Re: [Cooker] mmx and such
[snip] It's not a matter of complaining or not, it's a matter of minimal support. Mandrake Linux is supposed to run on any i586 or newer processor. Period. Even if that means a crazy person is encoding videos on a i586(no MMX), say, 166MHz (or somewhere near) without hardware acceleration Must be surely someone with paaaiinnnccce! Guy
Re: [Cooker] Re: War (Warning: Even more OT)
On 13 Mar 2003, James Sparenberg wrote: On Thu, 2003-03-13 at 15:03, Adam Williamson wrote: On Thu, 2003-03-13 at 19:56, Francisco Alcaraz Ariza wrote: [snip] If you're going to expect the Bushes to know about Spanish and Russian politics, maybe you should learn to spell George Bush? Fair's fair. :) (At least he knows who he is talking about. Nothing Fair is fair. Do make a fair COMPARISON.) if it's daddy it's George Bush If it's the son it's GeeDubya or Das shrubenmeister Thank you so much!! Now I finally know why he's called Dubya i.e. Dubble U i.e. W. in The Daily Dirt :-) Guy
Re: [Cooker] Cooking on cooker [ Was : WAR ]
On Wed, 12 Mar 2003, Michael Scherer wrote: And of course neither french fries nor french toast have a thing to do with France. In the first case, IIRC, the dish in question originated in Belgium Are you sure sure ? Some cooking books of my grand mother said it comes from Paris, near the Pont Neuf in 1890. It was imported by a Bruxellois around exactly that time :-) They may not be right, but, why do we call this french fries, if it comes from belgium No idea! We always felt this was unfair to us. I am happy though that someone as official as a spokeswoman from a French embasy admitted this :-) Of course, this could be just politics to make you look childish... Anyway, the story goes that the Spanish are the inventors of fried potatoes. It was a 16th century Spanish nun working in a sanitorium for poor people (-malnutrition) who recognized the nutritional value of potatoes (brand new at that time) when fried (chopped in pieces) in oil. (Note that skinned potatoes are low on vitamines but still high on CH's and fat (-the oil).) It spread to our regions during the Spanish Occupation but it was only much later that the fries were served in fancy rods like now. I am not sure but I think that Brussels fellow was the first when he opened his Friterie (Frituur) in Paris (causing a French Fries Frenzy :-). I am not sure where I read this, but I think in a history of recipes book (which I never bought, BTW). So, consider this anecdotal... Sorry, my last post on this OTS, Guy Bormann
Re: [Cooker] Re: War
On Wed, 12 Mar 2003, Francisco Alcaraz Ariza wrote: Don't you think they are absolutely crazzy?. Well, actually I am sure that there are lots of Americans that don't think in this way. Here in Spain more than 90% of the people don't wont the war, but for our president this doesn't matter :-(; I hope in the next elecctions we will give him very, very longs holidays. Since when has Spain become a republic again??? :-) (You mean Minister-President??) Guy
Re: [Cooker] ACPI Testing
On 22 Feb 2003, Adam Williamson wrote: On Fri, 2003-02-21 at 22:45, Buchan Milne wrote: [snip] I get the feeling if someone would just write some damn *documentation* for the current ACPI implementation we'd find out all sorts of cool things it does. As it stands, anyone who isn't a developer is fumbling around in the dark... Anyone who can get a copy of the Jan/Feb 2003 issue of c't (Dutch/German magazine) (and understands either language :-) : there is a nice introductory article on ACPI and Linux. Very informative!! Guy
Re: [Cooker] Learning from windows ...
On Wed, 19 Feb 2003, Buchan Milne wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Interesting reading: http://www.winsupersite.com/reviews/winserver2k3_gold2.asp Have you seen Tux next to the TV set on the first picture? :-) [snip] Guy Bormann P.S.: I think any large project has a large amount of branches when you have to maintain old release versions. On the other hand, excessive branching off the main trunk is asking for trouble. How do they handle all the critical conflict resolution near release? (I can see all these security fixes sitting in the version control system, waiting and waiting to never make it in the release version :-)))
Re: [Cooker] [Bug 1649] [drakxtools] localedrake: can't chooseBelgium as country when choosing English language
On 17 Feb 2003, Guillaume Cottenceau wrote: fhimpe [EMAIL PROTECTED] writes: https://qa.mandrakesoft.com/show_bug.cgi?id=1649 [snip empty lines] --- Additional Comments From [EMAIL PROTECTED] 2003-02-15 23:30 --- I had installed locales-nl. As Dutch is one of the three official languages of Belgium (French and German being the other languages), I don't see why I should also install locales-fr to be able to choose Belgium... You chose british english - locale for that is en_GB (is you chose american english, that doesn't make any difference). Then you chose Belgium as a country - default locale is fr_BE. Oh right, and how do you get nl_BE??? Have you actually read Frederik's message? If you chose Belgium (or any other officially multi-lingual country such as Switzerland) you should be presented with a choice for the official language instead of making one language the default. If that's impossible or hard to do, so be it but at least show a message on how to get the other locale. Don't make assumptions based on common impressions... Now the program tries to see if en_BE exists. Since it didn't exist in our list of available locales, it splitted between the locales for languages (taking the default, en_GB) and for country (taking the default, fr_BE). So we need to install locales-fr so that glibc has necessary information about fr_BE locale. I bet he knows, this is not his point! See above... Guy P.S.: This is a new attempt at clarification. I didn't find it useful to put it in the database. You can add if you want or just resolve it. Sorry, if I am wrong.
Re: [Cooker] [Bug 1705] [Installation] RFE- Change /root/drakx/report.bug.gzto /root/drakx/bug-rprt.gz
On 17 Feb 2003, Guillaume Cottenceau wrote: mrmazda [EMAIL PROTECTED] writes: DR-DOS 8's release is imminent: http://www.drdos.com/ mwh ahahaha eheheehh. hum, sorry. Don't laugh, it is still extensively used in embedded devices and process control and it is nowhere as bad as MS-DOS (yuck!). I haven't used it for a long time, though. Anyway, it has support for FAT32 and long file names so this should be out of Felix' list (I guess you needed some weight in your list, Felix :-) ). Guy P.S.: Since when has IBM taken up OS/2 development again? I thought they only provided support and updates?
Re: [Cooker] A new Mandrake Baby!
On 11 Feb 2003, Jean-Michel Dault wrote: I'm now a new daddy.. it's a girl, Gabrielle. Join the club! Congratulations! She is a whole lot of baby :-) Is she as geeky as her daddy? My daughter was 46cm and 2kg 990g, a cute little packet :-) Guy Bormann See all the pictures at http://www.advx.org/baby Name: Gabrielle Relocations: (not relocateable) Version : 1.0 Vendor: Similac Advance with iron Release : 1mdkBuild Date: wed 22 jan 2003 10:39:00 EST Install date: (just installed) Build Host: CHUS.qc.ca Group : System/Diapers Source RPM: parents-1.0-1mdk.src.rpm Size: 53.5cm, 4.170kg License: GPL Packager: Jean-Michel Dault [EMAIL PROTECTED] URL : http://www.advx.org/baby Summary : A new Mandrake baby Description : Little Gabrielle Dault-Deblois was born January 22nd 2003 at 10:39 am at the Sherbrooke University Hospital Center (CHUS). At 9 pounds 3 ounces (4.170kg), 21 inches (53.5cm), that's not what I call little ;-)
Re: [Cooker] Re: Creation of a community ( was : the end is inevitable )
On 10 Feb 2003, Gustavo Franco wrote: On Mon, 2003-02-10 at 02:47, Greg Meyer wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Sunday 09 February 2003 06:05 pm, Gustavo Franco wrote: On Sun, 2003-02-09 at 15:36, Steve Fox wrote: On Sat, 2003-02-08 at 13:36, Gustavo Franco wrote: [snip] [edited for clarity; NO CHANGE IN WORDS] Stupid estimation! My name isn't Martin Schulze!!! ^^ ?? I was sent a mail to [EMAIL PROTECTED] ^^^^ This is at the least ambiguous. You cannot blame people for misreading your messages if the language is not unambiguous. Did you mean to say I was sent a mail FROM [EMAIL PROTECTED] ... Blame it on bad language first before blaming it on bad intentions! (And that counts for both sides in this selective reading competition subthread...) Guy Bormann P.S.: Don't bother throwing baits, I won't bite...
Re: [Cooker] Re: Creation of a community ( was : the end is inevitable )
On Mon, 10 Feb 2003, Greg Meyer wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Monday 10 February 2003 12:22 am, Steve Fox wrote: The only thing in doubt is the reference to 'Mandrake developers', which makes it sound like Mandrakesoft employees were in this discussion, which they were not. The article insinuates that the community was planning what to do in the aftermath of the failure of MandrakeSoft, which is not why the thread was started. It is an inaccurate representation of the original discussion. Blablabla... All you write is most probably true and if so I totally agree the author should be summarly executed (for the humor impaired : *flash* *flash* sarcasm! *flash* *flash*). However, Gustavo was also misrepresented as being the originator of the comments in the referred article while he was actually referred to this mailing list AFTER the fact. aaaiiipp (*gasp of air*) On the other hand, he was getting obnoxious about the fact that he was misunderstood about this while in fact his not so complete mastering of the English language was to blame. And now I am REALLY out of this stupid selective reading competition. If you don't believe, go back to the mail archive and REALLY read and read and read and read what was actually typed (in this subthread)(and I am really not so stupid to believe that all of it represents what was intended)! Guy Bormann
[Cooker] Re: Creation of community
Hi Greg, I am a complete jackass myself to not check the addressees in the reply and funny things go on with the Subject field. It was not directly pointed to you and now that these messages arrive out-of-order it seems so irrelevant. Sorry! However, my point about the pointless blaming still stands. Guy Bormann
Re: [Cooker] Kernel: ACPI/APIC bug
On Tue, 4 Feb 2003, Levi Ramsey wrote: On Tue Feb 04 19:51 -0500, Gregory K. Meyer wrote: I am seeing posts on A.O.L.M, bug reports (including mine - 115) and some posts on various other web forums related to the onboard network, sound and usb of many newer motherboards not working. I have asked people in these cases to try booting with noapic and/or acpi=off and in most cases it allowed those periphs to work. See bug 115 for a detailed report I filed on the issue on a Soyo KT400 board. When I could identify the hardware, it seems these reports came from VIA chipsets for both Athlon and Intel platforms, but I cannot confirm that universally. [snip] will be...). The onboard sound, however, apart from the annoying buzzing that has a history of plaguing VIA sound chips, works perfectly. You can get rid of it by putting the option PciRestart in XF86Config-4. At least it works for me. No more hissing, echo's, buzzing, ... Guy P.S.: I have a A7V133 (KT133A) motherboard in the computer at home. I am not completely sure of the option name, I have to look it up at home (so not for today unless someone helps me out). I also have to look up in what section exactly you have to put the option.
