== sid77 [~sid77@151.11.187.66] has joined #ubuntu-devel
pittidaniels: here?01:56
=== lupus_ [~lupus@kn-ivl-2.kuleuven.net] has joined #ubuntu-devel
lupus_if I click on an executable file the application is not run in hoary02:04
lupus_for which package should I open a bugreport?02:04
Treenaksdoes anyone know kernel/isapnp stuff?02:07
pittiKeybuk: around?02:16
pittilupus_: try nautilus02:18
Treenakspitti: how afraid are you of kernel internals? :)02:19
pittiTreenaks: ?02:19
Treenakspitti: or, in other words: can you help me add an entry for a PNP card to a driver? :)02:19
Treenakshttps://bugzilla.ubuntu.com/show_bug.cgi?id=478702:19
pittiTreenaks: no idea, I cannot test this02:20
Treenakspitti: ok.. I'll try some things tonight then02:20
Treenaks*hopes for the best*02:20
pittiTreenaks: the source code looks easy enouhg02:21
Treenakspitti: yeah, but there are a few things I don't understand :) the "name" in the first part seems like it's from the hardware, but I don't know how to get it :)02:22
Treenakspitti: I'll look around a bit.. thanks anyway02:22
pittihaggai: ping02:23
=== mvo [~egon@ip181.135.1511I-CUD12K-01.ish.de] has joined #ubuntu-devel
pittimvo: still an unstable net link?02:25
mvopitti: yes :/02:25
mvofast but pretty unstable02:25
=== Astharot [~isager@82.52.100.18] has joined #ubuntu-devel
=== trulux [~lorenzo@67.Red-80-25-56.pooles.rima-tde.net] has joined #ubuntu-devel
pittiHi trulux02:34
pittitrulux: I packaged libssp02:36
pittitrulux: however, I still have some questions about it02:37
truluxpitti, feel free to ask02:39
truluxthanks btw02:39
trulux;)02:39
pittitrulux: will this lib be linked to every SSP-compiled executable?02:39
truluxyes02:39
pittifine02:39
pittisecond, I need a real upstream URL02:39
pittiand a proper upstream tarball would be nice02:40
truluxpitti, you mean libssp one?02:40
pittiyes02:40
pittisomething to put into the copyright file02:40
truluxthat's me and andrew dobbie02:40
pittitrulux: the tarball you sent me already contained an useless debian/ directory02:40
truluxi wrote the Makefile02:40
truluxpitti, yeah02:40
pittitrulux: no, I mean an URL02:40
pittitrulux: http://foo.com/libssp/ or so02:41
truluxok, http://wiki.debian-hardened.org/libssp02:41
truluxthere's nothing there but i will write the info there02:41
pittitrulux: is the tarball downloadable there?02:41
truluxadobbie and me worked out on the libssp, he managed the first pre, then i fixed the code and added a good Makefile02:41
truluxpitti, http://lorenzo.debian-hardened.org/debhard/libssp-0.1.tar.fgz02:42
trulux.gz02:42
truluxthat one is old (just a bit)02:42
truluxlet me upload it02:42
pittiwait a bit02:43
pittican you please clean up the tarball a bit?02:43
pittii. e. remove debian/02:43
pittiand remove the empty files (README and INSTALL or so)02:44
truluxsure02:44
truluxi'm doing it right now02:44
pittiright now I repackaged the orig.tar.gz02:44
pittibut that isn't so nice02:44
pittiI'd prefer having a _really_ original tarball :-)02:44
truluxpitti, kay, let me do it now02:45
truluxa few minutes and done02:45
=== dasdas [dasdas@tkp-ip-nas-1-p24.telkom-ipnet.co.za] has joined #ubuntu-devel
dasdashi02:48
Treenakshallo02:48
dasdashow do you / can you execute windows .exe files on ubuntu?02:48
HcEdasdas: use wine?02:49
Treenaksdasdas: you can ask support-questions on #ubuntu, but try the program "wine"02:49
dasdasok02:49
dasdaswhere can i download wine?02:49
Treenaksdasdas: use the package manager02:49
Treenaksdasdas: from the menu02:49
dasdasthanx02:50
truluxpitti, done02:51
truluxpitti, let me upload it02:51
truluxit will take really less time ;)02:51
pittitrulux: no problem, take your time02:51
pittitrulux: can you please remove INSTALL and README or write something useful into it?02:53
pittitrulux: oh, and can you please add a real copyright to ssp.c?02:54
pittitrulux: like "(C) 2004 Lorenzo ...."02:54
truluxyes02:54
truluxi did02:54
truluxalso a license related note02:54
truluxas specified on the GPL v202:54
pittitrulux: well, ssp.c already says it's GPL, that's not the problem02:55
pittitrulux: it's just lacking a copyright statement02:55
truluxit has no copyright info, yeah02:55
truluxnow it has02:55
truluxpitti, https://sourceforge.net/project/showfiles.php?group_id=118309&package_id=13875702:57
truluxi know SF.net is crap but no other place to host our scm02:58
