Re: kde volume control question
On Thursday 03 July 2008, Kent West wrote: How do I get the volume control in the systray to control the PCM control instead of whatever other control it is presumably controlling? Right click on the speaker .. see the menu ... Select Master Channel . and pick one.The little slider is one of the sliders in the mixer, and you get to pick which one. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: CAD software for PCB engineering and routing
On Wednesday 25 June 2008, Michelle Konzack wrote: Am 2008-06-23 22:31:12, schrieb David Goodenough: Have you looked a gEDA? Its available on Debian. Yes, but it is NOT intuitive as programs I have used in Enterprises. It is nearly imposible to make bigger projects with SIMPEL CPUs like ARM or MIPS. Have you looked at kicad? Supposedly, it is less powerful than PCB (geda) but more intuitive. My experience is that kicad is less intuitive, because it has a MS-windoze look and feel, with the extra baggage. I'm curious what you think about it. Again .. some friendly communication with the developers should help a lot. Making software intuitive is hard. It's mostly in how it behaves when you do something wrong. With gnucap, almost all of the bug reports ever received involve some kind of user error. The bugs are real bugs, and the reports are very much appreciated. That kind of bugs is very hard to find without users who are willing to try it, report the experience, and work with the developers to improve things. What I need is a PCB-Autorouter which do the routing for me!!! Someone is working on it. From what I see, he is making good progress. I want to tell the program the pinlayout of the used Microchips, the position of the parts and let the Computer work for me. This is HOW intuitive software should work, specialy for such simple tasks Autoroute is not a simple task. For a long time, all of the commercial PCB programs used the same outsourced autorouter. This lasted until the maker was acquired by Cadence, and became unfriendly to Cadence's competitors. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: CAD software for PCB engineering and routing
On Tuesday 17 June 2008, Michelle Konzack wrote: I am using *** but those programs are crap and you can not . Even my 18 years old MS-DOS software works better. So, my requirements are: 1) PCB-Layouts up to Extended ATX and 18x11 2) Only ARM and MIPS CPU's (yeah, no i386, sparc powerpc) 3) double sided 4) double layer 5) autorouter 6) design template per chip including auto placing of capacitors and such x) y) z) 997) Export/Import all common formats 998) Must definitivly work with Debian GNU/Linux 999) Not expensive! I have already VariCAD runing on a Dual Opteron 280 with 16 GByte of memory plus some 300 GByte SCA-Drives on a ICP Raid-5 and I do not need a second program of this caliber. Any recomandtions? Yes. Be more respectful of people who are trying to help you. You say not expensive indicating your willingness to pay something. How about paying for improvements you want? You say you are a system administrator, EE, and consultant. How about providing some services to a project you would like to see improved? You say are crap. How about some detailed info on what is wrong and how it can be made better? The projects you mention are starved for resources, as are many free software projects. The subtleties go a long way. In most significant free software projects, the authors are fully aware of the problems, but lack the resources to do much about it. You have been here a while, and certainly must know how free software works. How about taking on one of the problems yourself, solving it, and contributing the solution? Just one will go a long way. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: CAD software for PCB engineering and routing
On Wednesday 18 June 2008, Michelle Konzack wrote: Am 2008-06-18 13:24:06, schrieb al davis: Yes. Be more respectful of people who are trying to help you. I have gotten no reponse from Developers and others... Did you ask? You have been on this here list long enough to know what kind of response you will get. You say not expensive indicating your willingness to pay something. How about paying for improvements you want? I have payed arround 20.000 Euro for I High-Performance CAD program which support simulations in real-time... I used it to construct my Bio-Fuel-Raffinnerie... Even it is complex, it is intuitiv! I can not say electric or geda are intuitiv the use of the interface is horribel... You would be amazed at how far 20.000 Euro goes to support a free software project. It might be enough to pay for a complete redo of the user interface, to exactly what you want. You say you are a system administrator, EE, and consultant. How about providing some services to a project you would like to see improved? I am developer and have my own enterprise (IT), working currently arround 16 hours a day and I am in creation of a NEW Enterprise in Morocco (Bio- Fuel) which went hit by three bombs on 2008-01-01 because je deranger les intrets francaise... I have not the time to bother with only half working software... You must realize that all software was only half working at some time. For commercial software, investors keep it going past this stage. For free software, the investors are the developers themselves, and the users that are willing to put up with it, and carry it through. There's also the factor of which half, and where is it going. The zero-cost non-free software gets its funding from investors, and carefully chooses which half to make you buy something bigger. Another point on half working software is that sometimes there are multiple competing projects, each half working but a different half. If only they would get together.. I need a software FAST! Even if I would spend 2000 Euro to a OSS developer, he/she would not develop a solution in a timeframe I need. And since I have NO time, I already outsourcing jobs over http://www.getacoder.com/. I give a fsck on WHO is working for me because things must be done. I agree that there are times when we must put idealism aside to get something done. Some projects can only be done in a commercial environment, with big investors. I understand your decision here. You just might need to use Windoze here .. Pads, perhaps??? Even so, with wishful thinking ... a little participation in the free project goes a long way, if it is done constructively. I would really like to use your x, but because of (some detail) I feel forced to use y. It's so close, but. Some commercial products cooperate with Free/open-source software. Some try to undermine it. It is important to know the difference. The zero-cost crippled demos undermine it. There are big packages that use Free/open-source commodity packages and provide added value at a price. They support it, especially if they help fund it. Sorry if I sound overbearing here .. I am working on a project (very close to geda) which is now half working and realize how little things like this impact progress on a project. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Portage, RPM or DPKG: number of packaged software comparison
On Tuesday 17 June 2008, Sven Joachim wrote: A quick web search shows that Debian sid and Gentoo are roughly at par with 12000+ source packages¹ each. Both are outnumbered by FreeBSD ports, though; they have more than 18000 packages available². Gentoo and FreeBSD (and others) include some packages that do not meet Debian free requirements. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Portage, RPM or DPKG: number of packaged software comparison
On Tuesday 17 June 2008, Steve Lamb wrote: The program must include source code, and must allow distribution in source code as well as compiled form. but there is the non-free section, which includes some closed-source products that are proprietary in every way. I see license statements like Use is limited to non-profit applications or evaluation purposes.. . Make it clear that this is a Freeware Light edition of a commercial product, and as such may only be used for non-commercial or evaluation purposes. -- crippled demo ... -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: text file from Linux to windows.