Re: [Cooker] win32 for 9.1? migration tools
On Sat, 1 Feb 2003, Leon Brooks wrote: On Saturday 01 February 2003 11:00 pm, Buchan Milne wrote: On Sat, 1 Feb 2003, Leon Brooks wrote: [snip] It'd make a great screenplay, and far more plausible than a Cassiopea uploading a virus to an alien spaceship... (-: actually, a virus on a Windows handheld is highly believable, but the upload, or the concept of aliens running anything vaguely Windows compatible... :-) A, come on, when will you guys realise they ARE aliens?? Guy Bormann
Re: [Cooker] mdkkdm - project goals? (So we can help)
On Mon, 3 Feb 2003, Gregory K. Meyer wrote: --- On Mon 02/03, Laurent Montel [EMAIL PROTECTED] wrote: Le Monday 03 February 2003 15:39, Jason Straight a écrit : [snip] New kdm is a just a redesign of kdm. It's a lot of work to redesign the UI of something that is very nice already. I would be more agreeable to spending time on this if there was a major feature add. I bet it will eventually have the polished user-switching feature but the competition is listening too so I'm sure you won't hear a word about it. However, sneakily changing kdm was a strategic blunder if my guessing turns out to be right... I guess I read too many Clancy/Ludlum/Carre type of novels ;-) Guy Bormann
Re: [Cooker] [Bug 1062] [drakxtools] New: No ip over 1394
On Fri, 24 Jan 2003, John Danielson, II wrote: [Bug 1062] wrote: https://qa.mandrakesoft.com/show_bug.cgi?id=1062 Product: drakxtools Component: DrakConnect Summary: No ip over 1394 Version: 9.1-0.13mdk Platform: PC URL: [EMAIL PROTECTED] OS/Version: All Status: UNCONFIRMED Severity: major Priority: P1 AssignedTo: [EMAIL PROTECTED] ReportedBy: [EMAIL PROTECTED] DrakConnect doesn't list eth1394 as a driver, and expert mode crashes when I try it. This is after I have done insmod eth1394.o.gz and re booted. Actually, this option should be offered during the install, automatically, so those of us who are relying on our firewire cards to connect two windows machines will be able to network our new Linux install to, say, our winxp machine. (My case). Even though my Linux bootup screen says eth0 is OK, my winxp machine can't tell it's there. xp says 1394 Connection A network cable is unplugged. The Linux machine is dual booted with winme which does work with ip over 1394, Internet and everything. Linux needs to do ip over 1394 out of the box, firewire has been around for some time now. And I can't find ANY documentation on the eth1394 module anywhere. Any ideas? --- You are receiving this mail because: --- You are on the CC list for the bug, or are watching someone who is. Firewire support is still being finalized, as is USB 2.0 which is being also finalized as far as things ever are in the Linux world. Kernel developers are working on this still, it is not fully and completely a part of any stock distro exept as one might consider LFS or Gentoo a stock thing (you compile from source and tweak right now to get ieee1394 support going right if you also have USB or especially USB 2.0 also, among other things). This is a linux-wide issue, not just a Mandrake one. I do not think that even a USB 2.0 direct-connect cable works in Linux commonly yet. At a guess, we will need to wait for things from 2.5 development branch of kernel to be back-ported(patched back into the 2.4 [snip] Cheapest is a tossup between 1 and 2, for Linux. John. Hi John, For some reason I like your technical posts but could you please use the Enter key some more to add some structure like paragraphs. That would make reading a lot easier ;-) Guy Bormann
Re: [Cooker] New rpmdrake
On 20 Jan 2003, Guillaume Cottenceau wrote: Benjamin Pflugmann [EMAIL PROTECTED] writes: On Tue 2003-01-14 at 05:31:22 -0800, [EMAIL PROTECTED] wrote: --- Chris Picton [EMAIL PROTECTED] wrote: [...] I feel that the radio buttons 'Normal information', and 'Maximum information' should be replaced with three notebook tabs reading: 'Information', 'Files' and 'Changelog'. The relevant information should be placed in these tabs. [...] Yes! Thanks for reminding him Chris. I write just to make clear, that that's not what all want. I hated switching between tabs all the time with the old (pre-9.0) version. I think it's probably stay like it is currently anyway: it's both more simple for newcomers and more efficient when you maximize in 1024 (you have everything in one look). How about adding fake links (since it is not HTML; or three buttons, or a context menu) so that when clicking them, the wanted section of the text is automatically scrolled to the top of the screen? This way you can mostly keep the current GUI structure while still satisfying the info junkies ;-) Guy Bormann
Re: [Cooker] 9.1 Beta 2
On Mon, 20 Jan 2003, Marcel Pol wrote: On Mon, 20 Jan 2003 11:14:50 +0100 Warly [EMAIL PROTECTED] wrote: [snip] Midnight Commander 83 (really useful ?) Yes, it's imo the best filemanager for console and X (xterm) Yep, as a filemanager on non-X servers and as a fallback when X is in trouble... Guy Bormann
Re: [Cooker] Mandrake Broke?
[snip] As far as IBM bailing out SUSE, that was also another first. They were the first of the majors to approach bankruptcy, I would hedge bets that if Mandrakesoft was the first approaching Bankruptcy, then IBM may well have bailed out Mandrake and SUSE would be declaring Chapter 11 style bankruptcy now. It was IBM's interest that Linux appear strong Or it could be a strategic decision by IBM to back SuSe since Germany does not have a Chapter 11-type regulation while France does... [snip] Guy Bormann
Re: [Cooker] 650 or 700 MB
650 MB, 700 MB, 680 MB? ^^^ ^^ Well, depends on what is meant. Thanks to marketing (It looks like we offer just a little bit more...) and serial comm there is confusion over the meaning of MB (in the context of data storage, not transfer!) : 650 MB = 650*1024^2 = 650*1024^2/1000^2 = 680 MB So, many users actually mean 650MB when they talk about 680 MB. Others mean extended CD-ROM format which is again pushing limits and which can again lead to problems on older drives. So, I would limit the choice to 650MB or 700MB... Just to set a measuring standard... Guy
Re: [Cooker] Software naming (was: Please don't make urpmi stupid)
On Wed, 8 Jan 2003 [EMAIL PROTECTED] wrote: hmm. must be from the days when you had to enter special codes for each key stroke. Please reply BELOW the message! Quoting Steve Fox [EMAIL PROTECTED]: On Thu, 2003-01-09 at 01:19, [EMAIL PROTECTED] wrote: the name urpmi does sound stupid, like somebody belching. why do linux programmers give their software such silly names? guess they dont have sales or packaging departments Because it's fewer keystrokes! :) GNOME 2.0 went and renamed all their tools with more sane names, and so did the new Red Hat 8 configuration tools. gcalc = gnome-calculator gfontsel = gnome-font-viewer redhat-config-samba redhat-config-httpd So there is still hope! Not really, it defeats the readline completion feature... I.e. Do you really want me to show all gazillion possibilities... :-) [snip] [repeat:] hmm. must be from the days when you had to enter special codes for each key stroke. No, being faithful to heritage and Unix philosophy... (which doesn't mean it is Old and Crusty as the Beast wants you to believe.) It stems from the day Unix could fit in an 20k core and 12 floppies. Programmers are lazy/louzy typists...obviously Thompson and Ritchie too :-) The real reason is probably the tool-philosophy of Unix for which you don't want long filenames when piping commands together in a string of commands. Of course, it does not directly apply to single end-applications... Anyway, if you really want to tie design to marketing (we know what that leads to) you better focus on the GUI these days ;- Open source should stay with the Genetic Algorithm approach : slow but steady, untrendy and total coverage. Sorry, is there a Mandrake-advocacy list? :-) Guy Bormann
Re: [Cooker] [OT] Re: [CHRPM] rpm-4.0.4-27mdk
On 9 Jan 2003, Chmouel Boudjnah wrote: Guillaume Rousse [EMAIL PROTECTED] writes: Is that some sort of 'my changelog is strongers than yours' challenge ? not really we don't really compete (or if we do i will kick them all). And pollute it with nonsense... Guy Bormann
Re: [Cooker] Re: [CHRPM] gstreamer-0.5.1-1mdk
On Tue, 7 Jan 2003, Charles A Edwards wrote: On Tue, 7 Jan 2003 13:18:29 -0800 Todd Lyons [EMAIL PROTECTED] wrote: [snip] [root@localhost gstreamer-plugins]# strace rpm -Uvh Someone might want to try strace -f rpm -Uvh relevant-gstreamer-package instead to follow this fork(): [snip] fork() = 27099 wait4(27099, The parent is obviously waiting for a child that never dies. Guy Bormann
Re: [Cooker] Bug in gcc-3.2.1 .src.rpm build
On 7 Jan 2003, Austin Acton wrote: $ rpm -bb gcc.spec worked perfectly on my dual Athlon 1900 $ rpmbuild -bb --target=athlonxp gcc.spec failed to even get underway, but that's because I have no athlonxp-mandrake-linux-gnu target on my machine Are you 100% sure your rpm/BUILD and rpm/tmp directories were clean? AFAICR, he is not using a full Cooker system. (I could be mixing him up with someone else, though. If that's the case : sorry, I'll shut up :-) Guy Bormann
Re: [Cooker] intop (ntop-2.1.2-1mdk) doesn't start
On 3 Dec 2002, Gilles Mocellin wrote: [root@guitare root]# intop -i eth1 -- intop 0.0.1 (Sep 12 2002) -- The first interactive ntop program. (C) Copyright 2000 L. Deri and R. Carbone. All rights reserved. It allows you to control the power of ntop using fingers rather than ^^^ mouse! [snip] Initializing intop engine Please wait. Segmentation fault (core dumped) Maybe it doesn't like to be touched by fingers. Try the mouse instead :-) [root@guitare root]# Sorry, couldn't resist :-) Guy Bormann
Re: [Cooker] WTF?
[snip] Because I do not trust tinydns to do the job. I know a guy that has been working several years with the dot se top domain..., and I do take his word for it... Ok... you don't trust tinydns to do the job. Fair enough. Can I ask why? I had the whole icq chat in my history file that was lost after a session with ez-drive ;) So..., I can't remember specifically where the problems lies. I can ask him again if you like? Mmm, Oden, that's double standards when you compare to ... this: [snip] As I'm not in the position to tell if bind does the job worse than whatever else name server software I can't really say. I do have to trust that the de facto standard name server software works. If it didn't work you would surely be notified from a bunch of angry customers. Switching to djbdns is not an option for me in the near future I'm afraid. I know the ISC support sucks, but what can you do about it? People do trust companies like Microsoft, so... ;) Mmm, as long as it is free as in free beer again? After the initial rise in temperature I see the tragic irony in this one, so leave the fire extinguisher on the wall. But continuation of this line of thinking would have kept us in the Dark Ages of Computing... However, I'm afraid, Vincent, the battle has been lost already now that ISC has corrupted the Internet decision making process. (That is, apart from a sudden increase in user base.) Only a major security crisis(*) in the root servers due to BIND screw-ups could potentially revert this. Even the big players who pay for advanced warning will be bitten one day because redrawing into secrecy will make the BIND programmers even more lazy/lousy... Guy (*)One that comes out in the open, that is. A really smart hacker would leave services intact after the initial break-in. Since named crashes once in a while anyway, nothing but excellent an NIDS can distinguish an attack from a genuine crash. It is tempting to believe that there are already a bunch of servers compromised and silently serve as cloaks for other attacks.