truluxwe were negotiating the sponsoring of the project by Tek Alchemy, current Gentoo sponsor, but still waiting for the server02:58
pittitrulux: cool, looks good now03:00
pittitrulux: I will use sf as the upstream URL for now03:00
pittitrulux: thanks03:01
truluxok03:02
fabbioneTreenaks: that's relative simple to do (4787)03:04
pittitrulux: do you want to have it in Debian proper, too?03:04
fabbioneTreenaks: check what other pnp modules do03:04
pittitrulux: I can download to Debian experimental or only to Hoary03:04
fabbionethere is something like MODULE_<something>_TABLE 03:04
cenerentolais there any kind of base-package policy for ubuntu?03:06
truluxpiti, yeah03:06
truluxDebian experimental would be great03:06
cenerentolai mean what kind of requirements should a distribution satisfy to be called ubuntu:03:06
truluxHoary & Sid, then i would try to make it for Sarge03:06
pittitrulux: I will not upload it to sid for now03:07
cenerentolathese days im working on X-Ubuntu [hehe, ubuntu for xbox] 03:07
truluxok, no problem03:07
pittitrulux: experimental is fine; Hoary can sync from experimental and it can pass the NEW queue03:07
cenerentolaand ive ended up running [something like]  ubuntu from an usb pen...03:07
truluxpitti, kay03:07
=== sivang [~sivang@80.179.82.41.forward.012.net.il] has joined #ubuntu-devel
pittiHi sivang!03:07
cenerentolawhat kind of requirements the thing on the pen should satisfy to be called ubuntu?03:08
cenerentolaciao sivang!03:08
sivanghey cenerentola , pitti03:09
truluxor something is wrong with the gcc packages or simply my specs file is broken03:09
pittitrulux: wanna take a look at the package before I upload?03:10
=== AndyFitz [~andy@220-245-97-227-qld-pppoe.tpgi.com.au] has joined #ubuntu-devel
cenerentolano-one can help me?03:11
truluxpitti, sure03:12
truluxthe name is not changed also03:12
truluxi need to have a look inside it, this is not good if something went wrong03:12
truluxgive me half an hour03:12
pittitrulux: http://www.piware.de/libssp/03:13
pittitrulux: sure, take the time you need03:13
truluxok, great, thanks03:13
pittitrulux: I'm off for a while03:14
pittitrulux: returning this evening03:14
=== chuck_ [~chuck@CPE0006258ec6c2-CM000a73655d0e.cpe.net.cable.rogers.com] has joined #ubuntu-devel
=== zul [~chuck@zul.developer.gentoo] has joined #ubuntu-devel
seb128doko: scons need to be updated for python2.4, are you working on it ?03:28
danielsfabbione: pfft, 3 hours delay is nothing, that's only 1/8th flight time03:32
fabbionedaniels: it's nothing when you can sit in a normal chair inside the airport03:32
dokook, can do that.03:32
fabbionenot when you are on the plane and can't move03:33
danielsfabbione: ah, suck03:33
fabbionedaniels: the landing gear broke down03:33
fabbioneand they noticed one minute before taking off03:33
danielsjesus03:33
fabbionewe parked on the runway for 2 hours03:33
fabbioneone hour to move the airplane back to the parking area and do the tests03:33
rburtonlucky they didn't notice a minute later :)03:34
fabbioneand only after that they allowed us to take off03:34
fabbionerburton: well.. when i was in flight school we used to say 2 things03:34
sivangfabbione, you're a pilot? 03:34
fabbioneone is that once the airplane takes off... in one way or another you will land03:35
TreenaksI witnessed my plane getting a pit-stop/wheel change before I left _to_ Matar03:35
fabbionetwo: "go and explode"03:35
fabbionesivang: i was..03:35
sivangfabbione, nice :) btw, what flight company was it?03:35
fabbionesivang: Mrsk Air03:35
_rene_doko: you don't want to help fixing pyuno from OOo to work with something more than python 2.2? ;) (and help with the packaging ;) )03:40
=== AndyFitz [~andy@220-245-97-227-qld-pppoe.tpgi.com.au] has left #ubuntu-devel []
seb128doko: thanks for scons03:45
truluxdone03:47
truluxgcc pkgs ready03:47
truluxuploaded also03:47
sivangwow pyuno is cool, since when it's existed?03:50
wasabi_Kamion, yeah, I got it working. Pretty slick. I submitted a patch to choose-mirror.c to not need PREFERRED_DISTRIBUTION.03:52
KamionI think there's a reason why that's there03:53
wasabi_Yeah. To test the mirror.03:53
Kamionmore likely netboot03:53
wasabi_It now uses preseeding.03:53
=== sivang wonders if this is the common interface to use py scripts to connect and script OOo, gnumeric, sodipodi etc..
