Re: kernal-image-2.0.0-0 + stuff...
On Sun, 30 Jun 1996, Andy Heroff wrote: > Okay, first off, thank you much for helping me out with the dselect probs I > was having. Mounting my D drive before running dselect worked wonders. :) > > Now, while configuring stuff, I ran into about 4 packages, including > kernal-image-2.0.0-0, that complained that while running --configure or > whatever, that it couldn't find a directory or file. I included the latter > because it said in the package selection option that it was required. Is > this true? And is there any reason in particular that dpkg is having > problems finding the right directories? I know I wasn't very descriptve of > the problem, but as none of the packages are that important to me to get > installed (except for kernal-image?) I didn't write down every single > message thrown at me. This sounds like the perl problem: the current version of perl installs without a necessary sym link. The simple solution is to make this link. Try: ls -s /usr/bin/perl /bin/perl The install should proceed properly. > > Next, I'm curious if the editor Pico is anywhere in the Debian distribution. > I use it frequently at school, and have found it fast and convenient. If > it's not anywhere in the distribution, is anyone currently working on it? > Because of a more restictive distribution license, I was force to put pine, imapd, and pico into the non-free section. You should be able to find them there. > And finally, I seem to have lost the XF86Config program... I recognized it > when I was configuring one of the X11 packages, but I didn't have time to > set it up properly, so I skipped it. I found XF86Config somewhere in > /etc/ but it was just a symlink to a file that didn't exsist. Is > XF86Config in one of the packages, or do I have to run a configuration > script again? (If so, which one?) Thanks for any help given on the above. > --- My memory says that when you install the x stuff (xbase I think) you are walked through a question and answer session that results in a properly configured system. (I believe you must install at least the server package you will need before you can install the base packages. Don't forget the fonts.) --- Hope this helps, Dwarf -- aka Dale Scheetz Phone: 1 (904) 877-0257 Flexible Software Fax: NONE Black Creek Critters e-mail: [EMAIL PROTECTED] If you don't see what you want, just ask --
/bin/perl? [was Re: kernal-image-2.0.0-0 + stuff...]
Hi! On Sun, 30 Jun 1996, Christian Hudon wrote: > On Sun, 30 Jun 1996, Andy Heroff wrote: > > Now, while configuring stuff, I ran into about 4 packages, including > > kernal-image-2.0.0-0, that complained that while running --configure or > > whatever, that it couldn't find a directory or file. I included the latter > > because it said in the package selection option that it was required. Is > > this true? And is there any reason in particular that dpkg is having > > problems finding the right directories? I know I wasn't very descriptve of > > the problem, but as none of the packages are that important to me to get > > installed (except for kernal-image?) I didn't write down every single > > message thrown at me. > > Just guessing wildly, make sure you have /bin/perl as a symlink pointing to > /usr/bin/perl... It's a known problem with the Perl package in 1.1. There > should be a new perl package in 1.1-fixes fixing this. Or you can just make > the symlink yourself.(If you want more than wild guesses, you really must > provide more info.) Of course, making a symlink is the easiest way to solve this problem, but wouldn't it be better if the installation scripts, that depend on /bin/perl, were changed to reference /usr/bin/perl? IMHO, we would someone need to have perl in /bin, if it is symlinked to /usr/bin/perl and therefore *ONLY* available, if /usr is mounted? And /bin should only contain "Essential user command binaries". Thanks Chris
Debian 1.1 and the select system call
Hi,
I have have compiled (kernel 2.0 ) on debian 1.1
(libc libc5_5.2.18-9).
In a program that I have written I am using a select
system call. The select waits indefinitely on input from a serial line and
a FIFO node.
for (;;){
FD_ZERO(&readfds);
if (fifo_fd != -1)
FD_SET(fifo_fd,&readfds);
FD_SET(fd,&readfds);
sel = select(100,&readfds, NULL, NULL, NULL);
if (sel > 0) {
if (fifo_fd != -1 && FD_ISSET(fifo_fd,&readfds)) fifo_read();
if (FD_ISSET(fd,&readfds)) {
}
as soon as the fifo queue gets an input, the select system call
keeps returning with a flag indicating additional fifo inputs.
even though that there are no additional fifo inputs and the previous
input was processed.