Re: [Cooker] Mandrake Made Easy: a modest suggestion for Windowsusers[linux , what sucks ?]
On Thu, 14 Nov 2002, Faraj Meir wrote: Somoen mentinned a tool called tranfugdrake, which do this. After all, it could be a great thing no ? At least for some basics things, such a internet connection. yep but not only ... generaly anything is better that nothing a tool like this will be great even it's uncomplete... even if you say there is remplacement on windows, they will say it sucks. Everybody use winzip, and it sucks. no , I doesn't like at all winzip (on windows I use windows commander) . it's not true each version of mandrake I go to it test each beta in hope to del the windows partition , I and a lot of other would be glad to do that but really linux is not simple as windows and not yet for daily use . On other thing , you can use flash without paying 350 euros to Macromedia, I mean, if you are honest. But most people are not honest. Sad but true. yep crack it ;-) or make your company buy it for you . So, what is the point of moving to Linux then, if Windows is so wonderful and Linux really sucks? The free as in free beer (which seems to be the case already with cracked stuff)? Trading in the uncomfty feeling of being a factual thief with the uncomfty feeling that you can tweak your machine beyond NextNextNextFinish? My advice is to either make a choice and stop whining in the wrong place or to profit from both worlds and pay $$ and/or a learning curve. ...like anything in life... Linux is like evolution, if you are lucky, what you need emerges, if not too bad, nobody forces you to use it. On the other hand, if you are a breeder(i.e. tweaker) or gene technologist(i.e. programmer), you create your own species or sponsor people to do so. You can curse all you want, it won't help. Relax and you will see the beauty that's already there or stay on your own cosy planet if you don't like to space travel. It is not our fault They don't cater to our planet. Guy
[Cooker] Choice! (was Re: [CHRPM] kdelibs-3.1-0.beta2.11mdk)
On Tue, 29 Oct 2002, Mario Vazquez wrote: You are right, a poll would be nice. No matter which one Mandrake will choose at the end, it must be (1) the one the most users want, and (2) apply the same to gnome to have a little uniformity between desktops. I still don't understand why uniformity between different desktop environments is required. Some people mostly use KDE, others mostly GNOME (, yet others anything else). Why eliminate differentiation and choice for the ones who like switching all the time? (If you only switch once in a while you should be able to sustain this apparent disorientating blow of difference...) I use GNOME because I don't like the smooth uniformity (and slow bloathedness but that changed lately I heard) of KDE. Please, tweak as much as possible in both environments but keep the difference. Otherwise, there is no point in wasting resources to maintain two environments when their look and feel is the same. Linux is about choice! And I'ld like to make that choice for myself, not by the majority of users, usability experts or bean counters. Individual users should not have to undo integration efforts. If it is a matter of support in an organisation, then integration is the job of the support department or a third party (not necessarily the distributor). But yeah, I realise that also companies like free as in free beer so everything should be uniform, integrated, grayed out, uninteresting and dumbed down at install time to keep support costs to the minimum. Package software as is configuration-wise, keep on providing excellent configuration tools and pre-package optional configurations separately : raw_pkg-ver.relmdk: minimal tweaks drone_station_config_pkg-... : fully unified, integrated, drone station configuration package ergonomic_config_pkg-... : ergonomic AND esthetic configuration package avantgarde_config_pkg-... : fully themed and right out ugly configuration package :-) anything_else_that_sells_or_is_requested_or_contributed_config_pkg-... : ... Also configurations can have dependencies, just like libraries. And since most configuration in a Unix-like system is through separate text files, this should not add too much space on the mirrors, CD's, (Anyway, the installer already makes a distinction between configurations, but mostly on the raw install level...) And before you argue who is going to have to do all this additional work, just think about it first... (and yes, I do realise that an awful lot of configuration is hard-coded, i.e. at compile time so this does not apply to that case. On the other hand, I also realise that with tools like libtool, alot of modern software is heavily modularised, making separation of configuration from raw install possible.) Just my opinion, Guy Bormann
Re: [Cooker] Re: [CHRPM] kdelibs-3.1-0.beta2.11mdk
On Tue, 29 Oct 2002, David Walser wrote: --- Adam Williamson [EMAIL PROTECTED] wrote: On Tue, 2002-10-29 at 21:30, David Walser wrote: [snip] Thank you for the second reminder that Gnome sucks. Keyboard crap for copy/paste == Ctrl-C/Ctrl-V, in That is completely a function of what you are doing the most frequent : typing or clicking. When typing a text or when programming I use the keyboard. It is much faster in this case (i.e. no aiming required). When cross-pasting program output f.i. from other windows, high-lighting and middle-clicking is much faster. Unless you click your day away, this is no basis to call GNOME crap for this. (You might have other GOOD arguments, though :-) Linux you just highlight and middle click. And yes ^^ what has Linux got to do with it??!!! You meant KDE? there's one more button press to initiate the selection process with single-click, that's a *hell* of a lot better that *lots* more mouse crap (aka double-clicks). I don't suffer from RSI and I do like double clicks because it gives you a chance to abort an action. When you quickly do very diverse tasks on your computer, it is much easier to make (or about to make) a mistake. If you are doing the same things over and over (and for some reason these can not be scripted), it becomes automatic and you want your actions to be minimal (i.e. single click). [snip] Yes, I understand you stick to Gnome because you like the Windows interface. BULLSHIT!! (at least in my case) The way the window manager is configured here does make the interface way different from Windows (i.e. auto-raise on focus, window scaling, virtual desktops, ...) What have basic GUI actions to do with one articular family of implementations? One may like the Windows GUI for its double clicks. But this does not automatically mean the opposite...a common error against logics. Some of us like to move on to better things. A patronising phrase based on erronous logic, that is... Anyway, one cannot argue about taste. Calling these things better is pointless. It is a matter of preference. Guy Bormann
Re: [Cooker] Re: [CHRPM] kdelibs-3.1-0.beta2.11mdk
On 30 Oct 2002, Adam Williamson wrote: On Wed, 2002-10-30 at 12:13, Guy.Bormann wrote: Linux you just highlight and middle click. And yes ^^ what has Linux got to do with it??!!! You meant KDE? No, actually, I think he was right. From my understanding, the concept of middle-click pasting in *nix OSes predates KDE. Anyone know for sure? No, this is X behaviour (and gpm on Linux) in combination with three-button mice, common hardware until Microsoft started pushing two-button mice (i.e. one more than one-button Mac mice, We have more!). It has nothing to do with Linux per se. This predates both KDE and Linux. Guy PS: There was a transition period in the late eighties/early nineties of last century when some mice (like Genius) carried a switch to change between PC-AT three-button and MS two-button mode (disabling the middle button (and protocol)). And for some time three-button mice where very hard to come by. Some software on DOS used the middle button, but Microsoft didnot use (read : require) it for Windows. Although it could be used, the middle button only made a big comeback with the advent of the scroll wheel.
Re: [Cooker] Re: [CHRPM] kdelibs-3.1-0.beta2.11mdk
[snip] This goes back to Warly's question of what needs to be improved in the development process, and openness could be even better. It seems some changes are just mandated to the developers we interact with from somewhere above, and they just seem to come from out of nowhere, with no discussion. Marketing? What a surprise! Not fun for us but not unusual for a company... Well, why not. No money, no Mandrake. Guy
Re: [Cooker] Re: [CHRPM] kdelibs-3.1-0.beta2.11mdk
[snip] why hasten it? Won't happen if it didn't in the past 15 yrs. If you keep the button near threshold you don't need much finger movement, just the last bit of activation energy. And you move the mouse using your elbow, pivoting on the lower arm muscle cushion (on the table side) instead of using solely the wrist while crushing the carpal tunnels. [snip] It's my experience failed attempted double clicks cause more mistakes. Again, you must be clicking like crazy the whole day long. Personally, I hate the occasional slowdown caused by wrongly activating a process/widget because of a single misclick. I guess I should sum up that the whole point of this thread is that, yes, though single-click may not be what most people are used to, they *can* adapt to it, and they'll be happier about it in the long run. Never will be! Otherwise maybe they'll change it, but don't doubt people's ability to adapt, before 1991, nobody in the world (almost) could double-click. I don't doubt the adaption ability (I was surprised to find out how easy hand-eye coordination sets in with a tablet). You exagerate. Fewer overall computer users means fewer double-clicks :-) (IIRC, there was a program called view (evolving later in QEMM DesqView) that came with DRDOS. It required double-clicking on an icon to activate it.) Guy PS: I leave this thread. You can still mail me privately if you want...
Re: [Cooker] Why ext3fs is a default fs, not ReiserFS?
[snip] ReiserFS people have an excellent vision of next-gen plugin-based filesystem, but it turns out it's not quite ready yet. When (if) it becomes ready, however, it will replace ext?fs without doubt (my doubt, anyway ;-) ). Mmm, that's what they keep telling about translators on the GNU/Hurd mailinglists and pages :-)) Wondering when it will become usable enough... Guy
Re: [Cooker] Marcos colome
On Thu, 3 Oct 2002, Todd Lyons wrote: marcos colome wrote on Thu, Oct 03, 2002 at 03:10:48PM -0700 : Please remove my name from your mailing list Handled offlist. I begin to suspect that this is some autoreply from an active Yahoo spamfilter since it is always exactly the same sentence by this guy. He also does post once in a while a very stereotypical message (the same message??) about how Linux is heading for World domination. But then he NEVER asks to be removed. Also pay attention to the wording 'your mailinglist' (typical choice for a message in your name) instead of 'this mailinglist' (typical message by oneself)... So, the best we can do is just ignore Marco unless he comes with a really good bug report :-) Just an idea, Guy Blue skies...Todd P.S.: Concerning my own good bug reports :-) : I plan to buy a boxed set of 9.0 and 'upgrade' to Cooker...