wasabi_OR preferred distribution, whicheher comes first.03:53
wasabi_So, to override "sarge", you just seed it.03:53
Kamionok; did you send that upstream or to our bugzilla?03:53
Kamionit should always be given a correct default anyway03:54
wasabi_debian-boot@03:54
Kamionyou know how to submit Debian bugs?03:54
lupus_seb128, https://bugzilla.ubuntu.com/show_bug.cgi?id=4828 and https://bugzilla.ubuntu.com/show_bug.cgi?id=4829 are maybe related?03:54
wasabi_yes. ;)03:54
Kamionmail to the mailing list alone will probably get lost ...03:54
wasabi_just wasn't sure if a patch should go there.03:54
wasabi_oye.03:54
Kamionyeah, patches generally in the BTS03:54
wasabi_ANyways, hoary could use that, just by providing a default preseed file...03:55
wasabi_instead of altering the code.03:55
Kamionaltering the code's fine, we already alter it all over the place anyway :)03:55
Kamionwe will be using default preseed files in hoary for other reasons anyway, though03:55
wasabi_Well, the preseeding stuff is super cool.03:55
Kamionmy main concern with default preseed files is how to get them in the right place automatically03:55
wasabi_I *almost* got Warty loading without any prompts. 03:55
wasabi_Trying to find the names of a few dialogs that won't go away.03:56
Kamionwarty didn't even have most of the preseed support code03:56
wasabi_I know, that's why I did this. =/03:56
Kamionthat should by and large be fixed in hoary, although I haven't had time to test it yet ...03:56
wasabi_need to get ubuntu deployed at work, unattended.03:56
wasabi_using FAI now03:56
_rene_wasabi_: python interface to OOo03:57
wasabi_?03:57
_rene_erm03:57
_rene_ENICK03:57
wasabi_heh.03:58
_rene_sivang: python interface to OOo03:58
=== AndyFitz [~andy@220-245-97-227-qld-pppoe.tpgi.com.au] has joined #ubuntu-devel
=== herzi [~herzi@c182041.adsl.hansenet.de] has joined #ubuntu-devel
seb128lupus_: dunno, I don't understand 482804:03
seb128lupus_: "map" ??04:03
lupus_euhm04:07
lupus_directory in nautilius04:07
seb128that's called a map ?04:07
lupus_in winhoos you speak of maps04:07
seb128or you are hacking nautilus with some weird patches ?04:07
lupus_not directories04:07
lupus_no04:07
lupus_if you pick system 04:08
lupus_from places04:08
lupus_in the gnome-panel04:08
seb128ok04:08
lupus_I mean04:08
lupus_Computer04:08
seb128is gnome-settings-daemon running ?04:08
seb128yep, I understand04:08
seb128"map" is just weird here04:08
lupus_what name should I use instead of map ? 04:09
lupus_7637 ?        00:00:01 gnome-settings-04:09
seb128in nautilus04:09
seb128or place04:09
lupus_seems to be running04:09
seb128or view04:09
seb128or dir04:09
lupus_k04:09
seb128ps ax | grep settings ?04:10
lupus_ 7637 ?        S      0:01 /usr/lib/control-center/gnome-settings-daemon --oaf-activate-iid=OAFIID:GNOME_SettingsDaemon --oaf-ior-fd=2204:10
seb128seems to be ok04:11
=== AndyFitz [~andy@220-245-97-227-qld-pppoe.tpgi.com.au] has left #ubuntu-devel []
seb128lupus_: if you killall nautilus, do you still have the problem ?04:11
lupus_I had the problem yesterday04:11
lupus_en turned pc of04:11
lupus_and today still seem problem04:11
seb128weird04:12
seb128dpkg -l capplets ?04:12
lupus_ii  capplets       2.8.1-3ubuntu6 configuration applets for GNOME 2 - binaries04:12
seb128dpkg -l nautilus ?04:13
lupus_ii  nautilus       2.9.1-0ubuntu4 file manager and graphical shell for GNOME04:13
seb128have you changed/updated something to get the problem ?04:13
lupus_I think it has something to do with the mime-share package04:14
lupus_don't think so04:14
lupus_hmm04:16
lupus_all mime stuff is not working04:16
seb128dpkg -l shared-mime-info ?04:17
lupus_ii  shared-mime-in 0.15-1ubuntu1  FreeDesktop.org shared MIME database and spe04:17
seb128running update-mime-database /usr/share/mime/ help ?04:17
=== pitti [~martin@195.227.105.180] has joined #ubuntu-devel
lupus_fixed it04:18
=== jon1012 [~jon@AMontsouris-151-1-39-185.w82-123.abo.wanadoo.fr] has joined #ubuntu-devel
lupus_both bugreports seem to be fixed04:18
lupus_you close them?04:19
seb128ok04:19
lupus_bleh04:19
seb128something has overwritten your mime database04:19
lupus_I wonder why it got broken04:19
seb128but dunno which package04:19
seb128usually that's a package which include an empty mimedatabase and overwrite the system one04:19
lupus_fun :)04:20
jon1012hello :)04:20
jon1012do you know when the mono packages in hoary will be fixed ?04:20
TreenaksSoon04:21
jon1012thanks :)04:21
seb128lupus_: dpkg -S /usr/share/mime/globs ?04:21
lupus_lupus@lupus ~/gnome-schedule $ dpkg -S /usr/share/mime/globs04:22
lupus_drivel: /usr/share/mime/globs04:22
