[Cooker] [Bug 4073] [libxine1] segmentation fault in libxine.so.1(xine_print_trace+0x21)
http://qa.mandrakesoft.com/show_bug.cgi?id=4073 [EMAIL PROTECTED] changed: What|Removed |Added Status|UNCONFIRMED |RESOLVED Resolution||FIXED --- Additional Comments From [EMAIL PROTECTED] 2003-18-06 10:42 --- I think it's fixed in xine-vcdx-1-0.beta12.2mdk, please confirm. -- Configure bugmail: http://qa.mandrakesoft.com/userprefs.cgi?tab=email --- You are receiving this mail because: --- You are on the CC list for the bug, or are watching someone who is. --- Reminder: --- assigned_to: [EMAIL PROTECTED] status: RESOLVED creation_date: description: With current cooker (it worked some days ago, so this is probably because of some recent changes in cooker?): $ LANGUAGE=c xine This is xine (X11 gui) - a free video player v0.9.21 (c) 2000-2003 by G. Bartsch and the xine project team. Built with xine library 1.0.0 (1-beta12) Found xine library version: 1.0.0 (1-beta12). XServer Vendor: Mandrake Linux (XFree86 4.3, patch level 9mdk). Release: 4030, Protocol Version: 11, Revision: 0, Available Screen(s): 1, using 0 Depth: 16. -[ xiTK version 0.10.3 ]- -[ WM type: (GnomeCompliant) (EWMH) Metacity {Metacity} ]- Display is not using Xinerama. Benchmarking memcpy methods (smaller is better): glibc memcpy() : 449546920 linux kernel memcpy() : 449568179 MMX optimized memcpy() : 448746106 MMXEXT optimized memcpy() : 297737976 SSE optimized memcpy() : 355031658 configfile.c:_xine_config_register_string:217: assertion `def_value' failed. Default value is NULL. This is a required argument. Obtained 1 stack frames. /usr/lib/libxine.so.1(xine_print_trace+0x21) [0x4008a681] zsh: segmentation fault (core dumped) LANGUAGE=c xine
[Cooker] Re: [CHRPM] tetex-2.0.2-1mdk
This update stomps on Olivier's fix (2.0.1-7mdk) Giuseppe Ghibò wrote: --=-=-= Name: tetexRelocations: (not relocateable) Version : 2.0.2 Vendor: MandrakeSoft Release : 1mdk Build Date: Mon Jun 16 19:36:29 2003 --=-=-= * Mon Jun 16 2003 Giuseppe Ghibò [EMAIL PROTECTED] 2.0.2-1mdk - passivetex 1.24. --=-=-= @@ -371,7 +375,7 @@ tar cf - texmf | tar xf - -C $RPM_BUILD_ROOT%{_datadir} %makeinstall texmf=$RPM_BUILD_ROOT%{_datadir}/texmf -export PATH=$RPM_BUILD_ROOT/%{_bindir}:$PATH + # jadetex man page (cd %{jadename}-%{jadeversion} install -m 644 jadetex.1 pdfjadetex.1 $RPM_BUILD_ROOT%{_mandir}/man1 @@ -666,8 +670,8 @@ %defattr(-,root,root) %changelog -* Wed Jun 18 2003 Olivier Thauvin [EMAIL PROTECTED] 2.0.1-7mdk -- fix build when tetex is not allready installed (can be usefull !) +* Mon Jun 16 2003 Giuseppe Ghibò [EMAIL PROTECTED] 2.0.2-1mdk +- passivetex 1.24. * Thu Mar 13 2003 Giuseppe Ghibò [EMAIL PROTECTED] 2.0.1-6mdk - Added missed xmltex.cfg (fixes problem reported by --=-=-=
[Cooker] perl-XML-LibXML
hello in 9.1, and possibly in cooker, perl-XML-LibXML is packaged to work with the libxml2 of the distribution, version libxml2-2.5.4. The problem is that this version of libxml2 is known to not work correctly with the perl binding. If you try to install perl-XML-LibXML from cpan, it will say that you have no libxml2 installed, because in fact, the Makefile.PL has a blacklist of know libxml2 version that don't work. 2.5.4 is one of them. the resulting perl-XML-LibXML (and common, and possibly other perl binding) doesn't work in many cases, and the errors are not obvious ( most of the times segfaults of the libxml2 lib). The solutions could be to provide a libxml2 that is know to work, and package the perl binding against it. Maybe have 2 version of libxml2 : one for C, and another for perl binding. I'm convinced that this problem is not restricted to libxml2. We could try to maintain a library blacklist as they do in the CPAN perl-XML-LibXml Makefile.PL. By the way, if you don't have time to repackage the whole thing, I can try to do it, but I think it has t be decided what to be done : don't change anything, put an older libxml2 in the distro, or have 2 libxml2, one for C, one for perl. my 2 cents -- dams
[Cooker] Re: Re[2]: [PATCH] Final mkinitrd support for 2.5 (on ix86 at least)
Andrey Borzenkov [EMAIL PROTECTED] writes: There are several questions here ... - no, I do not work together with either of them. Juan did not reply to any of my mails the past several months so I assume he is just too busy and I do not bother him anymore. Ok. - it is up to respective maintainers to evaluate changes and decide. I tried to make them as non-intrusive as possible, but regression is always possible. At least gkrellm patch has been accepted by author :) As for myself, mkinitrd seems non-intrusive enough to be accepted as well. But the reason for me to be mkinitrd maintainer is not so clear/logical, since I'm not a kernel guy. It was done originally when I had to choose some packages to maintain, because I knew and used mkinitrd on my RH box before joining mdk, that's all :). But