On Thursday 29 May 2008, Miles Fidelman wrote: Actually, it dates back further than that, to ASR33 teletype machines, where you needed to issue separate carriage return and line feed characters to end a line - to i) physically return the carriage to the beginning of the line, and ii) feed a line of paper (turn the platten). (Anybody else out there old enough to remember when ASR33s where THE standard i/o device? :-) One reason for that was that it took extra time to do a carriage return. Having a non-printing character required after a CR made sure that the carriage had returned before printing another character. Remember .. those things had no buffer. The character decoding was completely mechanical. The electrical feed consisted of a current loop that operated an electromagnet in sync with the serial code. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: OT: C++ help
On Wednesday 21 May 2008, Jordi Gutiérrez Hermoso wrote:
The following code will not compile:
class foo{};
class A{
public:
void f(int a ){a++;};
private:
virtual void f(foo a) = 0;
};
class B : public A{
private:
virtual void f(foo a){};
};
int main(){
B b;
int a=0;
b.f(a);
}
The problem seems to be that all of my functions being named
f are somehow colliding with each other. It seems to me that
the call b.f(a) is unambiguosly pointing to A::f(int), but
gcc disagrees.
You might as well have written:
class A{
public:
void f(int a ){a++;};
};
class B : public A{
private:
virtual void f(foo a){};
};
f in B has the same name as f in A but different args.
This is usually a mistake, so they made it illegal.
If you override one f, you need to override them all.
(or change the name)
To fix:
class B : public A{
public:
void f(int a){A::f(a);}
private:
virtual void f(foo a){};
};
Now, try this:
class A{
private:
virtual void f(foo a)=0;
};
class B : public A {
private:
virtual void f(foo a){}
};
This doesn't work either, because f in A is private, so B
doesn't know it exists. I don't know what it will do. I think
you get 2 f's, and still can't make an instance of B because f
is pure virtual. Change the private in A to protected, so B
will see it.
It really is off topic
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: mailman install trouble
On Thursday 08 May 2008, NN_il_Confusionario wrote: * From: al davis [EMAIL PROTECTED] When I do sudo apt-get install mailman .. it gets as far as the language selection, and then loops. ideas? [naive idea] try changing the debconf interface: for example, from dialog to text or whatever That got it to install. I changed it from editor to readline. Although it installed, it still didn't work without lots of fiddling. The README files told about some of it, but still, I had to change some permissions in ways other than what was documented. I did enough fiddling that I am not sure what I did and what was really necessary. I don't know why to took so much fiddling. Maybe there is too much cruft on the machine. Maybe I missed something. Maybe I messed up something a while back, and fiddled to work around it. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
mailman install trouble
I am having trouble installing mailman. When I do sudo apt-get install mailman .. it gets as far as the language selection, and then loops. I enter en (or nothing) where it belongs, save it, exit the editor, or exit without saving, then it comes up with the same screen again. Repeat forever. (until I kill -9 ...) The system is rather old, with Etch, kept current. Everything else has been working fine for a long time. It is set up as a web and email server, with apache2, exim4, python 2.4.4, I checked archives, google, etc.. and I didn't see any reference to this problem. ideas? -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Commercial-grade application software
On Tuesday 15 June 2004 10:17 pm, jack kinnon wrote: That's true, there are a lot of possible mix-and-match. I am looking for a completed one that may suit my needs. Basically I have in mind an integrated environment like those available fr Borland or MS. Some of us think the MS or Borland approach is undesirable. Mix-and-match allows members of a development team to customize their personal environment, while maintainting consistency in the package being produced. Mix-and-match also allows the free software model to work. As someone else said, there is kdevelop and anjuta. Apt-get will find all the other stuff they need, and make it look integrated. That's actually what MS and Borland do. They just don't tell you. They make it difficult for us who don't like the GUI. I think either will meet your needs. Please don't make me use them. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: locale errors
On Thursday 25 March 2004 01:04 pm, Ashley Graham wrote: how do I fix this: perl: warning: Setting locale failed. perl: warning: Please check that your locale settings: LANGUAGE = (unset), LC_ALL = (unset), LC_CTYPE = en_US.UTF-8, LANG = (unset) are supported and installed on your system. perl: warning: Falling back to the standard locale (C) I can't remember what I did, but this error wasn't around the whole time. I think it started appearing after I installed something, but for the life of me, can't remember what. Log out, then log in again. I get this sometimes after an upgrade (apt-get dist-upgrade). dpkg-reconfigure locales alone doesn't do it. Logout, then in, then it's back to normal. If someone can explain why this happens, that would be nice. I am guessing that it has something to do with locale being updated, but some applications or dynamic libraries or something like that are already running with the old one, so it gets a mix and complains. As a more general hint .. I have found that often when things look messed up after an upgrade, logout and in fixes it. Most other systems require a reboot after an upgrade. If all else fails, try that. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Remove attachments from mbox
On Thursday 26 February 2004 01:39 pm, Shaun Jackman wrote: I want to remove all the attachments from an mbox. Can anyone suggest a good script or program for this purpose? Please cc me in your reply. awk. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: ext3 why?