Re: [Cooker] 9.0 - Voodoo 3 - No devices detected
On 1 Oct 2002, Wes Kurdziolek wrote: No, the NV03 is the RIVA128 chipset. The NV04 is the chipset for the TNT, the TNT2 (a sped-up version of the TNT w/ support for more RAM), and the Vanta (cheap-ass version of TNT). NVidia started marketing the Ok, thanks for the details. So, the NV04 is newer then the NV03 and that one didn't need an IRQ#... Guy
Re: [Cooker] 9.0 - Voodoo 3 - No devices detected
The Diamond Viper V550 is a TNT card, not a RIVA128 card, and there's no such thing as a TNT128 chipset despite the fact that the TNT is a 128-bit chipset. Mmm, how do you explain the following lspci -v output then? See attachment I don't believe I've ever come across an AGP card or a PCI card with a chipset that supports AGP (the TNT chipset for instance) that requires an IRQ. So... Guy 01:00.0 VGA compatible controller: nVidia Corporation Riva TnT 128 [NV04] (rev 04) (prog-if 00 [VGA]) Subsystem: Diamond Multimedia Systems Viper V550 Flags: bus master, 66Mhz, medium devsel, latency 64 Memory at de00 (32-bit, non-prefetchable) [size=16M] Memory at af00 (32-bit, prefetchable) [size=16M] Expansion ROM at unassigned [disabled] [size=64K] Capabilities: [60] Power Management version 1 Capabilities: [44] AGP version 1.0
Re: [Cooker] 9.0 - Voodoo 3 - No devices detected
On Tue, 1 Oct 2002, Ron Stodden wrote: Guy.Bormann wrote: [snip] 9.0 is unusable on this machine, 8.2 is fully functional. Am I not to complain? Until you show us /var/log/XFree86.some number, likely 0.log, lspci -v, /var/log/dmesg and /proc/iomem?...Nope, you can not! (Oh, BTW, how does /etc/X11/XF86Config-4 look like???) If you provide us this, we can find out if it is you or Mdk9.0... Guy (yes, me) PS: It is possible Mandrake did not include the Glide3 stuff that is needed for Voodoo cards... Adding the BusID line fixed the problem. However, if a BusID line is necessary with Voodoo3 (Generic) the installer should have generated it into XFConfig-4, surely? Well, it is not the Voodoo3 per sé that needs the BusID but those few machines that have multiple primary PCI devices the X init routines find (by whatever criterion it uses). Why the installer can make the driver stick to PCI:1:0:0 all by itself I don't know but the XF86Config-4 file is probably not generated through the same process. If it is an X tool, then you better take this issue to the XFree86 people... Guy, thanks for your help. Never mind... I am relieved that the fault was not mine! Well, that is a matter of opinion in this case. I can not put the blame on Mandrake as I'm too much in the RTFM-camp (a.k.a. know your system; of course for that you need a working system). On the other hand, I cannot blame the user because I understand many users seem to think they actually only have a working system when it has a GUI. And where does one start looking for the clues? (The Net of course using lynx :- ) However, calling 9.0 Mdk useless (instead of not really up to snuff) you went a bridge too far in my opinion, especially when you realize what effort goes in a Linux distribution aimed at a moving (hardware and userbase) target. Look what happens (see : Debian) when you try to iron out all actual and potential (your case) issues. You will always be left behind... Guy PS: I may be giving redundant answers by now but for some reason I answer the messages in the order I read them...
Re: [Cooker] devfsd.conf
On Mon, 30 Sep 2002, Biagio Lucini wrote: On Mon, 30 Sep 2002 [EMAIL PROTECTED] wrote: Now that 9.0 is out ... Minor spelling corrections in English version. Initialize not initialise Minimize not minimise Maximize not maximise Did you say English or American? :-) Well, according to Inspector Morse, not using the z-spelling shows you're not educated (at Oxford :-). I guess he's no authority on spelling though :-)) Biagio Guy
Re: [Cooker] Broken psaux sndconfig (cs4235)
I checked the CD that came with the sound card. There is no DOS software to be found on it. Well, no problem. I think the mouse is OK. However, you compared apples with oranges by doing the tests on two different platforms (i.e. different motherboards from different manufacturers) and the tests were not exhaustive (incl. swapping hardware between computers to rule out card problems). But see below... (Anyway, I think we should take this off-list...) [snip] I see the numbers, but they really have no meaning to me. Well, you provided me with the number anyway... [snip] without taking devfs into the equation of which I haven't heard until recently on this list. I don't see anything in this output that tells me anything. I'm not a programmer. Duh, pretty obvious, so why don't you pass on the output to me??? :-) Well, don't need it anymore, I think. [snip] new /etc/modules.conf. I put in the sound lines from the RH7.3 manually. That produced working CD player and KMedia, but non-working volume control in KMixer and CD player. I changed to the sound lines from RC3, but that produced no changes. Back to RC3 sound/psaux diagnosis. That's because you probably need a separate mixer device or maybe the OSS compatibility devices, don't know; ask the sound experts... NOTE ADDED : read SoundPro in kernel-doc-version/sound; it talks about mixer control problems for certain Windows Sound System cards (like your cs4232 card) and advises to try out the ad1848-like mixer chip driver (as included in the last modules.conf you included in this message). This reminds me to next time also provide the output of lsmod... The whole of RH7.3 /etc/modules.conf is: [snip] alias scsi_hostadapter sym53c8xx alias eth0 8139too Could be useful to enforce ordering maybe in the IRQ assignment by the kernel... [snip : see below] In strong contrast, /etc/modules.conf from RC3 contains: [snip] options cs4232 io=0x530 irq=5 dma=0 dma2=0 mpuio=0x330 mpuirq=9 !!!Only do this when the card is not detected by isapnp during boot. !!!However, you might want to add the mpu settings to the automatic line !!!if you need a MIDI synthesizer (and see below for yet another sound !!!device, the wavefront synthesizer...). Again, read !!!kernel-doc/sound/SoundPro cat /proc/interrupts from RH7.3: CPU0 3: 30 XT-PIC eth0 -- happy 5: 3151 XT-PIC Crystal audio controller -- happy [snip] 10: 12091 XT-PIC sym53c8xx !!!-- NOT happy!!! see below 11: 0 XT-PIC usb-uhci -- happy 12:488 XT-PIC PS/2 Mouse -- very happy :-) [snip] cat /proc/interrupts from RC3, without sound in /etc/modules: CPU0 [snip] 3:123 XT-PIC usb-uhci, eth0 -- not so happy :-) 10: 76 XT-PIC sym53c8xx !!!NOT happy!!! [snip happy rest] Note working mouse, but absent sound, and free IRQ 11 Because according to your previous post you assigned IRQ 11 in the BIOS to the SuperVGA card in slot 3. However, this type of video card doesn't need an IRQ since it is a completely memory-mapped device. So, turn of the BIOS IRQ assignment to VGA and additionally switch off the IRQ 11 allocation to slot 3 (i.e. by putting it back to Auto; in fact, you should do that for all the slots!!!). See below... [snip] Differences between machines: [A world of difference, but from this I can infer that the Tyan is giving you the hasles] [snip] # of SCSI HD's1 2 Boot hda sdb [snip] That's all the differences. Same include: ET6100, sym53c875 PCI slot 1, single IDE HD, CS4235 ISA slot 1, ALN-325 (Realtek 8139) PCI slot 2, SCSI CD CD-RW. I tried changing BIOS settings to stop the usb/eth0 sharing on IRQ 3, but nothing worked. IRQ 11 stays unused. No wonder, see above!! Next I tried using the different sound options from from the RH7.3 modules.conf in rc3. That produced another no-mouse result and the following cat /proc/interrupts: CPU0 [snip] 3: 36 XT-PIC eth0 -- happy 5: 0 XT-PIC usb-uhci -- happy 10: 68 XT-PIC sym53c8xx !!!NOT happy!!! 12: 0 XT-PIC Crystal audio controller !!!-- probably silently stealing IRQ 9 for automatically setting up !!!another audio device; I need lsmod and cat /proc/isapnp to be sure [snip happy rest] (Note sound on IRQ 12, and missing PS/2.) Mmm, I more and more start to believe the aliases for the other devices in modules.conf does influence the device driver probing or(/and) (so) the IRQ assignment order. and the following dmesg (from which I've stripped out most of what I believe is irrelevant): I hope so!! :-)) [snip] SCSI subsystem driver Revision: 1.00 PCI: Found IRQ 11 for device 00:08.0 IRQ routing conflict for 00:08.0, have irq 10, want irq 11 ?The MP3's don't happen to be on a SCSI disk, do they??? (See below!!)
Re: [Cooker] 9.0 - Mozilla 1.1 - [snip] 9.0 - Voodoo 3 -[snip]
On Sat, 28 Sep 2002, allen wrote: I have an AMD 1Ghz, about 300MB RAM, Voodoo3 AGP, about 20GB HDD IDE. ^ look Ron, get in touch with him... [snip : not failsafe though...] Guy (yep, me again)
Re: [Cooker] 9.0 - Voodoo 3 - No devices detected
[snip] 9.0 is unusable on this machine, 8.2 is fully functional. Am I not to complain? Until you show us /var/log/XFree86.some number, likely 0.log, lspci -v, /var/log/dmesg and /proc/iomem?...Nope, you can not! (Oh, BTW, how does /etc/X11/XF86Config-4 look like???) If you provide us this, we can find out if it is you or Mdk9.0... Guy (yes, me) PS: It is possible Mandrake did not include the Glide3 stuff that is needed for Voodoo cards...