currently, we don't have enough kernel guys so I think they can't maintain one more package anyway. - yes, I think, they shoud be uploaded. First, here is clear interest in 2.5 now and it is in quite usable state (this or that driver may be broken, but core appears to work). Second, and it is the main reason - we do need experience and testing and it needs time. There are many things that need clarification. Please remember devfs - it took for over a year and two releases before it was put in some usable shape. I'm asking myself if it would not be good to add a hackkernel2.5 in contrib? - finally, it all depends on Mandrakesoft policy. If you (meaning Mandrakesoft) intend to ship 2.5 with next release - not as main kernel of course, but as alternative - we must start to evaluate what is needed to support it as soon as possible. My RPMs are just bare minimum - someone has to inegrate 2.5 support into drakx, and that is hard work, and there are a lot of applications needing updating/patching. I have no idea about that. Frederic Lepied in CC should be able to answer that question. From the top of my head, I think alternatives for 9.2 are: 1- ignore 2.5 (well that's already not really the case, module-init-tools being in cooker) 2- provide a 2.5 in contrib but call it hackkernel2.5 and don't allow installing it from the install, so that it's clear it's bleeding not supported stuff 3- provide a 2.5 in cooker and allow installing it from the install (probably not as default boot image) Since 9.2 is to be frozen mid-august maybe (2) is the best solution? And yes, I think next release should include support for 2.5. At this time given the development rate it should be near to final form. Ah, ok. That said - I do use those RPMs for myself, I will continue to work on them, but I have almost reached my limits - they do work for me so unless I will get feedback I do not have anything really important to change ... (I do have some plans :) Ok. -- Guillaume Cottenceau - http://people.mandrakesoft.com/~gc/
[Cooker] mail?
Re: [Cooker] rpmdrake and newbies: they sometimes miss *installed* software
1 - The original point I was commenting on was that the 'drake' installer was getting more complicated by being broke into several sections. This is a fundamentally flaw concept for newbies to take on. We can disagree on this, but having more than one program to take care of installing and removing software might just be more involved. so, i guess you propose to have only one rpmdrake program ? Now, , for the same reason, should we unite all wizard related to Network in one super wizard ? No, i don't think. This would be confusing. But the same applies, why should people have one program to share the connection, and one to edit the firewall rules ? Don't forget that rpmdrake should not be run from the command line, but from the MCC or from kde-gnome-whatever menu. So, of course, they are two separate program, but, since they are designed to be launched as module in the mcc, this make sense. I agree, maybe mandrake lacks a powerful package management graphical tool, but, since ressources are scarce, they needed to choose between newbie and more advanced user, and clearly, they favored the newbie. -- Michaël Scherer
[Cooker] Evolution locks hard when a hyperlink is clicked
When I click a hyperlink in the Evolution summary screen or in a mail message, Evolution locks up hard. I have to kill the processes to terminate it. I have reproduced this problem on 2 separate Cooker machines, both fully up to date. Thanks, -- Jason Komar [EMAIL PROTECTED] Lubetec
Re: [Cooker] rpmdrake and newbies: they sometimes miss *installed* software
Le mar 17/06/2003 à 22:10, andre a écrit : On Tuesday 17 June 2003 22:21, Guillaume Cottenceau wrote: Argh. another addition to the GUI.. More like rpmdrake and rpmdrake-remove are just not the right tool for non-root users to select rpms. That's sounds sensible to me IMHO. Indeed users need to know which packages are installed and which ones can be installed, and for this need search features ( name, category, summary, origin ). So the best way is to do a separate tool. keep rpmdrake and rpmdrake-remove small ( so don't need to increase startup time ). Do a separate tool ( application finder, Drake Carpet, or whatever u want ) this one will have a longer startup time ( to compute installed and uninstalled packages ), one window, search criteria, grouping. Search results precise if package is installed or not ( with a color or with a checkbox ) and of course description of the package. Now I just want to make an assumption : users should not have access to package file list or at least for core or important or server software. To my mind it is a security risk as users doesn't need to knwo where they are. Maybe binary/doc for common packages but for core packages/server/critical packages they should not see them.