On Tuesday 24 February 2004 11:02 pm, Jan Minar wrote: Once upon a time, Jan converted his ext2 partitions to ext3. His understanding of the journalling options was even weaker than it is today. Days passed, several crashes occurred. Jan didn't care. He thought the journal would. The Fsck Day came. One filesystem error message after another started appearing on the screen, revealing that the journal was not as journalling as Jan originally thought. Jan was mystified. A journalling file system is of no help when the cause of the crash is bad memory, bad disk, disk controller bugs. If your hardware is really bad, the best file system is a simple one with no buffering. You might consider FAT. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Debian Sarge MIMEDefang with Exim
On Tue, Feb 17, 2004 at 05:40:48PM +1000, Peter A. Cole wrote: Does anyone know if there are plans to produce a MIMEDefang package with dependencies on Exim rather than sendmail? It seems a little odd that Debian uses Exim by default, but the MIMEDefang package provided with Sarge depends on sendmail... On Tuesday 17 February 2004 05:09 am, Colin Watson wrote: File a bug report? The maintainer probably doesn't read this list, and he's best placed to know the answer to your question. Here's a better idea. Ask him why. Probably he doesn't consider it to be as bug. Maybe upstream requires sendmail, and doesn't work with exim. Ask him why. Probably he doesn't consider it to be a bug. Maybe he never thought of it. Actually, it doesn't seem odd to me at all. Debian has lots of packages, and often provides several alternatives for a job. There are many applications that rely on a particular set of other applications. Please don't file a bug report when something is working as advertised. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Rejecting viruses the Right Way[tm]
on Mon, Feb 09, 2004, Derrick 'dman' Hudson wrote: If a message is either rejected (during the SMTP dialog) or bounced (after accepting and queueing the message) then the same innocent third party receives some junk mail.[1] The difference is only in which server is sending the bounce message. On Friday 13 February 2004 01:18 am, Karsten M. Self wrote: Not so. Unfortunately, Karsten, you are wrong here. Few viral SMTP servers will generate and forward a bounce. SMTP servers holding an open connection with the originating MUA (or the virus itself) will pass the reject message to the originating client. Only misconfigured smarthosts will generate a spurious bounce. True, but I think misconfigured smarthosts are more common than not. Comcast is one such misconfigured smarthost. Consider this ... A customer gets internet service by cable. The technican configures the modem. Customer doesn't bother to get an email address in the provider's space. Instead, uses some other address and picks it up by POP or whatever. When sending mail, puts that address in From. Having nothing else to go by, that becomes the Return Path. That is what is considered to be proper operation. As a result, for improper action, anything goes for that address, so spammers and viruses can do what they want. This is another reason why it is bad to force the use of a smart host. They launder the mail, so it becomes impossible to find where it really came from. Perhaps this is the intent. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Rejecting viruses the Right Way[tm]
On Saturday 14 February 2004 04:39 pm, Paul Johnson wrote: On Sat, Feb 14, 2004 at 03:45:24PM -0500, Al Davis wrote: On Friday 13 February 2004 01:18 am, Karsten M. Self wrote: Not so. Unfortunately, Karsten, you are wrong here. Back up your statement, please. It is easy to find out if your ISP's relay works this way. Use a MUA that lets you configure to use a remote smarthost, bypassing your local MTA. As far as I know, all of the graphic ones let you do this. On Windows, it is the usual way to configure them. Compose a message, but change the From to where you want the bounce to go. Put a bogus address in To. Send it! As a result, for improper action, anything goes for that address, so spammers and viruses can do what they want. Not quite. The return-path only screws with broken virus scanners that aren't rejecting at SMTP-time, but making up a bounce message incorrectly assuming that the return-path is actually accurate. Paul: If you would like proof, let me know, and I will arrange to have comcast bounce you a message. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
EXIM configuration question
I am puzzled about how to configure this. It is probably one line but I missed it. I want to configure it so that it first attempts to deliver directly. Then if it fails delivers through the ISP's relay. Background ... It's easy to configure it to always deliver directly. That works except for certain sites (AOL) that block addresses that are listed in dynamic IP blocks. (well known problem) It's easy to configure it to always relay through my ISP's relay. It's easy to have certain listed destinations delivered through my ISP's relay, and those not listed delivered directly. I am using: Exim version 3.36 #1 built 11-Dec-2003 11:35:23 on sid. What I would like is for something to make the decision automatically without explicitly listing the problem sites. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Partitioning Scheme for New 120GB Hard Drive -- Comments?