seb128that's the broken one :)04:22
jon1012i know that it's not he right place to ask, but I'll try anyway :p04:22
jon1012when I change a few times the image of a gtkimage, my program core dumps...04:22
jon1012how can I avoid that ? :/04:22
lupus_can I see the date drivel was last updated somehow?04:22
seb128zless /usr/share/doc/drivel/changelog.Debian.gz04:23
lupus_with ls -l  /usr/share/doc/drivel/changelog.Debian.gz I see when it was installed04:25
lupus_and it was yesterday :)04:25
lupus_can you open a bugreport on debian for that package or tell me what to say to the maintainer of the package04:26
lupus_I don't really know well how to explain what he does wrong04:26
seb128http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=28648204:27
seb128already done04:27
seb128(not by me)04:27
lupus_k thx04:28
lupus_but for next time what does the globs file show?04:28
seb128?04:29
seb128that's the mime database04:29
seb128it looks in the files in /usr/share/mime/ to know the mime to use for an extension04:29
seb128the file is overwriten here04:30
seb128so no association04:30
seb128and nautilus doesn't know which one to use04:30
=== trulux [~lorenzo@67.Red-80-25-56.pooles.rima-tde.net] has joined #ubuntu-devel
pittiHi trulux04:32
pittitrulux: is the package okay?04:32
truluxsure04:35
truluxit's not okay, it's bright ;)04:36
truluxgcc-3.4 done04:36
truluxwrapper in progress04:36
pittitrulux: I'm not sure that we should modify the gcc package to include the wrapper04:36
pittitrulux: I would prefer having a gcc-ssp-3.4 package providing a gcc-ssp executable04:37
pittitrulux: at least for the time we are still testing that in universe04:37
pittitrulux: okay, then I upload libssp now04:38
truluxgreat04:41
jon1012is there someone working with gnomeiconview here ? :/04:46
jon1012just to ask a little question :)04:46
=== stratus [~stratus@200.198.184.97] has joined #ubuntu-devel
=== wasabi [~jhaltom2@64.25.11.66] has joined #ubuntu-devel
=== mvo [~egon@ip181.135.1511I-CUD12K-01.ish.de] has joined #ubuntu-devel
=== sivang [~sivang@cablep-179-87-196.cablep.bezeqint.net] has joined #ubuntu-devel
=== mvo__ [~egon@suprimo-201.ping.de] has joined #ubuntu-devel
=== enrico [~enrico@enrico.developer.debian] has joined #ubuntu-devel
=== lupus__ [~lupus@kn-ivl-2.kuleuven.net] has joined #ubuntu-devel
mojo_ah06:00
mojo_gnome-panel-screenshot crash when click Save06:00
mojo_it seems to me it can't find the Save as Path06:01
=== _rene__ [~rene@dsl-082-083-171-197.arcor-ip.net] has joined #ubuntu-devel
=== mvo__ [~egon@ip181.135.1511I-CUD12K-01.ish.de] has joined #ubuntu-devel
=== Treenaks wonders what broke his panel
Treenaksor WHO06:23
=== _rene__ is now known as _rene_
Treenaksdid anyone else get a broken panel?06:32
=== GotD0t [~GotD0t@24.48.147.43] has joined #ubuntu-devel
=== trukulo [~trukulo@62.57.66.120] has joined #ubuntu-devel
TreenaksOK, so removing ~/.gnome2/session fixes the "panel displays as a gray bar" problem06:39
Treenaksbut why did ~/.gnome2/session break it in the first place?06:40
danielsTreenaks: blame seb06:40
Treenaksoh I will06:40
=== boglot [~logbot@gw.workaround.org] has joined #ubuntu-devel
=== pitti [~martin@dialin-212-144-018-001.arcor-ip.net] has joined #ubuntu-devel
=== pitti finally has internet access again. Gosh
pittiHey dudes, today one of the most famous German computer magazines reported about Ubuntu! :-)06:50
smurfixpitti: which one? c't ?06:50
jdubpitti: positive?06:53
jdubpitti: scan! url! ;-)06:53
pittismurfix: yes, C't06:53
dokovery short notice, about 1/3 page06:53
pittijdub: unfortunately it is only 1/3 of a page, and they forgot about all the good features06:54
jdubheh06:54
jdubpitti: like pmount? :)06:54
pittihowever, at least there is something06:54
pittijdub: security, utopia, nice default configuration, TRLS etc.06:54
jdubcool06:54
smurfixpitti: c't isn't one of the most famous computer magazines, it's the only one that still deserves that name06:54
pittismurfix: agreed :-)06:55
dokosmurfix: any news about your backpack?06:56
smurfixBy the way, I got my left middle finger back today. Mostly.06:56
=== ironwolf [~ironwolf@c-24-6-169-124.client.comcast.net] has joined #ubuntu-devel
pitticool!06:56
pittismurfix: did you also get back your backpack?06:56
smurfixNope :-/  I know somebody in Barcelona who said he'd ask again today, the lost+found is closed on weekends.06:57
dokojdub: please could you have a look at chinstrap:~doko/python2.4-doc, if that's what you want about gnome devhelp integration?06:57
lamontpitti: you were looking for me earlier?06:58
jdubdoko: oh! (downloading)06:59