Re: [Cooker] 9.0 - Voodoo 3 - No devices detected
The XF86Config-4 and /var/log/XFree86.0.log are attached. Both look good to me, yet no X appears. Not to me, since it finds more than one primary device so it doesn't know which one to associate the driver with. Either try setting the BusID (see man XF86Config for the format of the BusID statement; in PCI speak your Voodoo3 has ID 1:0:0 (see below)) or if you happen to have an onboard video chipset, disable it in the BIOS. So, first try adding BusID PCI:1:0:0 to the Device section of your XF86Config-4 file... [snip : calling a speckle of shit in the underpants a dunk heap; in this case, that might be true though :- ] XFree86 Version 4.2.1 / X Window System (protocol Version 11, revision 0, vendor release 6600) Release Date: 3 September 2002 If the server is older than 6-12 months, or if your card is newer than the above date, look for a newer version before reporting problems. (See http://www.XFree86.Org/) Build Operating System: Linux 2.4.18-23mdkenterprise i686 [ELF] Module Loader present [snip] (II) Bus -1 prefetchable memory range: [0] -1 0x - 0x (0x0) MX[B] (!!) More than one primary device found ^^ (--) PCI: (0:10:0) BrookTree unknown chipset (0x036e) rev 2, Mem 0xdb022000/12 (--) PCI: (0:11:0) Creative Labs unknown chipset (0x0002) rev 8, I/O 0xac00/5 (--) PCI: (1:0:0) 3dfx Interactive Voodoo3 rev 1, Mem 0xd400/25, 0xd800/25, I/O 0x9000/8 \ --- and this is the one you need (II) Addressable bus resource ranges are [0] -1 0x - 0x (0x0) MX[B] [1] -1 0x - 0x (0x1) IX[B] (II) OS-reported resource ranges: [snip] (II) Loading extension XFree86-DRI (II) LoadModule: tdfx (II) Loading /usr/X11R6/lib/modules/drivers/tdfx_drv.o (II) Module tdfx: vendor=The XFree86 Project compiled for 4.2.1, module version = 1.0.0 Module class: XFree86 Video Driver ABI class: XFree86 Video Driver, version 0.5 [snip] (II) TDFX: Driver for 3dfx Banshee/Voodoo3 chipsets: 3dfx Banshee, 3dfx Voodoo3, 3dfx Voodoo5 (II) Primary Device is: ^^^ (WW) TDFX: No matching Device section for instance (BusID PCI:1:0:0) found ^^ stands for Warning (see legend at the beginning of the file) That's why the BusID is needed... (EE) No devices detected. At least not unambiguously as required per the man page! Do you actually read the log completely before you judge it looking good? Since when it actually looks good, also your screen should look good, which it obviously doesn't and so in fact the log does NOT look good... (A.k.a. NEVER ignore Warnings (even not compiler warnings)) Hope this helps, Guy
Re: [Cooker] gnome integrated su
On Mon, 30 Sep 2002, Frederic Crozat wrote: On Sun, 29 Sep 2002 15:37:18 +, Sascha Noyes wrote: When clicking on the mcc icon on the top taskbar in the gnome environment, one is prompted for root password. But when exiting mcc, and starting it again from the same icon, one is not prompted for a password, this seems to be a pretty serious fault, no? Ok, i checked: this does not happen to any other application besides mcc. And it happens to mcc no matter whether i launch it from the panel or from the simplified menu. Root authentication is kept for 5 minutes.. And how is it kept???!! (Is it encrypted? In a file?) Guy
Re: [Cooker] 9.0 - Voodoo 3 - No devices detected
[snip : see other mail] cat /proc/interrupts shows no allocation to the video board (?). Shouldn't be a problem. AFAIK, graphics board are completely memory-mapped (through AGP, don't know about the Voodoo3, though) and don't need an PCI IRQ. This is not the problem. See my previous message on this subject... Guy
Re: [Cooker] 9.0 - Voodoo 3 - No devices detected
Typically one can assign an IRQ to the video card in the BIOS -- mine has an option named Video IRQ. Most video cards do not need them but work fine if one is assigned. Mmm, but they don't turn up in /proc/interrupts (and in this case it doesn't mean the device is not working) and it is a waste of a critical resource if you don't need it (sharing is a stopgap measure and reserving IRQ's through the BIOS makes life harder on the assignment algo). IRQ's are not really an abundant resource on the typical x86 platform... If you can prevent trouble, do it. Some, for one reason or another, require an IRQ and won't work without it (NVidia RIVA128 cards, for instance, tend to require one). Nope, I have a RIVA TNT128 (Diamond Viper V550) in my machine here at work. In fact, after Felix's problems I started analysing my BIOS settings (I inherited the machine with settings, no real tuning yet) and I disabled IRQ assignment to VGA. The card is obviously still working happily. That might change when I start playing videos or so, I don't know... Guy On Mon, 2002-09-30 at 12:28, Guy.Bormann wrote: [snip : see other mail] cat /proc/interrupts shows no allocation to the video board (?). Shouldn't be a problem. AFAIK, graphics board are completely memory-mapped (through AGP, don't know about the Voodoo3, though) and don't need an PCI IRQ. This is not the problem. See my previous message on this subject... Guy
Re: [Cooker] 9.0 install - no hdlist2.cz
On Thu, 26 Sep 2002, Levi Ramsey wrote: On Fri Sep 27 11:36 +1000, Ron Stodden wrote: I repeated the call just now: [ron@small ron]$ rsync rsync://ftp.uninett.no:873/Mandrake/ drwxr-xr-x4096 2001/11/04 07:36:22 . drwxr-xr-x4096 2002/09/26 03:29:03 Mandrake drwxr-xr-x4096 2002/04/16 00:11:05 Mandrake-devel drwxr-xr-x4096 2001/06/11 21:59:48 Mandrake-iso [ron@small ron]$ Again, there is no Mandrake-old on this primary mirror. Then switch your fscking mirror... Mmmh, but this sunet mirror tells him how beautiful and smart and experienced he is. How tragic, some find it hard to cope with the fact that the years have passed and that they missed an opportunity to leave a big impression on humanity and that their, no doubt deep, cathedral experience is worthless in todays bazaar. Admitted, it feels shitty to be surpassed in fame by an unexperienced young student in his 20's... Concerning the problem : Either something went wrong in your new rsync script (missed 12 rpms??), or wait for the fix if it is a genuine problem. It will be quicker than anything you experienced in commercial software (except for another load of support money)... Guy
Re: [Cooker] 9.0 rc3 problems with upgrade from 8.2
Why is install about 8 times longer than it should be? Dependency checks I guess, causing a roughly exponential increase in time instead of your linear extrapolation. In the pre-configured cases, the RPMS could be installed using --nodeps if that is not already done and fixing the dependency info in the DB for the hardware configuration- dependent packages. In the hand-picking mode and/or Expert case, however, dependency checking is still needed. Another contributing factor is the ever increasing size of type-characteric packages (i.e. development, productivity, desktop, ...). Guy
Re: [Cooker] 9.0 and next
[snip] 2. a trust metric system for bug reporters. (2) would be useful because it would give a fast indication of where the problems lie (I'm thinking of tying this into drakbug or something), and automatically filter out noise How would you handle aging of the trust metric? When there is aging, it relieves the problem of quality posters turning crappy due to lack of time, lost interest, ... On the other hand, it punishes quality posters that irregularly send consistently good reports. Guy
Re: [Cooker] Re: Nondeliverable mail
On Wed, 25 Sep 2002, Todd Lyons wrote: Marcel Pol wrote on Wed, Sep 25, 2002 at 09:17:26AM +0200 : Could it be hat he is subscribed with this adress? Just a blind guess, the name looks so familiar. janmaris at zonnet.nl A feeler email has been sent. Will let you know the results soon. I hope you don't get a slap on the wrist :o) Guy
Re: [Cooker] Broken psaux (was 9.0 final when ?)
[snip] 2-OS/2 doesn't find it necessary to reserve IRQ 15 when there is only one IDE device in the system and the secondary IDE controller is disabled to free IRQ 15 for something that wants or needs it. Fair enough. [snip] accomodate all extension cards. But on the other hand, only use IRQ sharing between non-sound card cards. This Award BIOS doesn't seem to like permissive PCI IRQ sharing, only Sharing is never a good idea anyway, but somehow it is even worse with a (CL) sound card present. [snip] There is one thing that DOS is perfect for on old hardware : using the setup floppy that should have come with your card. Usually there is a small port setup and test utility with which you can figure out the IRQ/IO combo or it tells you if the card needs jumpers set. There should be a leaflet explaining on how to turn also the card from automatic (f.i. through PNP) to manual setup. I never tried it the DOS software, and the ISA card is jumperless. No windoze here, and don't use DOS enough to bother making sound work in it. Sound card came only with a CD, which is a PITA in DOS. OS/2 installed the ISA sound drivers with no fuss or lockups. You obviously have NO idea what I'm talking about :-) Getting it WORKING under DOS is indeed a super PITA. But I was talking about little low-level test utilities one finds with many ISA cards. (Concerning CDROMs in DOS, I never found it a PITA unless I forgot to put MSCDEX.COM(?) on the boot floppy.) [snip] No cards have manual jumpers. Great! [snip : will look at the config tonight (CET zone)] OS/2 has never listed the Realtek 8139 in hardware manager. But it is present and it works?? Cheap Realtek chipset-based cards also don't like sharing. In fact, el cheapo cards usually don't support IRQ sharing. Adding to the trouble is that they sometimes don't fail completely so that they appear to work when you don't strain them. There are other devices listed, but none have IRQ or DMA allocations. I/O addresses are listed. If you want those too, let me know. If they work and don't have a IRQ, they are either pseudo-devices or polled through I/O ports. Don't need the data... I tried to find something to list resources in Linux using apropos, and had no luck. What do I use at the cli to list them? cat /proc/interrupts cat /proc/ioports lspci -v cat /var/log/dmesg (Just redirecting the output of the command 'dmesg' to a file only gives you the last few lines of the kernel hardware messages. I need the boot messages.) (Please, put in tar.gz for attachment if you think about sending them to the list.) Guy
Re: [Cooker] `death ears' (-:
On Tue, 24 Sep 2002, Leon Brooks wrote:
On Tue, 24 Sep 2002 20:50, Michael Holt wrote:
[snip : too much honor]
battery chickens ran on size D drycells (I
kid you not!).
Ok, I give up - what are battery chickens??
A battery of anything is a long row of them. So a D cell (or for that matter
A, AA, AAA and C as well) is not actually a battery, but the little square 9V
models and the ones in cars are. A battery mill is a bank of cam-driven
Not necessarily true! Although it is usually hard to see, A{A{A}} cells
sometimes clearly consist of a battery of small watch cell-like cells when
you look close enough to the casing. Or you could take off the casing to
take a closer look (see WARNING below!!!). In fact, I don't know of any
compound combination that produces an (electrochemical(*)) potential of
1.5V! 900mV is already quite a potential, needing 6 units in series to
reach 1.5V. Please enlighten me (off-list) on modern battery technology
and/or such compound combination.
Guy
P.S.: WARNING : you should know what you are doing and you need special
protection against physical and/or chemical hazards. Short-circuiting can
cause fire and most types of batteries contain poisonous organic
compounds, heavy metals and/or strong acids. Better keep in mind the
warning on the casing.