On Friday 13 February 2004 02:49 pm, Karsten M. Self wrote: You've also left off a recovery partition. I keep a 256 MiB - 512 MiB partition on which a relatively minimal installation is kept. I would go farther than that. My preferred setup is to have enough space to completely install twice. That way if I decide to reinstall or make a major upgrade, the other one is there for recovery. Here's mine / 250 mb /usr 5 gig /tmp 250 mb /var 1 gig (cache fills up occasionally) /home /alt 250 mb /alt/usr 5 gig /alt/var 1 gig /alt/home The /alt stuff has helped many times. It was a way that I could try Debian without losing my old Mandrake system. It was a way that I could switch to unstable, while keeping stable in case of trouble. It was a way to keep a working system when the switch to gcc 3 and kde 3 left unstable very broken a few times. I used it as a recovery partition, when bad memory trashed the one I was using. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: apt-gey: I can't find anywhere this answer! :(
On Thursday 05 February 2004 01:10 pm, Johann Koenig wrote: 82roby [EMAIL PROTECTED] wrote: is there a way to see a list of installed packages, with the INSTALLATION DATE? No. but you can tell the installation date of a particular file: ls -lc -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: A letter for Mr. Darl McBride
On Sunday 01 February 2004 03:44 am, Nano Nano wrote: ... As employees of the store, we were allowed to check out one piece of software per day, and return it in the morning. A hairdryer was used to gently separate the sticker without breaking it and retaining its glue-iness. We had a shrink-wrap machine in the back. No one would know we'd made a copy for ourselves. You took. You gave nothing back. We were a small community, and as the witches say The Whole Of The Law Be An Ye Harm None, Do As Thou Wilt. Parasites. Then we used to trade MP3s on UseNet in 1997. I used to suck down 1 Gig per day and burn 2 CDs. My goal was to have 1000 CDs. No one cared. You took. You gave nothing back. Then Napster came along and ruined it for everybody but letting their 12 year old brother abuse the same power. Napster took. It gave nothing back. . Stay in the shadows, and there is nothing you cannot have. Stand in the light and be smited and lose. Smited by who? By those in power? By the people? Once upon a time there was slavery. In some parts of the world there still is. We have freedom only because of those willing to stand in the light. Once upon a time women couldn't vote in the U.S. In some parts of the world they still can't. The got the right to vote only because of those willing to stand in the light. I guess my point is we are a smart bunch of people and if we stay quiet we can have all we want and no one will bother us. That's all. I've seen bigger movements than this go down. The free software / open source movement isn't a movement that takes and gives nothing back. We share our own and give back in kind. We have no desire to take that which is not ours. The movement is growing only because of those who give back more than they take. What has SCO contributed? They take. They give nothing of value back. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: A letter for Mr. Darl McBride
On Sunday 01 February 2004 03:47 pm, [EMAIL PROTECTED] wrote: Ok I am sorry for all the rant, if I have personally insulted anyone I am sorry. When you use free software, you are giving something back when you spread the word, and when you redistribute as GPL says you should do. What bothered me is that Nano's parable seems to lead to the conclusion of shut up and be happy you have anything at all, and likened free software developers to people making illegal copies of proprietary works, in defense of SCO and their approach. If I misinterpreted it, please accept my apology. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Isolationism is history.
On Fri, Jan 30, 2004 at 11:45:49AM -0800, Nano Nano wrote: ... Microsoft does tend to use gotos for error exits. Nothing wrong with this. It is called throwing an exception. C++ and java have keywords try, catch, throw to make it official. On Fri, Jan 30, 2004 at 11:45:49AM -0800, Nano Nano wrote: You can probably read the Windows code by looking at the Windows CE.NET source which is freely available. It's a fork. You can get a flavor of it. If you try real hard, you can probably via a University take a look at the source -- Microsoft is handing out read-access grudgingly. Work within the system and you can accomplish that goal. On Friday 30 January 2004 04:11 pm, Colin Watson wrote: But be very careful about doing that; you may well end up tainted if you sign source licence agreements, and writing free software thereafter could be difficult. This is the original basis for the SCO vs. IBM lawsuit. Writing any software that is in any way similar thereafter could be different. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Isolationism is history.
On Friday 30 January 2004 07:58 pm, Al Davis wrote: On Friday 30 January 2004 04:11 pm, Colin Watson wrote: But be very careful about doing that; you may well end up tainted if you sign source licence agreements, and writing free software thereafter could be difficult. This is the original basis for the SCO vs. IBM lawsuit. Writing any software that is in any way similar thereafter could be different. difficult.I hate it when the typos correctly spell a different word. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Derivative effects.
On Saturday 24 January 2004 01:43 am, Day Brown wrote: There is one other example from computer history that applies to our power to control our own system: .zip. Years ago, the BBS networks were setup with archived files available with the .PAK extension. It was .arc . When Phil Katz crafted a new archive tool, he offered it to BBS users for free, to extract their .pak downloads. The corporate owners of PAK had the money and the lawyers, and found a judge who saw things their way, and sued Phil, saying that they owned 'pak' as a copywrite. Arc was released as source, before GPL was accepted by the community. Here is the original ARC license, in full: You may copy and distribute this program freely, provided that: 1) No fee is charged for such copying and distribution, and 2) It is distributed ONLY in its original, unmodified state. If you like this program, and find it of use, then your contribution will be appreciated. You may not use this product in a commercial environment without paying a license fee of $35. Site licenses and commercial distribution licenses are available. A program disk and printed documentation are available for $50. If you fail to abide by the terms of this license, then your conscience will haunt you for the rest of your life. PKARC was derived from the original ARC sources, in violation of the license. PK never released source. The original was just a fast ARC, then PK extended it to use different methods of compression, making it incompatible. The PK version had some critical parts hand coded in assembly language, and by far outperformed the original. So- Phil sent out an email to all the BBSes, announcing that his software would no longer be able to extract '.pak' files, and suggested that we all use .zip instead. PKZIP/PKUNZIP is still the defacto dos/win archive standard, and PAK INC... went out of business. Point being, that it was not up to the judge, nor the lawyers, it is up to us. PK's first change after losing the lawsuit was pak which was the same thing changed only to make it incompatible. Then it was replaced by zip. The source for zip was never released, but PK did release specs that someone could use to make another program that was compatible. The InfoZip package available on Debian, and WinZip are both non-PK, from the specs. At the time, I believed like the majority, that Henderson was just jealous of his competition, because he couldn't keep up. In hindsight, now I see it Henderson's way. How is this case different from GPL violations today? http://www.esva.net/~thom/philkatz.html http://www.was-ist-fido.de/doks/fnews/fido540.txt apt-get install arc -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Derivative effects.