pittilamont: yeah, I needed a native speaker to review the xine USN. However, Kamion showed up, so he took the job :-)06:59
lamontpitti: ah, ok06:59
truluxhey pitti07:00
pittiHi trulux 07:00
truluxpitti, howya?07:00
pittitrulux: sorry, my WLAN is down07:00
truluxdid you uploaded the pkgs?07:00
trulux:(07:00
pittitrulux: I borrowed a modem from my neighbour and now work over ssh :-/07:00
pittitrulux: yes, I uploaded it to experimental. However, it's in NEW now07:01
truluxok07:02
jdubdoko: dude07:03
jdubdoko: sweet!07:04
dokohowever that's the printed documentation, not extracted from the docstrings in the sources. For that kind of thing we would have to add a pydoc -> devhelp mapping.07:05
=== lamont heads to town for lunch and bandwidth
jdubhttp://cairographics.org/pipermail/cairo/2004-December/002342.html07:10
danielsjdub: sweet!07:12
danielsjdub: apparently alexl has been making it mad fast as well07:12
jdub"dumb optimisation" :)07:12
jdubit will be interesting to see how gtk+ changes, with cairo07:13
daniels'These two changes brought down the page rendering speed to about 2.407:13
danielsseconds, which is about 11250% faster (likely even more, since a07:13
danielssignificant part of the time is spent reading and decoding the pdf and07:13
danielswriting out the resulting png).'07:13
pittisjoerd: here?07:15
truluxpitti, what do you want to do now with the pkgs?07:15
pittitrulux: sorry, I did not yet find time to look at the gcc packages07:15
pittitrulux: too much to do today07:16
truluxpitti, ok, np , where's the libssp exactly? any upload logs available like in Debian?07:16
pittitrulux: libssp is currently in the NEW queue, i. e. it must be manually processed by a Debian FTP master07:16
pittitrulux: if this is done, further uploads go directly into the archive07:17
truluxpitti, i'm confused, where i can see that? did you uploaded it to a Debian queue or to an Ubuntu one?07:19
=== trulux is confused
pittitrulux: I uploaded it to Debian experimental07:19
pittitrulux: you cannot see the package right now07:19
lifelesspitti: thank you very much, pmount is happy for me now.07:20
sjoerdpitti: yeah07:21
pittilifeless: I'm happy that you are happy now :-)07:21
pittisjoerd: with the newest hal, labels with spaces still don't work07:22
pittisjoerd: I added a trivial gvm patch which makes them finally work07:22
sjoerdpitti: you need both hal and gvm.. works here07:22
=== boglot [~logbot@gw.workaround.org] has joined #ubuntu-devel
pittioh, ok. I did not notice a new gvm07:22
sjoerd1.1.2-5 :)07:22
truluxpitti, ok, sorry of my stpudity ;D07:24
truluxpitti, i'm reading the debian mentors site07:26
truluxpitti, how somebody can apply for it?07:26
pittitrulux: no worries07:26
truluxi know i need an already working debian dev07:26
truluxbut just nothing known about somebody who wants to bet mine ;)07:27
pittitrulux: this is more or less DD-internal knowledge07:27
pittidarn, working with a modem is a PITA07:28
truluxheh07:30
truluxso, pitti, there's no place where see what packages are in NEW state or even the uploads logs07:30
truluxnot linked from mentors site indeed07:30
=== maskie [~maskie@196-30-111-147.uudial.uunet.co.za] has joined #ubuntu-devel
pittitrulux: upload logs are in the pts07:31
pittitrulux: but not for NEW07:31
sjoerdpitti: lemme know if the new package doesn't work for you 07:32
pittisjoerd: should, I just not yet upgraded to it :-(07:33
pittisjoerd: I worked offline for two hours and finally found a modem link now07:33
sjoerdpitti: ubuntu is so outdated ;P07:33
pittisjoerd: my WLAN will be unavailable until tomorrow07:33
pittisjoerd: no, it is already in ubuntu :-P07:33
sjoerdpitti: still your usb wlan card that's broken ?07:34
pittisjoerd: but not yet on _my_ ubuntu :-)07:34
sjoerdhehe07:34
=== boglot [~logbot@gw.workaround.org] has joined #ubuntu-devel
pittisjoerd: no, my provider has to make some repairs on the central router node07:34
sjoerdah07:34
pittisjoerd: at home I don't use my USB wlan stick, I'm in an intranet07:34
pittisjoerd: btw, when you write Sjoerd's Totally Rad New GVM Fork (tm), can you please make it react on, say, SIGUSR1?07:36
pittisjoerd: which causes it to reconnect to dbus?07:37
pittisjoerd: this should smoothen upgrades07:37
sjoerdpitti: who is gonna send it SIGUSR1 ?07:37
pittisjoerd: postinst?07:37
sjoerdit already reconnects to dbus nicely07:37
pittisjoerd: darn, I should sleep07:37
sjoerddbus postinst ? hal postinst ? random dbus using program postinst ?07:38
pittisjoerd: I mean on upgrades of the g-v-m package07:38
pittisjoerd: but there sigusr won't help, sorry07:38