[Cooker] skel.spec
Is this file reserved to MandrakeClub users? https://people.mandrakesoft.com/~lenny/files/skel.spec :-)
[Cooker] Re: [PATCH] Final mkinitrd support for 2.5 (on ix86 at least)
On Wednesday 18 June 2003 20:37, Pixel wrote: Andrey Borzenkov [EMAIL PROTECTED] writes: BTW David found that DrakX creates version 1 swap and it is not supported by 2.5 anymore. Pixel, any reason for that? nope, i changed it to use mkswap and so v1 by default erm sorry, I meant version 0 of course. Quoting: Well, just as I was writing that last sentence I took another look at my log, and noticed something I hadn't seen before: Jun 14 12:20:08 mario kernel: version 0 swap is no longer supported. Use mkswap -v1 /dev/hda2 Did DrakX really create a version 0 swap? It could be formatted ages ago. Have you ever said you want your swap to be formatted? It turns out that DrakX must have created a version 0 swap (how weird!). I did a mkswap -v1 /dev/hda2 and now booting 2.5 it doesn't complain anymore.
Re: [Cooker] 2.5 Mandrake Kernel
On Wednesday 18 June 2003 21:43, Edward Tandi wrote: On Wed, 2003-06-18 at 06:48, Andrey Borzenkov wrote: 3) rmmod still not working -Can't open 'analog': No such file or directory. I presume this is fixed by Andrey's recent fixes. As I have never seen it before I cannot be sure. Actually I have not touched utilities themseves (only packaging). You have compiled 2.5 with module unloading support, have not you? It is not default. Yes. Both module unloading and module-force unloading. ^^ may be it was the last straw? :)) Ed-T. -andrey
Re: [Cooker] rpmdrake and newbies: they sometimes miss *installed* software
andre [EMAIL PROTECTED] writes: On Tuesday 17 June 2003 22:21, Guillaume Cottenceau wrote: Argh. another addition to the GUI.. More like rpmdrake and rpmdrake-remove are just not the right tool for non-root users to select rpms. I would remove your May 28 addition. Buchan's suggestion was not for selecting rpm's, it was for allowing non-root users to browse, so that afterwards they can request software installation to the machine's administrator. -- Guillaume Cottenceau - http://people.mandrakesoft.com/~gc/
Re: [Cooker] rpmdrake and newbies: they sometimes miss *installed* software
I think that urpmi need to introduce meta-package like we have on debian with apt... For exemple, it will be cool to do an urpmi kde to install all kde package or an urpmi Xwindow, urpmi gnome, ... With this, installation of software will be easiest for newbies... The actual problem is that newbies don't know what to install: kdebase, kdenetwork, Le mer 18/06/2003 à 03:37, w9ya a écrit : Yesterday I was at a client's. I was installing software using the 'drake' gui tools while they watched. The clients remarked that they were sure glad they didn't have to install software on linux. They went on to say the they were use to a much easier install process. They are not stupid. They are not slow. Bob Finch On Tuesday 17 June 2003 02:14 pm, andre wrote: On Tuesday 17 June 2003 15:39, Guillaume Cottenceau wrote: Buchan Milne [EMAIL PROTECTED] writes: And of course, at present rpmdrake can only be run as root. If it were Untrue! This time you suck.. you asked for that feature, I've said I would add it, I've added it on May 28, but you didn't even notice :(. The way it is implemented sucks. Not only would you expect that one would find a request to get root but also to find installed and uninstalled software if you did a search.