On Saturday 24 January 2004 07:11 pm, Bijan Soleymani wrote: On Sat, Jan 24, 2004 at 06:28:17PM -0500, Al Davis wrote: You may copy and distribute this program freely, provided that: 1) No fee is charged for such copying and distribution, and 2) It is distributed ONLY in its original, unmodified state. How is this case different from GPL violations today? The GPL gives you the right to modify the program and distribute modified versions. That means that you'd be allowed to modify the program to make it run better. The major requirement of the GPL is that you have to distribute the source along with the binaries. The GPL also allows a fee to be charged. Good point. My point was that both are the same in that the issue was that a free program with source distributed is illegally forked and taken proprietary.It is an example of a case that GPL is designed to prevent. Remember .. this happened at a time when GPL was not well known. The more usual was a shareware license, like arc. Perhaps if it was originally released under GPL this would not have happened. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: locales - C
On Monday 19 January 2004 05:22 am, Wolfgang Lonien wrote: in unstable, some upgrade must have shot my locales-config. I 'dpkg-reconfigure'd it, but the default 'C' locale is missing. Does anyone have a clue what happened and where and how I can get it back? Programs like Thunderbird complain since then... I see that often. Perl complains. I still have no clue what happened, but when I log out, then log in again, I get the locale back. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: My email is rejected by some sites
On Thursday 11 December 2003 03:56 pm, Joerg Rossdeutscher wrote: Yes, many ISPs do that, and it's a good thing. We all would drown in spam if they accepted mail from everywhere. There is absolutely nothing you can do except to use your providers mailserver. On Saturday 13 December 2003 04:46 am, Joerg Rossdeutscher wrote: It is a good thing. What kind of mail comes from dynamic IPs? In 99% it will be spam from open relays, misconfigured adresses ([EMAIL PROTECTED]),... .. Not everyone should use a private mailserver. Hobbyists and Freaks should not run such service, it's a job for professionals, and those have a static ip. I'm really tired of writing a lot in mailinglists an get lots of You mail to [EMAIL PROTECTED] could not be delivered... So you believe the ISP should censor our mail, or at least be given the opportunity? You believe that Hobbyists and Freaks especially need for their mail to be subject to screening, but corporations don't? They do spam filtering. What's next? Adding advertisements? Harvesting the addresses and selling them to spammers? Allowing certain spam to get through, for a price? Discarding your own mail as spam, because you used a naughty word? Discarding your own mail as spam because of your political beliefs? It is worth putting up with some spam to get a free, uncensored, fast email system. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: My email is rejected by some sites
On Sunday 14 December 2003 12:41 pm, Magnus von Koeller wrote: On Sunday 14 December 2003 17:21, Al Davis wrote: It is worth putting up with some spam to get a free, uncensored, fast email system. Free, uncensored and worthless would be a better description for my email account if it wasn't for my Spam filtering - considering that 95% of my email is Spam. YOUR spam filtering. Actually ... I have several email addresses, and my mail comes in from several paths. MY spam filtering seems to be nearly 100% effective on the freeelectron address, which I control. It seems to be nearly 0% effective on the ieee address, which someone else controls. A couple of years ago, I tried configuring my home system to apply that rule about tagging dynamic ip that doesn't go through a relay. It was 0% effective. 100% of what it tagged was non-spam. 100% of the spam passed. It was effective at tagging viruses, but on Linux they were not a problem. It is easier and more effective to reject microsoft executables. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Cell Modem and Linux
On Friday 10 October 2003 04:16 pm, Bill Moseley wrote: Anyone have information on using a cell modem with linux? It's not for a desktop (i.e. not pcmcia). What kind of drivers are needed, if any? I have one and use it occasionally. It is a modem, so it interfaces like a modem. No special software is needed. Some warnings . 1. It doesn't work everywhere. There are some places where it doesn't work at all, no matter how good the signal is. It has something to do with what kind of system you are on. Even if it is supported locally, you need a very good signal. 2. It is slow. At best, it connects at 9600. Some dialup ISP's will not connect that slow, leaving you with no way to connect. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: more OT: search-and-replace commands (was Re: OT: Why is C so popular?)
On Thursday 28 August 2003 03:08 am, Alex Malinovich wrote: As much as I hate to admit that MS can do some things right, their 'ren' command (rename, instead of doing a mv from name to name like we do) is actually quite intelligent. Wasn't that part of QDOS? -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: COBOL compiler
On Tuesday 26 August 2003 04:35 am, Alex Malinovich wrote: I've, unfortunately, been forced into taking a COBOL class as a requirement for getting my BS. (And that's just what it is, a load of BS...) What's worse is that I can't seem to find any Free COBOL tools. 'apt-cache search cobol' returns 3 hits, all of which are documentation utilities. Any suggestions? Or am I going to be stuck using school PC's loaded with Windows and more proprietary software than you can shake a warez-kiddie at? You didn't say what your major is, so I am assuming it is CS. It is worth studying many languages, including those that do not seem to apply to your present interests. You are not expected to like them all. COBOL should be studied for several reasons. First, it uses a syntax that is more conversational than other languages. Also, it is has been and still is of significant historical interest. By learning it, and learning its style, you learn programming techniques that you would probably not be exposed to otherwise. Too many CS programs today put too much emphasis on a single style of program development, usually object oriented. You can't really know what that is, or what its benefits are, until you contrast it against the others. Windows, too, has its place. Too many CS schools use MS-Windows exclusively. In my mind, this destroys their credibility. As much as I like GNU, Linux, free software,.. I must admit that an environment that is 100% GNU, Linux, free also is of limited credibility. You see only one side. A truly good program will use a mix of systems. A few years ago I taught a senior CS course comparitive programming languages. There were assignments in FORTRAN, AWK, Prolog, and Lisp. We also spent some time on FORTH, and a little (too little) on COBOL. If I ever do that course again, there will be assignments in COBOL and FORTH. The only reason there wasn't is that I couldn't find appropriate free software in time. I do know where to find them now. In this course, it wasn't enough to just get a program working in each of these languages. I required them to be in the style of the language. Although you can make believe they are all C, I didn't accept that. I even enforced a no assignment statement rule in Lisp. FORTRAN here meant FORTRAN-IV, that is the 1966 version, that didn't have block if, so they had to use goto's. So, stop griping about COBOL. Do your best. Learn why it is the way it is, and how it influenced the more modern languages. Learn the style, so when someone gives you a COBOL-style program in C++, you will understand it. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: [OT] C++ question re. dyn. mem.