pittisjoerd: just forget about that07:38
sjoerdhehe07:38
=== pitti got up at 0630, which isn't to be considered healthy
=== bluefoxicy [~bluefox@pcp485126pcs.whtmrs01.md.comcast.net] has joined #ubuntu-devel
=== boglot [~logbot@gw.workaround.org] has joined #ubuntu-devel
sjoerdpitti: i'm currently convincing the kernel to put enough information in sysfs for hal to nicely handle dm devices07:42
pittisjoerd: cool; what does it already provide right now?07:43
sjoerdpitti: basically nothing. It needs to provice information about it parent devices.07:44
sjoerdunfortunately hal currently can't handle device with multiple parents.07:45
sjoerdnot a problem for crypt devices luckily07:45
pittisjoerd: ah, multiple parents is for RAID devices?07:45
sjoerdyep07:45
sjoerdand multipath and other kind of cool stuff07:45
sjoerdpitti: biggest problem with the gvm rewrite is a nice name ;)07:53
pittisjoerd: ghotplug?07:53
Treenaksgotplug?07:54
pittigrdm?07:54
pitti(removable device manager), since it also manages Cameras etc.07:54
pittinight, dudes07:56
pittiI'm tired07:56
=== dasdas [dasdas@tkp-ip-nas-1-p87.telkom-ipnet.co.za] has joined #ubuntu-devel
dasdashi08:07
dasdasi installed ubuntu - i wiped out an entire hard disk and installed it.08:07
dasdaswhen i decided to clean it - format the entire hard disk and run and install windows again, it gives the error message:08:08
dasdasGRUB loading, please wait ...08:08
wasabi /join #ubuntu08:08
dasdasthe above problem is serious08:16
dasdasplz someone give attention to it08:17
Riddelldasdas: it's a user problem, please ask in the correct channel #ubuntu08:17
danielssssssssseeeeeeeeeebbbbbbbbbbbbbbbb08:30
danielsseb128: i can't set .pngs as the background, seemingly08:30
danielsseb128: (hoary; i can add them, click them and see a preview, but when i select them and hit close, it stays on chocolate ubuntu)08:31
seb128daniels: killall nautilus ?08:35
danielsseb128: background remains the same08:36
danielsah no wait, it changed after a while08:36
seb128weird08:36
seb128ok08:36
danielscheers08:37
danielsright, dinnertime now08:37
sivangpeople people, do we have an eta for gnome return for hoary? and am I the only one not affected because of not doing dist-upgrade? :)08:46
dokosivang: what's wrong?08:47
sivangdoko : let's check :)08:49
=== lamont_r [~lamont@mesaradio41.customer.frii.net] has joined #ubuntu-devel
truluxdoko, woo woo :)08:51
truluxdoko, did you read the last reply on the bug thread?08:51
dokotrulux: which bug thread?08:55
lamont_rgrumble.  someone else is hogging the coffee-shop bandwidth08:55
truluxdoko, GCC one08:55
dokotrulux: these are not separate packages09:02
truluxdoko, pitti didn't said diff packages, just that it could be good09:03
truluxmut again09:03
truluxdoko, tssp & pie are not enabled by default09:03
dokotrulux: no .dsc and .diff.gz on your site.09:04
truluxdoko, ok, let me upload it09:05
truluxdoko, i was busy with a talk in the Umeet09:05
trulux3 hours of talk09:05
lamont_rhrm... forkbomb.  oops.09:06
=== Capri [~makolb@217.188.235.109] has joined #ubuntu-devel
truluxlamont_r, heh09:09
=== pitti [~martin@195.227.105.180] has joined #ubuntu-devel
truluxpitti, woo woo09:10
lamont_rtrulux: launching 500 wgets in parallel can be a bit bad... :-)09:11
truluxlamont_r, O_o09:11
lamont_rOTOH, the coffee shop's AP has only fallen over once, that I can tell... :-)09:11
ogralamont_r:  launching 500 wgets in parallel ? is it an amsterdam coffee shop ?09:14
lamont_rogra: fort collins colorado, usa09:14
ograi thoght so :)09:14
=== lamont_r watches wave 2 of the timeouts go through.
truluxpitti, hey09:23
truluxpitti, when do you we could have uploaded the gcc packages?09:23
truluxi'm uploading the diff09:23
truluxbtw09:23
pittitrulux: I promise I will take a look at them tomorrow09:26
pittitrulux: where are your packages?09:29
truluxon http://d-sbd.alioth.debian.org/apt/hoary09:30
=== mvo__ [~egon@ip181.135.1511I-CUD12K-01.ish.de] has joined #ubuntu-devel
pittitrulux: this directory only contains the debs; can you please put the source packages there, too?09:32
truluxpitti, umm, ok, but BIG upload09:32
truluxlet me do it09:32
pittitrulux: ahem, and can you please rename the packages to *-ssp?09:32
pittitrulux: we must not touch the current packages in main right now09:33
pittitrulux: so big?09:33
lamont_rtrulux: if you're uploading to ubuntu, it's source-only...09:33
pittitrulux: you do not need to put the orig.tar.gz there09:33
pittilamont_r: no, it's a private archive for testing purposes09:33
lamont_rah, ok09:33
=== lamont_r was concerned there for a minute..