In other words the program goes to an infinite loop because it thinks
that there is additional unprocessed data in a fifo queue, when actually
there is none (there was only one input that was processed).
The problem goes away if I just reinstall the old kernel (1.3.72)
-Any help will be appreciated
-Oz
--
<
NAME Oz Dror, Santa Monica, California
EMAIL [EMAIL PROTECTED] <>
PHONE Fax (310) 396-5798
-BEGIN PGP PUBLIC KEY BLOCK-
Version: 2.6.2
mQBtAzA/tLQAAAEDAKUy/TEjQ/jiZ+9/WJb/+NHxqkvOxGZ3W/F2JCNm5v5ZTZz+
BVZC9GM/I+plQ8xz+7B+KhDSVax8gxNTAkJ+I7P/zAP2ZDMwVf4lq5ZFxMJC+7c7
ET+hNtmQUt8vCVR8hQAFEbQZT3ogRHJvciA8ZHJvckBuZXRjb20uY29tPg==
=EU23
-END PGP PUBLIC KEY BLOCK-
>
gcc and libg++27
I have been trying to use gcc to compile some C++ code. Libg++27 seems to be the supplier of standard headers for C++. I have both gcc-2.7.2-8 and libg++27-2.7.1-2 installed. When I try to compile something, I get: g++ -c -I./source/include -I/usr/include/g++/std ./source/parody2/tnode.cc In file included from /usr/include/string.h:33, from /usr/include/g++/std/cstring.h:18, from source/include/parody.h:8, from ./source/parody2/tnode.cc:7: /usr/include/g++/std/stddef.h:15: parse error before `=' /usr/include/g++/std/stddef.h:15: warning: ANSI C++ forbids declaration `NPOS' with no type or storage class /usr/include/g++/std/stddef.h:15: warning: implicit declaration of function `int size_t(...)' It goes on like this, complaining about syntax in many of the standard headers... /usr/lib/g++-include/std/bastring.h:67: `__bsrep >::operator [](...)' must take exactly one argument /usr/lib/g++-include/std/bastring.h:67: confused by earlier errors, bailing out make: *** [tnode.o] Error 1 untill it finally errors out. This is my first attempt at C++ on Linux, (I have worked in Borland up til now) so pardon my ignorance. Do I need to do any pre-processing here? Or have I just left out a necessary switch for gcc? Any help would be appreciated. TIA, Dwarf -- aka Dale Scheetz Phone: 1 (904) 877-0257 Flexible Software Fax: NONE Black Creek Critters e-mail: [EMAIL PROTECTED] If you don't see what you want, just ask --
Re: gcc and libg++27
Dale Scheetz <[EMAIL PROTECTED]> writes: Are you using STL? If so, I was completely unable to compile certain bits of STL code until I modified the bottom of /usr/include/g++/std/bastring.h like this: #if !defined (_G_NO_EXTERN_TEMPLATES) //#include breaks stl (RLB) -- ambiguous instantiations // same as complext.h including cinst.h #endif and complext.h like this: //#include breaks stl (RLB) -- ambiguous instantiations. // same problem as bastring.h including sinst.h Don't know if this is your problem, but I thought it might help. Hopefully libg++ will get this straightened out upstream soon. -- Rob
Re: netpbm package?