I think all of the responses are missing something very
important. Let me try
On Sunday 03 August 2003 08:57 pm, MJM wrote:
Will the free store be properly maintained when the following
is executed? // a simple object is defined
typedef struct
{
uint32_t a;
uint64_t b;
uint8_tc;
} t_my_type;
OK so far. It's more elegant to say:
struct t_my_type;
{
uint32_ta;
uint64_tb;
uint8_t c;
};
but I won't dwell on that.
It is not clear what uint32_t, uint64_t, and uint8_t are. Let
me assume that they are names of classes.
// allocate some memory for an instance of this object
t_my_type * p_a = (t_my_type) new t_my_type;
You left out the * in the cast. I assume this is a typo.
Actually, the cast isn't necessary. The C++ new is not just
another name for malloc. It gets memory using malloc, then
invokes a constructor to initialize it. This difference is
important. The cast isn't necessary because it already is the
correct type.
// change the way it is accessed to prove a point
int * p_b = (int *) p_a;
// p_a and p_b point to the same block of dyn. allocated
memory;
Hopefully. In this case, probably, but it cannot be guaranteed
to be true in general. When you use multiple inheritance, it
will probably not be true.
// return the memory to free storage using the second
// pointer
delete p_b;
Delete is not the same as free. It first invokes the
destructor, then returns the space using free. Since you
changed the type of the pointer, you get a different
destructor.
Since you didn't define a destructor, by default it just calls
the destructors for all the members.
I think the free store will be maintained properly because
there is a control block attached to the allocated block of
storage that defines the length of the allocated memory.
For this specific example, probably, but you can't count on it
in general.
Since you didn't say what uint64_t is, let me make a diabolical
definition of it.
class uint64_t
{
long* x;
uint64_t() {x = new long;}
~uint64_t() {delete x;}
};
Now, when I do:
t_my_type * p_a = new t_my_type;
I actually get the struct, but since uint64_t has a constructor
it is initialized. In this case, the initialization uses new
to get even more memory. So, you actually have 2 blocks of
memory.
If I do:
delete p_a;
the destructors will be called, deleting the hidden memory
block, then the real one gets freed.
If I do:
int * p_b = (int *) p_a;
delete p_b;
It doesn't call the destructors, so the internally allocated
piece is still allocated, and no longer accessable. It's a
leak.
Actually, you get the destructor for int, whatever that is.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Modem: LSR safety check engaged
On Thursday 29 May 2003 06:47 am, Sebastian Canagaratna wrote: ANyone know what this means? ANybody successfully got this card to work with Linux? I tried ActionTEc's PCI 56K V92 Call Waiting modem, which is supposed to work with Linux; its existence is not recognized either by windows XP or by Linux. No experience with that card, but I had a similar experience with another. I was trying to get a PCMCIA modem to work on my laptop. I was getting that message on boot, and it didn't work. syslog showed the modem (strangely) as /dev/ttyS04, which also didn't work. Meanwhile, it worked on Knoppix, Mandrake, and Red Hat as /dev/ttyS01 . The problem was a misconfigured setserial. Not knowing any better, somehow I installed it, with manual configuration. The problem was solved by changing the configuration to #kernel. Then I questioned why do I need setserial? So ... apt-get --purge remove setserial . Works perfectly now, as /dev/ttyS01 . Apparently, setserial is used when you have lots of modems, or a multi-port card. I still don't know what it means. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: KNOPPIX as an installer for Debian
On Sunday 30 March 2003 10:38 pm, Paul Johnson wrote: I gotta wonder why everybody's harping so badly on Memorex. I've got to really abuse the media to burn a Memorex brand coaster. So far, the only Memorex CDR I couldn't burn was one I made a fractal out of in the microwave with. Do y'all microwave your CDs before using them or something? Maybe it is a problem that has nothing to do with the quality of the CD's. Here's my own experience ... I was having all kinds of trouble with bad disks. The procedure seemed to work fine, but the disks had errors. There were a few (about 3 on average) corrupted files on most disks. I had a cheap drive, and thought that was the problem, so I bought a new one. I was told that Plextor was the best, so that's what I got. It still had errors, about the same as before. I use mkisofs and cdrecord. After getting a bad one, I tried again from the same image file. Same errors. Check the file. It's bad too. (CD-RW, of course, for the experiments) It turns out the problem was the VIA DMA bug. It worked with DMA disabled, but a large partition with ext3 was now too slow to burn a CD at 4x. (Reiser on the same size partition was ok at 12x, with DMA off.) Fortunately, my motherboard also has a Promise controller on it. Use that instead, problem solved. No more coasters. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: How to manage a large number of JPEG images?