pittitrulux: I can download the orig.tar.gz from archive.u.c09:34
truluxok09:34
truluxthen i put the diff09:34
pittilamont_r: you should rather be concerned about if we uploaded them into main right now :-)09:34
pittitrulux: and the dsc09:34
pittitrulux: but before you do that, can you please rename the package?09:35
pittitrulux: right now I cannot install it without overwriting my current gcc09:35
truluxpitti, it wouldn't change anything09:35
truluxi mean, nothing enabled by default09:36
truluxwithout the wrapper that gcc is same as gcc-3.4 one09:36
truluxor without explicitly enabling things09:36
truluxthat's because i haven't renamed it09:36
pittihmm, okay09:36
pittiThen I can _test_ the packages as they are (please send the source anyway)09:36
truluxok09:37
pittiI thought the patch would modify gcc proper to add something like -fstack-smash-protector?09:37
pittitrulux: I now see two *.dsc files, but no diff.gz09:39
truluxuploading09:39
truluxmy badnwidth sucks09:40
truluxpitti, 83%09:44
trulux87%09:44
truluxpitti, DONE09:47
=== pitti downloads
pittitrulux: but AFAIUI the patch _has_ to modify gcc, right?09:49
pittitrulux: only providing a wrapper is certainly not sufficient to spit out different binaries09:50
pittitrulux: are you sure this is everything? the interdiff is only 6 kB09:52
pittitrulux: and most of the diff is bureaucracy09:52
pittitrulux: is the actual SSP code already contained in the original package?09:52
pittidoko: ^09:52
pittitrulux: thanks for preparing the package, I will play around with it tomorrow.09:54
=== pitti grabs some food
truluxpitti, yes09:57
truluxpitti, everything is inside09:57
truluxpitti, anyway, to prevent unexpected errors, i'm uploading also the gcc-3.4 original (or seems to be the orig one) diff09:58
=== Simira [~rpGirl@ti511220a080-1777.bb.online.no] has joined #ubuntu-devel
pittitrulux: that means that the required patch only needs to activate the stuff that is already in the package?10:11
pittitrulux: I will scrutinize the interdiff tomorrow, but from a first glance it seems to be pretty safe10:12
truluxok10:12
truluxyes , supposed to ;)10:12
=== Ygster [~ygster@82.67.136.38] has joined #ubuntu-devel
dokopitti, trulux: the size of the patch is not the problem. are you willing to recheck every bug report with an unpatched gcc version, on every architecture, so that it can be forwarded and accepted by upstream?10:17
pittidoko: as long as this is in universe, I don't really see a problem with this10:20
lamont_rpitti: right.  supporting this is certainly beyond our scope...10:20
pittidoko: if the patch really just adds a wrapper around gcc, then we might even put the package in main later and deactivate it by default10:20
pittidoko: but I did not look at the patch so far10:20
truluxdoko, it's the idea10:21
truluxpitti, to make it breaking something you need to acticvate it explicitly10:21
pittidoko: as soon as the patch modifies anything in gcc (i. e. compiled binaries with deactivated SSP can be different than with the current gcc), the package should stay in universe for Hoary10:21
pittitrulux: is there any chance of modifying behaviour if SSP is turned off?10:22
lamont_rpitti: if the code is present in the compiled gcc-binary, whether on or not, then you have the support issue with upstream.10:22
lamont_rthat is, the binaries for gcc can't really change if we want to be able to support it in hoary.10:22
truluxpitti, it's not on GCC as lamont_r says10:23
truluxit just needs to load it from libssp10:23
truluxno issues about that10:23
pittilamont_r: but I thought it already is present somehow? because the harden patch does not add real code10:23
truluxonly HPPA seems not fully supported10:23
trulux(not working...)10:23
lamont_rif it doesn't add real code, then it's possibly OK.10:23
truluxupgrowing stacks archs10:23
truluxit loads code from libssp10:23
truluxno crap in libgcc10:23
truluxnor glibc10:23
lamont_rLD_PRELOAD?10:23
truluxnope10:23
truluxshared lib dinamically linked10:24
pittilamont_r: it links to an additional library libssp10:24
lamont_roh, the wrapper just preloads it. ok10:24
trulux-lssp -fstack-protector10:24
pittilamont_r: I don't think it uses preloading10:24
truluxwrapper justs sets if using default PIE and default SSP or SSP all10:24
lamont_rso you're telling me that the gcc-binary is different (extra library)?10:24
pittilamont_r: it seems to be normal linking10:24
truluxit does not10:24
truluxyeah, pitti right10:24
dokowe can add your additional dpatch file (if you document it ;-P), but not enable it by default. this way you can rebuild an ssp enabled gcc with less effort (to put in in universe). what's the problem having it available with another name?10:24
pittidoko: you mean a dpatch which changes debian/control?10:25
dokolamont_r: the gcc binary is different, because trulux applies the protector patch.10:25
lamont_rdoko: and that probably creates pain for upstream support, yes?10:26
truluxdoko, you forgot it uses libssp10:26
trulux_LIBSSP_PROVIDES_SSP_10:26
lamont_rand gcc links with it.10:26
truluxmeans, no code tweaking except for specs10:26
dokolamont_r: yes10:26
doko+# Hardened Debian patches10:26
doko+# lorenzo@gnu.org10:26
doko+debian_patches += protector10:26
truluxdoko, libssp.dpatch10:26
pittitrulux: please don't get us wrong; I really love to see this working, but I'm paranoid about breaking _anything_ and supporting it afterwards10:27
dokotrulux: what about having a separate binary in universe and getting some experience with it?10:28