(*) or any other intrinsic potential gradient
Re: [Cooker] Just joined - Alan of the MandrakeSoft support teamsuggested i join
[snip] The ps/2 mouse problem bit me again this morning. Just a fropzen pointer. It happened again but I was watching the boot messages this time. Said /dev/psaux doesn't exist which is interesting as it does. It worked O.K. on the next boot. Mmm, also Felix has trouble with psaux. It's to soon to say they are related though... Here's some info cd /dev ls ps* -l | more lr-xr-xr-x1 root root 10 Sep 24 06:53 psaux - misc/psaux cd misc ls ps* -l | more crw-r-1 root root 10, 1 Sep 24 10:56 psaux vi psaux psaux is not a file Uhm, psaux is a character device. In principle, you can open it like an ordinary file but I guess vi stat's the file first to determine its type. Better to look at /proc/interrupts f.i. to see if interrupts are being caught. Not sure whether you need dmesg, but here it is [snip] IRQ 14, IDE0 IRQ 15, IDE1 [snip] IRQ 5, USB [snip] eth0: RealTek RTL-8029 found at 0xa800, IRQ 11, 00:E0:7D:76:E0:4E. IRQ 11, 10Mbit card? Mmm, can you show us lspci -v? eth1: ns83820 v0.18: DP83820 v1.2: 00:04:5a:72:2a:cc io=0xe500 irq=5 f=sg eth1: link now down. IRQ 5, 10M/100M/1Gbit card? Sharing with USB device! Show us ifconfig eth1 if you can get the link up... [snip] es1371: found es1371 rev 2 at io 0xa400 irq 10 es1371: features: joystick 0x0 ac97_codec: AC97 Audio codec, id: 0x8384:0x7608 (SigmaTel STAC9708) IRQ 10, Sound card... [snip] parport0: PC-style at 0x378 (0x778) [PCSPP,TRISTATE] parport0: irq 7 detected (IRQ 7, parallel port..., standard IRQ...) It probably all works but it is an ugly IRQ setup, at least according to my experience and taste :-) We need /proc/interrupts anyway... (make sure to move the mouse a bit after boot even if it doesn't move the cursor, we want to know if it is a hardware/driver problem or a software problem(gpm/X).) Anyway, if it worked before and stopped working recently, it is either : * the mouse or the PS/2 mouse port has become flaky * the kernel PS/2 driver changed * isapnp changed and ISA probing is screwing up the hardware even harder than it was notoriously able to do before (that's why probing should be avoided once the resources are known and isapnp.conf is updated accordingly). * the PS/2 mouse packet driver in X has changed and now contains a rarely tripped bug. (It is most likely adapted to the new XF86 4.0 architecture but it is rather hardware independent so most people using PS/2 mice should have experienced this bug if there is one.) * new hardware was recently inserted and it (or its driver) is interfering with the PS/2 mouse port. This breakdown counts for everybody experiencing (PS/2) mouse problems... Guy PS: Maybe in some critical cases, PS/2 doesn't like cooperating through devfs. That would be option 6...
Re: [Cooker] Just joined - Alan of the MandrakeSoft support teamsuggested i join
It worked O.K. on the next boot. Mmm, also Felix has trouble with psaux. It's to soon to say they are --too related though... Sorry, typo! I shut up for now... Guy
Re: [Cooker] Broken psaux (was 9.0 final when ?)
Still haven't learned how yet, so you get .zip attachment instead (shrunk only about 50%, total 4545 bytes). The mouse device is active as per the /proc/interrupts info. All cards seem to be detected and initialized alright (from dmesg). (I would try to get the NIC on IRQ 9, the sound card on IRQ 5(despite it apparently refusing to work there) and keep USB on 3 but that's just my obsession. :-) This does not exclude interference at a later stage. From the low interrupt activity I infer that you did the measurement right after boot. What does /proc/interrupts and dmesg output (not file!) give right after the mouse blocks (don't need the rest of the info)? (Only move it within a terminal window to keep keyboard focus and full exposure.) I should check the kernel change log to find out if something changed recently with regard to PS/2 but I doubt it. So either the mouse (port) hardware is flaky or the XF86 4.2 mouse packet driver is buggy (but why only for you). What does /var/log/XFree86.0.log show? To rule out other SW problems, you should strace -p PID of X server the X server to find out if it blocks and if so, if it is on the mouse device file. If it does, it really is the mouse. Of course, this is without taking devfs into the equation of which I haven't heard until recently on this list. It is hard to judge from your messages what you did right before or after sndconfig to determine that this really is the root cause... Happy hunting, Guy
Re: [Cooker] [OT] batteries
Not. I disassembled a fair few of them in my childhood, e.g. for carbon anodes for chemistry experiments. Yep, you should talk to Igor, he is sharing the same passion for batteries... :-) (As I told him, it could have been a reinforcement frame in a light casing...) In fact, I don't know of any compound combination that produces an (electrochemical(*)) potential of 1.5V! Carbon-Zinc does that nicely, and has done for centuries. Well, since Carbon and Zinc got formed in the Sun, I guess. (Yeah, yeah, actually, until Galvano or Volta put them togther in the 18th century...:-) (Anyway, I was confronted off-list with a gaping hole in my memory concerning redox and fysical chemistry and was pointed to a nice site on battery technology.) Off to bed now... (5am here and in 4hrs I have to get up, phhhewww Thanks! Guy
Re: [Cooker] cron.daily: rpm
On 20 Sep 2002, Adam Williamson wrote: On Fri, 2002-09-20 at 13:24, Götz Waschk wrote: Am Freitag, 20. September 2002, 13:58:31 Uhr MET, schrieb rcc: do we need to have rpm in cron.daily? Because like slocate it brings older computers almost to a halt. Stop whining. How long can a rpm -qa take? Not more than 10 Seconds on an old Pentium I guess. About 20 on this p2-400 with 4200rpm hard disk. Same order of magnitude... :-) Guy
Re: [Cooker] APIC support
On Fri, 20 Sep 2002, [iso-8859-15] Stéphane Teletchéa wrote: It seems my mobo is an APIC-only one. ACPI, you mean? Why doesn't it be included in the kernel ? (Too much work, unstable, ???) Bad, unconsistent implementation in BIOSes, although it is getting better, 'they say.' My old win98 can handle power off the cpu without any trouble. Yeah right, unless it is acceptable for the computer to hang in the power down process, making you wonder if the sync-to-disk has finished so that you can savely switch it off manually. And maybe there is a big switch-case statement in Win98 for all the motherboards around. Hmm, maybe that's where the performance drop comes from :-) Hey, we could mimick this to better reflect Windows behavior so that we don't scare those users from switching to Linux. We have performance in(?) spare anyway. BTW, my computer switches off quite nicely, I haven't tried sleeping or hybernation lately though... Is linux older ? Yes, in fact it is (foundation-wise) and not so full of half-working features, although it is changing (both ways). Stef, a bit disappointed. Well, that's the nature of the game I'm afraid. The question is how long do you stay disappointed with each OS. Guy, a bit less disappointed since he made The Switch (a long time ago).
Re: [Cooker] `death ears' (-:
[snip] ROFL! (-: `Larry Boy and the Ears of Death?' :-) It's `deaf ears' as in ears that do not hear. (-: Thanks for lightening my morning a bit :-) [snip : you only found out after getting married] battery chickens ran on size D drycells (I kid you not!). A few things like this helped to soften the blows of an otherwise traumatic experience. Yep, a sense of humor is essential... If a friend has the bandwidth, get them to fetch the SRPMs down for you. You can also try rpmfind.net as sometimes they have links to non-Mandrake RPMS Or try rpm.pbone.net!!! The downloads might be slow (when there's no local mirror) but the searches are lightning fast... Guy
Re: [Cooker] Broken psaux (was 9.0 final when ?)
First, I don't understand why OS/2 also assigns IRQ 15 to the sound card since IRQ 14 and 15 are reserved for the Primary and Secondary IDE channel resp. IRQ 3,4 and 5 can be used by either the serial port (internal and external) or ISA cards (check BIOS). Using them for PCI should only be done when there is NO ISA card in ANY slot. Keep them legacy otherwise and you have enough IRQ's 8 to accomodate all extension cards. But on the other hand, only use IRQ sharing between non-sound card cards. I thought I had it figured out about the time this post showed up. I had to move the forced NIC IRQ from 5 to 3, disable the #2 serial port, and set IRQ 5 to legacy. Apparently this sound card and Linux just won't play nice together if the sound card can't have IRQ 5, regardless of what setting you try to give it in modules.conf. Once running, I decided to actually try to do things. I logged in started Mozilla Messenger. That used to be typical for ISA Sound Blaster (or compatibles) cards. For older PCI cards (and motherboards), they show some affinity for IRQ 9 (not only in Linux). [snip] Next I tried changing io=0x530 to io=0x534 to match one of the OS/2 allocations. Repeat the app open steps above, except tried opening Messenger after starting the music, and got the hard lock immediately. Don't know what to try next. There is one thing that DOS is perfect for on old hardware : using the setup floppy that should have come with your card. Usually there is a small port setup and test utility with which you can figure out the IRQ/IO combo or it tells you if the card needs jumpers set. There should be a leaflet explaining on how to turn also the card from automatic (f.i. through PNP) to manual setup. Concerning PS/2, it usually is routed to IRQ 12 (check your BIOS, sometimes it gives you the option of freeing IRQ 12 up for regular PCI usage). ISA. Phooey! All I said is pretty general. In order to help, can you give me : ANY current BIOS setting related to IRQ's/IO ports and, if supported, binding to INT#'s, occupied serial ports, all extension cards (including order in slots) including assigned IRQ's after boot. Also, if you happen to have info on which card (ISA and PCI) have manual setting jumpers... Guy
Re: [Cooker] Re: [Contrib-Rpm] hot-babe-0.1.0-1mdk
On 19 Sep 2002, Adam Williamson wrote: On Thu, 2002-09-19 at 20:00, Guillaume Rousse wrote: [Contrib-RPM] Name: hot-babe Relocations: (not relocateable) [snip] Description : Hot-babe is a small graphical utility which display the system activity in a very special way. When the CPU is idle, it displays a dressed girl, and when the activity goes up, as the temperature increases, the girl begins to undress, to finish totally naked when the system activity reaches 100%. Of course, if you can be shocked by nudity, don't use it! I see absolutely *no* need to turn Mandrake into a porn distributor. Is it true after all what they say about British men? :-))) *jumping in my flamesuit* Guy
Re: [Cooker] Stackguard, FormatGuard for future Mandrake versions?
[snip : Stackguard fart] This should be taken up with the ISO C standardization committee so that boundary checking becomes an optional language feature (giving us a well thought out choice) instead of forcing a stopgap measure on everybody... [snip : to the point security reply by Ben] why not with a standard install have it use the guarded versioon, with the majority of new Mandrake users coming from a windows environment, they won't think there was a performance hit at all, since anything over a 1/100 rate will be faster than windows. [snip : ridiculous benchmark] Then it would be Goodbye Mandrake. I thought Linux was about CHOICE, why would I, as a scientific user who needs all the performance he can reasonably get, be forced to get substandard performance just to suite people who don't patch their machines? Cloning Windoze (in feature and in phenomenon or culture) is making Linux LESS attractive every day, not so far from now the only difference will be price (and that's not necessarily a Good ThingTM)... Jaqui Guy
Re: [Cooker] Cooker way to much...