On Monday 31 March 2003 08:13 pm, Leo Spalteholz wrote: In case you are using KDE, go to the directory that has your pictures in it and hit CTRL-I. That'll bring up a dialog of options for your gallery and then generates a gallery style webpage that you can save. One big thing wrong with it... It generates the thumbnails in .png format. As a result they are about 10x the size they should be. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: How should we handle people who can not unsubscribe?
On Fri, Mar 28, 2003 at 04:04:38AM -0800, Osamu Aoki wrote: Oh, as for subscription address, maybe we need to tell outlook user how to read their mail header. People forgets where they subscribed from and outlook users tend to lack skill to find it or read web site. On Friday 28 March 2003 15:22, Nathan E Norman wrote: style=mostly joking Or, we could require users to pass some sort of comprehension test before successfully subscribing to the list /style On Friday 28 March 2003 08:56 am, Joerg Johannes wrote: Such as changing the confirmatory mail when subscribing? It does not longer contain the instructions in the body, but in the signature... (To subscribe, just reply to this e-mail...) Change the confirm procedure The way it is now: When subscrbing the list sends you a message. You must confirm by responding to it, keeping the subject intact. Just reply does it. The body doesn't matter. Proposed change: The list sends you a message. You respond, keeping the subject intact, but in the body you must copy the Received headers from the confirmation message. The body must consist of that and nothing else. How's that for some sort of comprehension test. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
download decelerator
With this discussion on a download accelerator ... What I really need is the opposite. I have a full time connection. The problem is that when I download something large, like a CD image, other activity slows down, often to the point of being unusable. Is there a way to slow down a particular download to improve the performance for whatever else I might be doing on the net? I need the equivalent of nice for downloads. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: howto verify burn?
On Thu, 6 Mar 2003 04:22, Ray wrote: ?how do i verify that the burned cd is correctly burned? ?i burned a set of Debian 3.0r1 cds and have md5sums of the isos, but they don't match the output of ?md5sum -b /dev/cdrom In my experience, the actual burn is often a little longer than the .iso it comes from. It is padded with something that might as well be garbage. If you only read the first N bytes, where N is the length of the .iso you used, it should work. How about head --bytes=426893452 /dev/cdrom | md5sum -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Switch from Gnome to KDE...
On Monday 03 March 2003 09:10 pm, Leo Spalteholz wrote: I would recommend the unstable sources.. As far as I can tell all of KDE is now in it. No. kmail and knode are still not there. I don't know what else is missing. On my system, noatun acts strange sometimes, but kwave is fine. I don't know what the problem is, but have not had the time to investigate. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: TFT monitor and XF86config-4
On Tuesday 04 March 2003 11:52 am, Mike! wrote: Dave Selby wrote: ?Have just bought a TFT monitor, works well except occasionaly some of text is blurred on top pixel line ? ?I remember when I was installing debian there were questions about TFT monitors, at the time I ignored them. ? ?How can I re-run the configuration again without re-installing ? I think it will be a ? ?dpkg-reconfigure ? ? dpkg-reconfigure -plow xserver-xfree86 When I took the easy way it configured mine wrong. You may need to select advanced and manually configure scan rates and resolution, eliminating those that work but poorly. I recommend looking in the monitor manual and selecting scan rates in a narrow range so it can only choose the one you want. When I left the range wide, it chose 1280x960 on a 1280x1024 monitor, even though 1280x960 was not listed as a proper choice and 1280x1024 was. This resulted in a vertical smear over the whole screen. Another option is to manually edit /etc/X11/XF86Config-4 . Just specify the exact rates you want, and take out the others. Another comment on TFT monitors It is not enough to match the scan rates. Every pixel needs to line up, in the correct phase, otherwise you get smearing. It must be exact. Most monitors have a means of automatically matching it. You might need to activate it by pushing a button on the monitor. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: GUI Mail Client
On Tuesday 25 February 2003 11:56 pm, Rus Foster wrote: Can anyone recommend a mail client which can take my existing pine config and folders? One that takes the whole config will be hard. Several will take the folders, including kmail and mutt. (but mutt isn't GUI.) I use kmail and mutt sort of interchangeably. I use kmail at home, and mutt away from home when the gui is too slow. Since they use the same format, I can go back and forth between them. I used to do this with pine, but for me mutt has replaced pine. One thing I like about kmail (compared to other GUI mail programs) is that it does only mail. One bug in kmail is that it uses index files. If there isn't one, it will create it. If you change a mail file (mbox format) with another program, only a little, sometimes it won't discover that the index doesn't match and will give a messed up display. The solution is to remove the index file. Sylpheed uses a different storage format. I couldn't figure out how to go back and forth between it and a text program like mutt. That means you will have the same problem moving from pine. I don't like evolution for the same reason I don't like Outlook. A mail program should do only mail. It would be SO nice if they would all use the same storage format. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: About to go all Deb