lamont_rpitti: in the ideal world, the package would Depend: gcc-3.4, and provide whatver shimming was needed to make it not have to change the gcc-3.4 debs at all.10:28
pittilamont_r: sounds like a matter of providing a runtime hook in gcc10:29
pittiif that were possible, it would really rock10:29
lamont_rpitti: or creative use of existing loader functions, maybe.10:29
truluxa good gcc-config like10:30
truluxto switch security profiles10:30
truluxwhich we are working on10:30
truluxbut gcc needs to be dependent of libssp if it explicitly uses it10:30
truluxor even include then the SSP inside the glibc10:30
truluxbut it's  not the right way in our opinion10:31
pittitrulux: why not? the gcc-ssp package would depend on libssp-dev and gcc10:31
lamont_rtrulux: another good thing to do would be to break the patch up into rational-sized bits and feed them upstream.  (splats of the full diff tend to not make it..)10:31
pittitrulux: and the binaries are linked against -lssp anyway, so would automatically depend on libssp10:32
truluxpitti, great10:32
truluxthen we can do it forward10:32
dokolamont_r: i think the upstream patch author is working on it10:32
lamont_rdoko: cool10:33
pittitrulux: AFAICS the only real patch to gcc.c is the change of const char* link_spec, right?10:33
=== Ygster [~ygster@marriere-1-82-67-136-38.fbx.proxad.net] has joined #ubuntu-devel
truluxright10:33
pittitrulux: if we would replace the static string assignment with something runtime evaluated, the hook should be possible without recompiling gcc, right?10:33
dokopitti: no, please look at the protector patch10:35
pittidoko: what I don't understand: the current package already contains the actual ssp code, but does not contain this simple #ifdef'ed patch to actually use it?10:35
pittidoko: ah, it overwrites protector.c with a new version, it seems10:35
pittidoko: so is the current protector.c just a stub?10:35
truluxpitti, right but ssp code stills need to be recognized10:36
dokothe whole patch isn't applied in the current packages. it just sits there.10:36
pittidoko: ah10:36
truluxdoko is right10:36
pittidoko: I was confused because I did not see a file protector.c in the interdiff10:36
pittithe gcc-spec-env.dpatch seems uncritical to me10:36
tsengit clears some environment variables10:37
tsengon SUID10:37
tsengthere are some info leaks using LD_*10:37
pittitseng: huh? I don't see anything like this?10:37
pittitseng: it seems to hook an additional spec file into the chain10:38
pittitseng: and reads the name from env["GCC_SPECS"] 10:38
tsengoh sorry, im thinking glibc-env.dpatch10:38
tseng(which i was just looking at)10:38
tsengtrulux: you put in the tigger specs-env thing?10:39
truluxtseng, yeah, it's not really good but useful anyway10:41
tsengyeah...10:41
tsengi dont think even h-gentoo is using it yet?10:41
truluxit's using it i think , no idea10:41
truluxour wrapper now works independent of that10:41
truluxwe are moving in a new way, studying things further10:42
tsengill finish reading the backlog..10:42
truluxfine10:42
jdubmorning all10:42
tsenglo jeff10:42
truluxbtw, my work with the NSA stuff (selinxu backport) is also distracting my fun times10:42
=== jdub boggles at stupid u-u threads
tsengwhy are you backporting to 2.4..?10:43
truluxtseng because of many things, mainly that enterprises are not going to migrate to 2.6 if the don't have moe than 4 cpus (on SMP related discuss)10:43
pittiHi jdub 10:43
truluxalso that 2.4 is well tested, works fine, it has no monkey/kamikaze changes in the source and so on10:44
seb128hey hey jdub 10:44
truluxpitti, i must go to sleep10:48
pittitrulux: me too :-)10:49
truluxall of you know my email so, send there any thing on them10:49
pittigood night and thanks for your work10:49
=== lamont_r goes to fetch kids. bbiah
truluxlorenzo@gnu.org10:49
truluxsend there any q's10:49
truluxbye10:49
ograhey jdub10:51
jdubyo ogra 10:52
ograjdub: http://www.grawert.net/xss_shot_1.png10:53
sivangyo jdub dude10:53
jdubogra: aha, now that's looking muuuuuch nicer :)10:56
ograjdub: everything xft (even the drawings including the cursor) ;) 10:56
ograjdub: next will be utf8 (i want the dots for the pw !)10:56
jdubheh10:56
ograjdub: and probably a fade to transparency ? (by just blending over the saved under pixmap)10:57
jdubthat's getting towards diminishing returns10:58
jdubonce we have a proper gtk+ version, and full composite support, we won't have to work too hard for it10:58
=== seb128 is curious to know what jdub call a proper gtk version :)
ograjdub: ok, i'll stop at utf8 (and probably rtl) there seems no .po file for the lock code10:59
jdubseb128: not the sun/ximian hack -> something acceptable to jwz11:00
ograjdub: in fact it looks like the thing was never touched since ages11:00
seb128jdub: hum. What kind of hack is that ?11:00
seb128jdub: BTW the debian menu is back in hoary :p11:01
ogralol11:01
jdubseb128: they just did a gross, relatively insecure hack to make the lock dialogue gtk+11:03
jdubseb128: isn't that what you showed me at matar?11:03
seb128yep11:03
seb128with a debian swirl as icon11:03
jduboh good11:03
seb128and you're right, the menu is kind of slow to be built after changes11:04
seb128I've noticed it while I was testing the changes11:04
jdubwhen popping open from the top level?11:04
seb128yep11:04
=== sivang had done dist-upgrade, to relogin again or not? this is the question....;-)
wasabianybody fooled with preseeding base-config in hoary?11:04
wasabioh. n/m. I just had an epiphany and realized my problem.11:05
=== mdke [~Matt|@mdke.user] has joined #ubuntu-devel

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!