[Cooker] Re: Re[2]: [PATCH] Final mkinitrd support for 2.5 (on ix86 at least)
Andrey Borzenkov [EMAIL PROTECTED] writes: BTW David found that DrakX creates version 1 swap and it is not supported by 2.5 anymore. Pixel, any reason for that? nope, i changed it to use mkswap and so v1 by default thanks
Re: [Cooker] mail?
The servers were down for about 20 hours. Just went back up about 45 minutes ago or about. --LX -- Kernel 2.4.21-0.13mdk Linux Mandrake 9.1 Enlightenment-0.16.5-12mdk Evolution 1.2.4-1.1mdk Linux User #268899 http://counter.li.org/
[Cooker] gcc doesn't rebuild (on i586 and alpha)
gcc currently doesn't rebuild on both i586 and alpha. It seems to be related to some yacc errors: gcc -c -g -DIN_GCC -W -Wall -Wwrite-strings -Wstrict-prototypes -Wmissing-prototypes -Wtraditional -pedantic -Wno-long-long -DHAVE_CONFIG_H -DGENERATOR_FILE-I. -I. -I../../gcc -I../../gcc/. -I../../gcc/config -I../../gcc/../include \ ../../gcc/gengtype-yacc.c -o gengtype-yacc.o gengtype-yacc.c: In function `yydestruct': gengtype-yacc.c:798: warning: traditional C rejects ISO C style function definitions gengtype-yacc.c: In function `yyparse': gengtype-yacc.c:861: warning: traditional C rejects ISO C style function definitions gcc -g -DIN_GCC -W -Wall -Wwrite-strings -Wstrict-prototypes -Wmissing-prototypes -Wtraditional -pedantic -Wno-long-long -DHAVE_CONFIG_H -DGENERATOR_FILE -o gengtype \ gengtype.o gengtype-lex.o gengtype-yacc.o ../libiberty/libiberty.a gengtype.o(.text+0x90d): In function `adjust_field_rtx_def': ../../gcc/gengtype.c:430: undefined reference to `lexer_line' gengtype.o(.text+0xb03):../../gcc/gengtype.c:473: undefined reference to `lexer_line' gengtype.o(.text+0xd3a):../../gcc/gengtype.c:530: undefined reference to `lexer_line' gengtype.o(.text+0xdde):../../gcc/gengtype.c:573: undefined reference to `lexer_line' gengtype.o(.text+0xf74):../../gcc/gengtype.c:620: undefined reference to `lexer_line' gengtype.o(.text+0x1053):../../gcc/gengtype.c:633: more undefined references to `lexer_line' follow gengtype.o(.text+0x52e0): In function `main': ../../gcc/gengtype.c:2609: undefined reference to `parse_file' gengtype-yacc.o(.text+0x25f): In function `yyparse': /home/cooker/RPM/BUILD/gcc-3.3/obj-i586-mandrake-linux-gnu/gcc/gengtype-yacc.c:1023: undefined reference to `yylex' gengtype-yacc.o(.text+0x41f): In function `yyparse': /home/cooker/RPM/BUILD/gcc-3.3/obj-i586-mandrake-linux-gnu/gcc/gengtype-yacc.y:70: undefined reference to `lexer_line' gengtype-yacc.o(.text+0x484):/home/cooker/RPM/BUILD/gcc-3.3/obj-i586-mandrake-linux-gnu/gcc/gengtype-yacc.y:72: undefined reference to `lexer_line' gengtype-yacc.o(.text+0x4a3):/home/cooker/RPM/BUILD/gcc-3.3/obj-i586-mandrake-linux-gnu/gcc/gengtype-yacc.y:73: undefined reference to `lexer_toplevel_done' gengtype-yacc.o(.text+0x4c6):/home/cooker/RPM/BUILD/gcc-3.3/obj-i586-mandrake-linux-gnu/gcc/gengtype-yacc.y:79: undefined reference to `lexer_line' gengtype-yacc.o(.text+0x529):/home/cooker/RPM/BUILD/gcc-3.3/obj-i586-mandrake-linux-gnu/gcc/gengtype-yacc.y:81: undefined reference to `lexer_toplevel_done' gengtype-yacc.o(.text+0x537):/home/cooker/RPM/BUILD/gcc-3.3/obj-i586-mandrake-linux-gnu/gcc/gengtype-yacc.y:89: undefined reference to `lexer_line' gengtype-yacc.o(.text+0x576):/home/cooker/RPM/BUILD/gcc-3.3/obj-i586-mandrake-linux-gnu/gcc/gengtype-yacc.y:94: undefined reference to `lexer_line' gengtype-yacc.o(.text+0x5b5):/home/cooker/RPM/BUILD/gcc-3.3/obj-i586-mandrake-linux-gnu/gcc/gengtype-yacc.y:99: undefined reference to `lexer_line' gengtype-yacc.o(.text+0x60a):/home/cooker/RPM/BUILD/gcc-3.3/obj-i586-mandrake-linux-gnu/gcc/gengtype-yacc.y:107: undefined reference to `lexer_toplevel_done' gengtype-yacc.o(.text+0x626):/home/cooker/RPM/BUILD/gcc-3.3/obj-i586-mandrake-linux-gnu/gcc/gengtype-yacc.y:119: undefined reference to `lexer_line' gengtype-yacc.o(.text+0x7e2):/home/cooker/RPM/BUILD/gcc-3.3/obj-i586-mandrake-linux-gnu/gcc/gengtype-yacc.y:162: undefined reference to `lexer_line' gengtype-yacc.o(.text+0x7e8):/home/cooker/RPM/BUILD/gcc-3.3/obj-i586-mandrake-linux-gnu/gcc/gengtype-yacc.y:162: undefined reference to `lexer_line' gengtype-yacc.o(.text+0x879):/home/cooker/RPM/BUILD/gcc-3.3/obj-i586-mandrake-linux-gnu/gcc/gengtype-yacc.y:173: undefined reference to `lexer_line' gengtype-yacc.o(.text+0x87f):/home/cooker/RPM/BUILD/gcc-3.3/obj-i586-mandrake-linux-gnu/gcc/gengtype-yacc.y:173: undefined reference to `lexer_line' gengtype-yacc.o(.text+0x976):/home/cooker/RPM/BUILD/gcc-3.3/obj-i586-mandrake-linux-gnu/gcc/gengtype-yacc.y:190: more undefined references to `lexer_line' follow gengtype-yacc.o(.text+0x1144): In function `yyparse': /home/cooker/RPM/BUILD/gcc-3.3/obj-i586-mandrake-linux-gnu/gcc/gengtype-yacc.c:1504: undefined reference to `yyerror' gengtype-yacc.o(.text+0x1156):/home/cooker/RPM/BUILD/gcc-3.3/obj-i586-mandrake-linux-gnu/gcc/gengtype-yacc.c:1508: undefined reference to `yyerror' gengtype-yacc.o(.text+0x1168):/home/cooker/RPM/BUILD/gcc-3.3/obj-i586-mandrake-linux-gnu/gcc/gengtype-yacc.c:1512: undefined reference to `yyerror' gengtype-yacc.o(.text+0x12fe):/home/cooker/RPM/BUILD/gcc-3.3/obj-i586-mandrake-linux-gnu/gcc/gengtype-yacc.c:1626: undefined reference to `yyerror' collect2: ld returned 1 exit status make[2]: *** [gengtype] Error 1 make[2]: Leaving directory `/home/cooker/RPM/BUILD/gcc-3.3/obj-i586-mandrake-linux-gnu/gcc' make[1]: *** [stage1_build] Error 2 make[1]: Leaving directory `/home/cooker/RPM/BUILD/gcc-3.3/obj-i586-mandrake-linux-gnu/gcc' make: *** [bootstrap-lean] Error 2 error:
Booting OS X off of 1394 drive with OF
OK, when I installed mdk, i bought a new 1394 external drive, and i soon realized i couldn't install mdk directly on to it. so i copied all my OS X stuff over to the 1394 drive, initialized my internal HD, and installed on there. now the problem is that i cannot boot OS X off of my firewire drive. Since OF does not recognize 1394 drives, I cannot just boot into OS X using yaboot. Stew suggested that i set the boot-device in OF to blank. I did this, and restarted, and it searched for a start-up volume for a few seconds, then loaded yaboot. I am completely stuck. Can anyone help me? Please note that I am a linux and OF newbie so you will have to give me step-by-step instructions for everything. Thank you, Micah Buckley-Farlee