> Does someone have a netpbm package for debian? I just found the pbmplus > package that looks pretty old. Two people had volunteered to set up new netpbm packages, one of them was going to make it a shared package. Since nothing has been said, and nothing was produced, I've gone ahead and uploaded a netpbm and netpbmdevel set of packages. > If there is no such package, I will probably create one. If you would like to take this one over, I would greatly appreciate it. I'm doing this out of a semi sense of obligation since I've let pbmplus sit still for so long, but I have no desire to maintain it. The source and packages should be uploaded within the hour. Jim
Floptical disk drives: Valuestor (Sysgen)
Hi there, I have looked in the how-tos, and have also done searches on the web.. What I need to know, is whether Debian Linux supports 21Mb floptical disk drives such as those produced by ValueStor ? Is support for such devices in the kernel, or will I have to go and recompile the kernel ? Will the floptical be detected and work straightaway without too much hassle if I put it in ? Excuse me if I am a dunce here, but I didnt see anything to tell me what I had to do in any of the online documentations, except that Linux works with floptical drives ... Perhaps someone out there could give me some kind of guidance... or at least tell me where to look thanks in advance Cormac --- Cormac McGuinness Experimental Physics Department,Tel:+ 353 1 706 2205 University College Dublin, Fax:+ 353 1 283 7275 Belfield, Dublin 4. e-mail: [EMAIL PROTECTED] Republic of Ireland ---
Re: color_xterm?
On Sun, 30 Jun 1996, Christian Schwarz wrote: > Is there a color_xterm in Debian 1.1 - or something similar? Right now, > I'm using the old color_xterm from Slackware. Yes. It's called xterm-color and is in the x11 section. Christian
Re: rev?
[EMAIL PROTECTED] (David Puryear), 28 Jun 96] > Does anyone installed linux-source using dselect? I'm getting > error when I try. Oh well, this is not important since I have > downloaded and untared 2.0.tar.gz. Just want to mention it. I had a similar problem, and it seemed related to having downloaded a a kernel-source-1.3.99 or some in dselect and a kernel-source-2.0.0 on the disk. It ended up installing fine with "dpkg --install ". I had some other problems with dselect, and ended up installing many packages with "dpkg --install" instead. Perhaps it has something to do with that Packages file synch. problem. Best regards. Anthony Godshall - Paradigm Shifting and Testing Digital Models and Transformations Project Engineer, Educator, Programmer
Re: ZIP-drive [parallel port zip drive as a portable Linux
Aslak wrote... > > > > I have one of the IOMEGA ZIP-drives,100MB, parralell version, > > > > and was wondering if it is, if it can, and if it will be > > > > possible to have one's linux system one of these ZIP disks. It > > > > _is_ possible to have a system on 100Megs, and it would be > > > > extremely convenient, at least for my purposes. Eric Hoeltzel... > I just loaded up Linux on a 1 gig iomega jaz drive last weekend. It > was quite painless, just dropped a friend's aha1515 into my 486dx50 > and ran the install and make him a bootdisk with loadlin. I would > say that it was somewhat noticably slower than an IDE drive, but > pretty acceptable nonetheless. X ran fine too. Highly convenient > arrangement I would say, and I had never seen one of these jaz > drives before. The SCSI version of the zip drive is a bit faster than the parallel port version the I assume your Jaz drive is a SCSI version too? Best regards. Anthony Godshall - Paradigm Shifting and Testing Digital Models and Transformations Project Engineer, Educator, Programmer
Re: Upgraded Motherboard
[Andrew Stephen <[EMAIL PROTECTED]>] > I have just upgrade the Motherboard in my Linux Box which is running > Debian 1.1 from a DX4/100Mhz to a Pentium 75Mhz. When I compare > /proc/cpuinfo it has dropped from 39.7 bogomips for the DX4/100 to > 29.79 bogomips. Any ideas why there is a decrease. "Upgrades" ain't always. Da facts, or da hype? It's no secret that the fastest 486's outperform the slower Pentiums. 486DX4's are clock tripled, and also L1 cache doubled (I think). Best regards. Anthony Godshall - Paradigm Shifting and Testing Digital Models and Transformations Project Engineer, Educator, Programmer