On Tuesday 25 February 2003 08:24 am, John Anderson wrote: I am relatively new to Linux, been running Mandrake 9.0 for about a fortnight, and after lots of reading I have decided to move to Debian. If you have enough space (which you do with 80 gb) it is a good idea to reserve a group or partitions for a re-install. What I have is : / 250 mb /var1g /usr5g that is unique to an installation. Others (/tmp, /home) are shared between installations. Then I have another /, /var, /usr for another installation. By doing this, I am able to do a complete reinstall while keeping the old one. This gives a fallback if the new one doesn't work correctly, and you keep your old config info. When I first installed Debian, the fallback was Mandrake. It is also handy if you are running an unstable system. Keep one before a dist-upgrade for when something gets removed or doesn't work. On my main machine, I have 2 drives, with 4 boot configurations. There is a duplicate /home on the other drive that I use for backup (with rsync). If you are setting up from scratch .. I have found that reiserfs works better on big partitions, and ext3 works better on small ones. If you are measuring in gigs, that is big. On a 20 gig partition, I get about a 4:1 performance improvement with reiser. I don't know about the other filesystems such as jfs. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
spam spam spam
A Spammer has the list. It took only one posting to this list to get spammed. Here's where it came from: 4.40.163.14 Server: 63.228.184.2 Address:63.228.184.2#53 Non-authoritative answer: 14.163.40.4.in-addr.arpaname = lsanca2-ar32-4-40-163-014.lsanca2.dsl-verizon.net. Authoritative answers can be found from: 163.40.4.in-addr.arpa nameserver = dnsauth1.sys.gtei.net. 163.40.4.in-addr.arpa nameserver = dnsauth2.sys.gtei.net. 163.40.4.in-addr.arpa nameserver = dnsauth3.sys.gtei.net. dnsauth1.sys.gtei.net internet address = 4.2.49.2 dnsauth2.sys.gtei.net internet address = 4.2.49.3 dnsauth3.sys.gtei.net internet address = 4.2.49.4 al@oliver:/home/alping 4.40.163.14 PING 4.40.163.14 (4.40.163.14): 56 data bytes 64 bytes from 4.40.163.14: icmp_seq=0 ttl=113 time=696.3 ms 64 bytes from 4.40.163.14: icmp_seq=1 ttl=113 time=365.6 ms -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: unsubscribe doesn't work!~
On Saturday 15 February 2003 11:15 am, Fer'had Erdogan wrote: Not following the footer instructions, not via the website, not even by sending multiple e-mails to the list manager. So, if there's anyone here who cares about keeping this list decent; take me out! And maybe even fix the problem while you're at it so that others can unsubscribe when they want to. My guess is that [EMAIL PROTECTED] isn't subscribed, and never was. There is some other address that you have used, possibly without your knowledge, at some time in the past that is subscribed. It is likely to be some variant. Perhaps your ISP changed its name, and you subscribed under the old name. Perhaps your ISP automatically changes all the names to the new one on all mail you send and receive. Perhaps your ISP told you to change your address, but keeps forwarding the old one. What do you get back when you try? If the address doesn't match, there is nothing anyone can do. The best way to trace this is to look at the headers. Most email programs hide most of the headers, so you need to do some extra work to see them. In there, you will find the complete routing that a message took to get to you, including intermediate sites it has been forwarded through. You may find a like like Envelope-To that probably refers to another or your email addresses, that you have forwarded so they all look the same to you. If you send a copy of the complete headers from the mail you get from the list, maybe someone can help. by the way I just changed the address I subscribe to this list with. Just unsubscribe the old one, subscribe the new. It worked as intended. Just for kicks, I tried to unsubscribe an address that isn't subscribed. I got back a reply that said the address was not subscribed. (as expected) -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
File systems -- reiser vs. ext3
I am trying to understand them. The basic question ... Why choose one over the other? Technically, what is different? My benchmarks seem to show that Reiser is superior, but they may be flawed. What I see is that it is faster and uses space more efficiently. I get about a 4:1 difference for an md5sum of a large .iso file. This was different partitions of the same drive. Later, I copied all out of an ext3 partition, reformatted it as reiser, and got the same 4:1 speedup. I have been using Reiser for a few years without trouble. (On the laptop since Suse first made it available as an option.) Does anyone know where I can find technical documentation on ext3? Does anyone have any hard data (as opposed to anecdotal like mine) to show the relative merits? (other than ext2 compatibility) Tech documentation of Reiser is available on his web site. I am looking for something comparable on ext3. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: IDE disk corruption - A7M266-D; which kernel patches? or other solution
On Sat, Jan 18, 2003 at 11:27:28AM -0500, Daniel Barclay wrote: I'm getting disk corruption if I try to enable DMA mode for my IDE disks. On Saturday 18 January 2003 04:14 pm, Pigeon wrote: If you have a VIA chipset try making sure that VIA chipset support is included in the kernel. How do I find out? I am using the one on the woody bf2.4 cd. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: IDE disk corruption - A7M266-D; which kernel patches? or other solution
On Sunday 19 January 2003 02:48 pm, Bob Proulx wrote: Wayne Topa wrote: Al Davis([EMAIL PROTECTED]) is reported to have said: How do I find out? If you are using the bf24 kernel then the config is installed in /boot. grep VIA /boot/config-2.4.18-bf2.4 CONFIG_BLK_DEV_VIA82CXXX=y CONFIG_VIA_RHINE=m CONFIG_AGP_VIA=y CONFIG_SOUND_VIA82CXXX=m Thanks. That is what I was looking for. My real reason for asking ... I was getting disk corruption with an older kernel (2.4.8 or 2.4.18, Mandrake). A colleague said it might have something to do with a hardware bug (south bridge VIA VT82C686). He also said there was a fix in recent kernels. I thought it was bad memory. It did have CONFIG_BLK_DEV_VIA82CXXX=y . The workaround was to disable DMA. As expected, it has now been reliable but slow. I have since installed Debian, with 2.4.20-bf2.4, and now I wonder if it is safe to re-enable DMA. Lots of good documentation about the Debian way of installing kernels in /usr/share/doc/kernel-package/*. That module wasn't installed here. I will look. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