On Wed, 18 Sep 2002, Todd Franklin wrote: I personally have almost 6000 unread emails in the cooker group alone, but i'm not upset about it! If anything it's like having a personal database for when i run accross a bug. Even if I don't have time to read them all, if I come accross a problem, i just run a search in my cooker folder and 9 times out of 10, there's a fix in there, or at the very least, somebody else with my problem. So I say, keep 'em coming, I'm loving every minute of it! My hats off to everybody for all the good work. Of course, you could use the archive for that :-)) On the other hand, you don't have to stay online for too long if you're still on 56K (in that case you have my sympathy :-)) Guy
Re: [Cooker] Minor visual glitches in Mandrake gtk2 theme
On 18 Sep 2002, Adam Williamson wrote: followup - after an urpmi.update (though nothing obviously related got installed) and a reboot, both my problems went away - scrollbar backgrounds now work, and the scrollbar block (come on, there HAS to be a proper name for this :) Index slider?? now has a minimum size so it always displays right. Guy
Re: [Cooker] OpenOffice affected by SetiAtHome
On Fri, 13 Sep 2002, Leon Brooks wrote: On Thu, 12 Sep 2002 21:40, Crispin Boylan wrote: This may or may not be fixable, but it seems that OpenOffice.org drastically slows down responsiveness when setiathome is running in the background, sure its easy enough to turn off but no other apps are affected! Can this be fixed? Yah, find us some aliens, make SAH either redundant or so important that it gets switched to SetiAtWellArmouredMilitaryBase. Hurr, hurrr, hurrr :-D Guy Bormann
Re: [Cooker] Cooker is frozen / 9.0 schedule
On Thu, 29 Aug 2002, Ron Stodden wrote: Levi Ramsey wrote: On Wed Aug 28 17:07 +0200, Guy.Bormann wrote: Why do you always smile the other way? :-))) He's Australian... it's kind of like how water drains in the opposite direction and they drive on the wrong (not the right...) side of the road... :o) And: The sun traverses the Northern sky. [which disorients anyone from the other hemisphere, since your innate sense of where north, south, east and west are by sunlight and shadows will all be 180 degrees out - Fun!] Mmm, me thinks it is not innate but by experience since otherwise most Australians would have died out already some time ago by getting lost in the desert :-) (Or never reached land in the first place...) (They didn't have GPS in those days, you know ;-) Guy P.S.: And anybody who tried to discover South-America f.i. would have ended sailing in circles :-)))
Re: [Cooker] Mailing lists dead
On Wed, 28 Aug 2002, Guy.Bormann wrote: On Tue, 27 Aug 2002, Jure Repinc wrote: David Walser wrote: Where'd everybody go? I usually wake up to over 50 new messages, but none? Even the archive shows none. A bunch of stuff was going on on the mirrors, but nothing on the changelog list either? What's going on? Why did a bunch of mplayer and php stuff disappear (contrib)? I think Mandrake mailing list server has quite some troubles. I also didn't get any mails from it and some messages I send to the list are lost forever. Some messages atke more then 10 minutes to appear on the list. I hope this message gets to the server and appears on the list. Very, very weird, I have never ever suffered from a lack of messages (quite to the contrary :-) or delays since people started complaining In fact, it seems I was struck by the last wave of lost messages after all... Never say never, Guy
Re: [Cooker] Cooker is frozen / 9.0 schedule
Yes, hush, don't tell rsync that it can possibly be doing this. (-: Why do you always smile the other way? :-))) Cheers; Leon Guy Bormann
Re: [Cooker] Off Topic
On Tue, 30 Jul 2002, Marcelo Gigirey wrote: http://techupdate.zdnet.com/techupdate/stories/main/0,14179,2871663,00.html A must read article Must read article? There's not even a decent comparison. Call me paranoid but this is a blatant plug bought for by Red Hat (Afficianados, Investors, ...?). Large market share, fat bank account, ... *tongue in cheek* Ffffinally a Linux company that is all-American and has success : Party Time!!! Hehe, finally rid of that Commie-tag, phewww! Gimme those dollars for a share! Oh well, NASDAQ can use a kick... I don't mind Red Hat being successfull but yes, indeed, very off topic on the ML of another distro trying to get ready...and a-waiste-of-time read. Guy
Re: [Cooker] mailcap requirisity (fwd)
Not that I needed it (as nothing goes wrong here without mutt :-) but thank you for the explanation. It's more authoritative than shouting Is! Is not!'s back and forth. I know MIME is for selecting view clients at the receiving end (which does require the right tagging at the sending end). I was focusing on the PDF format. Of course, I didn't realise that the wrong tagging is not supposed to damage the message payload. [snip; self-censorship on speculation of potential cause] Sorry Ben for tickling your challenge bone and triggering such a long but friendly response from you. I think this should do, Tibor (not suggesting you should listen to me ;-). Greets, Guy Bormann (PS: Some PDF files can only be read by xpdf, others only by gv and still others only by Acrobat Reader. They usually fail with exotic error messages... So much for Portability :-)
Re: [Cooker] BUG with gcc,glibc,clone
Hi,
clone() does get called but with -lpthread it segfaults (currently, only
main() threads core dump; this is being worked on by the LKD's).
(To see this try strace -f ./cl1 )
Anyway, like I asked you in a private message already why would you mix
the clone() system call with libpthread calls (or why else link
libpthread)? At the moment pthread_create is a wrapper for clone().
I guess Mandrake barfs at your cl.c because libsafe is automatically
linked. Playing with header files (adding memory.h) and allocation types
(i.e. valloc???, malloc, alloca, ...) I can get the clone process to run,
so it must be some stack/pointer issue. Normally you should be able to
allocate stack space on the heap but I guess (!!! not sure) libsafe
doesn't like this. Allocating stack space on the stack of the current
thread is obviously allowed since it can be monitored. See attachment.
Unless you are developping a new thread library, I advice you to read the
last part of the clone() man page if you are trying to invent your own
threading wheel...
I think this is not a problem with Mandrake so we should take this
off-list.
Greetings,
Guy
#include sched.h
#include signal.h
#include memory.h
#include stdio.h
#define STSZ (4*1024)
int pslave(void *data)
{
puts(slave runs );
return 0;
}
int main(int argc,char** argv)
{
char *stack;
stack = (char *)alloca(STSZ);
if (stack == NULL) {
printf (Not enough memory\n);
exit(1);
}
puts(about to clone...);
clone(pslave, (void *)(stack+STSZ-1), CLONE_VM|CLONE_FILES|SIGCHLD, NULL);
puts(clone ok);
wait(0);
puts(slave done);
/* free(stack);*/
return 0;
}
Re: [Cooker] BUG with gcc,glibc,clone
Oops, forgot : and you should always compile your programs with -DREENTRANT (or add #define REENTRANT), at least that's how it used to be, it is not in the man page (anymore?). Maybe, gcc is now configured to recognize -lpthread and automatically emit REENTRANT save code... Guy
Re: [Cooker] mailcap requirisity (fwd)
On Mon, 17 Jun 2002, Ben Reser wrote: On Tue, Jun 18, 2002 at 07:55:42AM +0200, Tibor Pittich wrote: this is BULLSHIT! pdf is binary format at all. there doesn't matter that containing graphics or only text. this is thing, that i testing about 2 weeks and i have big problem with sending correct pdf document (without graphics, only text!) to my business partner. at first, i think [snip] Try opening a text only pdf file with a text editor. It's text. If you see binary it's not a text only pdf file. Even still as long as all the *EERR* wrong answer! It's not because you can read a large part of a file that it is text/plain (i.e. pure ASCII TXT). PDF files can contain binary objects and streams. Even text-only files can contain binary bitmap font info, digital signatures, DRM info, logos (more often than not in business documents), print info, ... In fact, unlike embedded objects in PS, the binary data is in raw format, not MIME encoded (binhex'ed, base64'd, ...) like in an email. The text you see when reading a PDF is the PostScript framework on which a PDF is based. While you might have a text-only PDF, this is hardly the general case and since it is NOT pure ASCII, you need a correct MIME type in order to guarantee its integrity (especially on non-ASCII foreign systems). So Ben, either you didn't look deep enough in your example files or your files are the exception, rather than Tibor's. (See attachment, look for stream and what follows. Also check it with a hex-editor and you will see that it even contains NULL characters for example. Yes, at the end it contains a simple line drawing. I don't think all the binary is related to that, in fact, it can be easily handled by EPS.) However, I could be completely wrong. Anyway, I thought this list was to report problems. Let Mandrake decide what goes in the distribution and don't shoot the messengers (even when they sound ignorant to you) because you think they are stupid (an impression (not by me) obviously based on bad English language skills). The distribution is targeted at the desktop (i.e. client (f.i. mutt) fine tuning and easy, complete configuration), isn't it? This is not clear to me anymore, lately... (notice *tongue in cheek* for the humor-impaired) Greets, Guy Bormann PS: Sympa didn't like the message :-) Everybody interested can request a copy or download it from http://xtl.sourceforge.net/ under the header Documentation. I also have a scientific report written using Lyx, converted to PDF. It is absolutely text-only and still contains stream objects (I suspect they are actually bytecode snippets used for rendering or filtering or such... To be really sure one has to check the source for xpdf or gv.) Alas, it has interesting unpublished results so I can't distribute it freely. However, if you really don't trust me I can produce another document. Note that it contains formula`s. Some consider this not text and think it uses bitmaps but LaTeX renders them using fonts (potentially bitmaps too, but irrelevant now) which looks like text to me.
Re: [Cooker] Real world testing (was Re: is gcc 3.1 rebuilding done yet?)
Still no reason to whine! Especially *not* when newbies (i.e. apparently less-priviledged as opposed to self-appointed experts) get bashed for exactly the same type of message (what triggered their message doesn't matter). Do your Real World testing and clearly report the problems. I've said enough. It's the only polemic I started. I can't afford to sacrifice a second machine to Cookerize (so I can't really contribute much apart from reporting intrinsic problems) but that doesn't mean I cannot lurk on this list to get a feeling how Mdk NG will look like and decide if I want to buy it or not. I really want to spend some money to support Mandrake instead of downloading it but I want to be informed... So, nothing wrong with lurking. Guy
Re: [Cooker] php-4.2.1-3mdk.src.rpm + php-xmlrpc-4.2.1-1mdk.src.rpm
(The f*cked up A RR for d-srv.com has expired now..., well I did it to myself, so what can I say other than I'm a dickhead... :-)) You're complete utterly mindless idiot? :oP :- Guy
[Cooker] XFree86 4.x memory leakage
Fred, Are you still directly involved with XFree86 development? (Sorry list, I know he's on this list and I forgot his exact Mdk mail addr...) If so, is there an explanation(*) why the X server's virtual memory size constantly increases while the resident set size stays reasonably low? (This is ever since 4.0 but recently it got so bad that every so often, i.e. sometimes once in a week, X crashes because it cannot get vm_page's, i.e. X VM gets so big it eats up all swap while the RSS is only about 40MB. This seems to be correlated with heavy pixmap usage such as with Netscape webbrowsing or apps with heavy GUIs such as anjuta. In fact, crash is immanent as X size is now 157MB, RSS 41MB, swap 119MB/128MB and almost no X apps are running apart from GNOME system processes.) Sorry again, although also Cooker's 4.2.0-12mdk version is affected it is not a Cooker issue. It even happens on RedHat 7.1 + RawHide, although it is a bit aleaviated with 4.2.99 (I'm writing a driver for the SummaExpression tablet for which I wanted to use the newest XInput API). Are there plans for a 4.3 release soon? Anyway, we should take this off-list... Guy Bormann P.S.:(*)I have an idea (which needs more involvement from the kernel) but I have to check the X sources first...
Re: [Cooker] Re: Re: is gcc 3.1 rebuilding done yet?
So what does this mean in terms of the usablilty of the current Mozilla/Galeon in Cooker? Without Java (and Javascript?), it seems quite useless for real world usage, no? So what?!! _AGAIN_, Mandrake Cooker is _not_ supposed to be used for real world usage! How hard is this to understand? Jeezzz, why do people keep complaining about this? Is it premature to use gcc 3.1 perhaps? What if this issue is not resolved by the next release of Mandrake Linux? Will it just ship minus a gcc-3.1 compiled Mozilla and Galeon? See, above : The raison d'etre of Cooker is exactly that, namely to get an up to date distribution and resolving all the dependency/interoperability issues that comes with upgrading. If you want to use Cooker in real world usage you are completely on your own and no guarantees are given (and _none_ should be given) that it won't break in the meantime towards full release. I'm here only a few months just observing the mailing list, I'm not a maintainer, nor Mandrake expert, but this far I got! (I am lucky though as I am using a mixed system (Arrgg, don't do that! :-) but I have no problems or resolved simple ones by only getting the really essential parts. But hey, I'm not complaining and I won't ) Now go read it again! Guy
[Cooker] xf86cfg crash on initial module probing
Hi, xf86cfg of the XFree86-4.2.0-12mdk packages crashes at the initial Loading/Unloading of modules with the following error messages : ... Loading /usr/X11R6/lib/modules/drivers/vesa_drv.o Module vesa: vendor=The XFree86 Project compiled for 4.2.0, module version = 1.0.0 Unloading /usr/X11R6/lib/modules/drivers/vesa_drv.o Loading /usr/X11R6/lib/modules/drivers/vga_drv.o Module vga: vendor=The XFree86 Project compiled for 4.2.0, module version = 4.0.0 Unloading /usr/X11R6/lib/modules/drivers/vga_drv.o Loading /usr/X11R6/lib/modules/drivers/trident_41_drv.o LoadModule: Module trident_41 does not have a trident_41ModuleData data object. Unloading /usr/X11R6/lib/modules/drivers/trident_41_drv.o Failed to load module trident_41 (invalid module, 0) Loading /usr/X11R6/lib/modules/drivers/ati_old_drv.o Duplicate symbol R128Identify in /usr/X11R6/lib/modules/drivers/ati_old_drv.o Also defined in /usr/X11R6/lib/modules/drivers/ati_drv.o Fatal Error: Module load failure This machine has a Diamond Viper550 card and nothing fancy (no TV/Decoder/... card). If you cannot reproduce this crash, ask me more specific info on my system (I don't have the slightest idea what could be useful info). Guy Bormann
!!!OT!!: Re: [Cooker] Is LKCD worth including in kernel?
On Thu, 2 May 2002, Oden Eriksson wrote:
On Thursdayen den 2 May 2002 14.19, Borsenkow Andrej wrote:
What I really miss in Linux is ability to take and analyze system dump.
Anybody have tried LKCD project? Is it mature enough to be included in
released kernel? Is there any interest to have this in kernel at all?
Yes, I would probably find it interesting if it's not too difficult to
understand. Likewise I find the LTT stuff very interesting too, but there
were no interest as it seems...
There was a big discussion in the Linux Kernel newsgroup/list about this
issue about 2 weeks ago and it seems that only industry contributors
(i.e. telco's, IBM, ...) wanted to pursue this. The rest of the hackers
in the wild and Linus ('s crew) were not at all convinced how event
logging could be useful enough to justify :
1) a major overhaul of the printk/syslog messaging system
or 2) a second, parallel event logging capability
and the associated performance impact for the average user. Even
people who volunteered to implement a compilation configuration
construct to minimize the hassle for uninterested users were turned
down rather bluntly. (The next one judging in place of the user will be
pointed to the historical Gates blunder of the 640KB limit, on the Linux
Kernel list of course :-)
So, standard kernel support for industrial strength event logging
won't happen in the near future. However, there seems to be a strong
pressure--from companies moving their core control networks to Linux--on
Linux consultancy firms for event logging capabilities. I fear this could
lead to forking or a big turn-down for Linux.
Guy Bormann
Re: [Cooker] images/readme out of date?
On 24 Apr 2002, Liam R. E. Quin wrote: The images/readme.txt file on the ftp mirrors is maybe old? Is it a wrong version? Also, maybe it could give more explanation on how to use the network boot image? For many people that's a lot easier than burning CDs, and the net install seems to work really well, so it'd be nice to publicise it a bit more. At the moment MandrakeSoft is struggling to increase its revenue from boxed sets so promoting free downloads is the least of their worries ;-) Guy Bormann Liam
Re: [Cooker] Why is xmatrix missing from xscreensaver?
On Wed, 24 Apr 2002, Levi Ramsey wrote: On Wed Apr 24 23:51 +1000, Geoffrey Lee wrote: Probably, or probably not. But we can't risk our asses just because we *think* that it's ok to use xmatrix without some sort of licensing. I'm sure that was what I felt when I removed it. Exactly. If Mandrake gets sued by the entertainment industry (who spend huge amounts on lawyers) Mandrake will go under. Not could, but will. Well, instead of speculating, it's maybe time to actually consult a lawyer for this type of issues. This kind of discussion will occur more often as media companies tend to sit on their property, or what they think should be their property. Companies which do their legal homework will eventually get an edge over the ones that drastically avoid difficult issues out of fear to be sued to oblivion. This doesn't mean I don't believe Geoff made a wise decision (see below) but it should not be his decision in the first place. Some day, something will escape scrutiny and slip through. Anyway, IANAL(*) but AFAIK copyright is about copying (in terms of reproduction) the original, not about mimicking it as long as you don't pretend to represent the original. So xmatrix is actually a bad choice of name (but if the coders can show that they didn't steal the patterns from some video sequence and withdraw any direct reference to The Matrix motion picture, which makes it less attractive of course, it should be save). Guy Bormann PS: (*), i.e. visit one and don't base any decisions on my words. IANAL! :-)
[Cooker] Serious imlib bug in some 1.9.x versions
!!Sorry, template message ahead!!
-
Hello,
There is a potentially serious problem with (lib)imlib for versions up to
and including 1.9.13 that crept in probably around 1.9.9 or 1.9.10
(haven't checked but it was probably first introduced between image loader
module support and the disabling of helper image loader processes). I am not
familiar with the maintenance organisation of imlib so this message is
posted to the major Linux bleeding-edge distros (I checked RedHat RawHide1.0
imlib-1.9.13-3.7 source). I couldn´t query the Bugs DB, so I apologize if
it is already known (it is, it is fixed in 1.9.14, and probably all SuSe
releases).
I include a simple fix, not a patch since version 1.9.14 (ximian
release) has a complete cleanup of the helper process support and
recognized the problem. (It is a silent fix (i.e. no ChangeLog entry)
since it exposes big mouthing by the author/maintainer about this is
2002... ;-) (Bugzilla hates me, not able to check bug#'s.)
Symptoms: on my system (Mandrake 8.0 with major 8.2 Cooker upgrades):
anjuta complains after a while about too many open files for a single
source and doc file project.
(same thing could have happened to my RedHat 7.1 home system but I didn't
upgrade yet (==pacifier for distro-cross posting :-P )
Diagnostics:
lsof shows that it is constantly reopening theme image files without
closing them, pointing toward the gtk libpixmap (I use DarkMarble ;-)
theme engine (or any other engine using libgdk_image) and thus
pointing to libgdk_imlib
due to this, I also checked other gtk apps : gnomecc, gimp, gedit, ...
showed the same behaviour
Why did only anjuta get sick?
It has a new feature (as of v1.8) : whenever a open doc in the editor
changes due to C/P or typing, the color of the notebook widget page
label changes.
Anjuta does this by setting a theme color rc and issueing a gtk_rc_style
reload.
(Mmm, that's weird, I have imlib caching enabled which should generate
hits for every gtk-widget opened or gtk_rc_style reloaded, inter-appl.-wide,
instead it reopens files...Ok, that's something else for the imlib
maintainers...) Other apps. only happen to create widgets (and thus gtk
theme loads) once in a while (you need to open between 20-30 widgets to
hit a limit of 1024 for an average theme).
Prognosis: Ok, but potentially dangerous (imagine minor scattered changes
to 10 source files in anjuta before the autosaver kicking in (and changing
the label) when you only have 10 fd's left).
Treatment : either fix Cooker imlib-1.9.13-1mdk sources at around line 305
of gdk_image/load.c (i.e. the unconditional #if 0 should only exclude
the else {...} clause of the if (!need_conv || !data) {} to allow file
closing for non-EIM format files) or kick Cooker up to 1.9.14 of imlib.
Sorry for this hefty mail, but this way people can recognize the bug as it
is only apparent in gtk apps. with heavy image loading such as anjuta or
thumb nailers/animators.
Guy Bormann
SuSe imlib's don't seem affected (checked 1.9.9-100 and 1.9.10-1)
Checking Debian as of now...
-
Re: [Cooker] Serious imlib bug in some 1.9.x versions
On Fri, 19 Apr 2002, Frederic Crozat wrote: On Fri, 19 Apr 2002 15:08:47 +0200, Guy.Bormann wrote: !!Sorry, template message ahead!! - Hello, [snip] so this message is posted to the major Linux bleeding-edge distros (I checked RedHat RawHide1.0 imlib-1.9.13-3.7 source). I couldn´t query the Bugs DB, so I apologize if it is already known (it is, it is fixed in 1.9.14, and probably all SuSe releases). I include a simple fix, not a patch since version 1.9.14 (ximian release) has a complete cleanup of the helper process support and recognized the problem. (It is a silent fix (i.e. no ChangeLog entry) since it exposes big mouthing by the author/maintainer about this is 2002... ;-) (Bugzilla hates me, not able to check bug#'s.) I'm currently building imlib 1.9.14.. Thanks for screaming, I didn't saw the imlib 1.9.14 release :(( Never mind! In fact, 1.9.14 seems to be fresh as 1.9.14.ximian appeared on http://rpm.pbone.net about the same day 1.9.13-1mdk(Cooker) appeared on http://rpmfind.net !! (I'll be buying 8.2 for work next month...if I can get my hands on a copy here in Belgium.) Salut, Guy
