Linux-Development-Sys Digest #633, Volume #6 Sun, 18 Apr 99 02:14:19 EDT
Contents:
problem with executable file (nturdali)
Performance question (Zoom)
framebuffer and svgalib (Chris Frost)
Re: Hp Kayak (Thomas Zajic)
Waiting for Red Hat 6.0 ("folkert meeuw")
Re: problem with executable file (Jim Roberts)
Re: Red Hat 5.2 RAMDISK questions (Sami Tikka)
Re: framebuffer and svgalib (Chris Frost)
Re: Can you make a driver that runs as an application? ([EMAIL PROTECTED])
Re: Minimum Hardware For Ghostscript on Linux (Hans-Joachim Baader)
Re: SMP Linux, Any Catches? (Mark Hahn)
Re: Can you make a driver that runs as an application? (Martin Maney)
Re: Waiting for Red Hat 6.0 (Frank Sweetser)
Re: PCI DMA to user space possible? (Peter Samuelson)
[Q] Linux pstat() or table() equivalent (Paul Roebuck)
Re: Epson Stylus Color 900 (Warren Young)
Re: Linux Servers, what to buy? (Warren Young)
----------------------------------------------------------------------------
From: [EMAIL PROTECTED] (nturdali)
Subject: problem with executable file
Date: Sat, 17 Apr 1999 18:38:12 GMT
Hi,
I am new to the Linux world.
Today I needed to run a C program. Compilation seems to work OK.
But I cannot use executable file.
After commands:
# gcc -o myprog myprog.c
# myprog
the system responds:
"dash:myprog: command not found".
Why? What should I do?
nturdali
------------------------------
From: Zoom <[EMAIL PROTECTED]>
Subject: Performance question
Date: Sat, 17 Apr 1999 08:24:15 -0700
I have red hat 5.2 with the 2.0.36 kernel. If I time a "dd bs=1024k
count=300 < /dev/hda > /dev/null", it takes around 29 seconds to
complete. If I reboot with the 2.2.6 kernel and do the same test it
takes around 34 seconds to complete. Any suggestions on why the large
time difference?
Thanks,
Zoom
------------------------------
From: Chris Frost <[EMAIL PROTECTED]>
Subject: framebuffer and svgalib
Date: Sat, 17 Apr 1999 12:06:58 -0500
I'm just trying out the framebuffer (matroxfb actually) and haven't been able to get
svgalib to run correctly (lines are
extremely garbled). Is there any way to have svgalib-based programs run when using a
fb device instead of the old
text-mode support? (Fwiw I'm running debian 2.0, which has svgalib 1.2.13).
thanks for any help,
--
- Chris
-- Visit Me at http://www.frostnet.advicom.net --
------------------------------
From: Thomas Zajic <[EMAIL PROTECTED]>
Subject: Re: Hp Kayak
Date: Sat, 17 Apr 1999 15:24:31 GMT
[ e-mail CC�ed ]
Anis Hanna wrote:
> Has anyone installed Redhat 5.2 on HP Kayak XA machines?
> I did, works fine, except my xfree86 version 3.3.3 hangs the machine
> within a few minutes of starting Xwindows. my configurator detects the
> video card correctly (cirrus clgd5465). I tried everything in the
> troubleshooting section of the manual, disabling blibt, and
> acceleration, nothing works. Anyone had this problem?
Yes, _everybody_ with a CL-GD5465. The solution is even in the
XFree86 FAQ by now. In short, add the following lines to the
"Device" section in your /etc/XF86Config:
Option "xaa_no_color_exp"
Option "no_pixmap_cache"
The first one eliminates the hangs, the second one gets rid of
screen garbage after moving windows around.
BTW was there any particular reason for posting this question to
c.o.l.development.system instead of c.o.l.x?
HTH,
Thomas
--
=---------------------------------------------------------------------=
- Thomas Zajic aka ZlatkO ThE GoDFatheR, Vienna/Austria -
- Spam-proof e-mail: thomas(DOT)zajic(AT)teleweb(DOT)at -
=---------------------------------------------------------------------=
------------------------------
From: "folkert meeuw" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.networking
Subject: Waiting for Red Hat 6.0
Date: Sat, 17 Apr 1999 22:35:23 +0200
Hi Dear Frindly Readers,
now, since I'm waiting for Red Hat 6.0, i 've got a Problem with Red Hat 4.2
A month ago I updated to the Kernel Source 2.1.43.
After, I wrote a shell script to automate the compilation of new Kernel and
delegate the job to crond on every night at 01:00 and 03:00 and 05:00,
'cause the compilation breaks with an error sometimes (a 486-Intel/Compaq
PC)
Always the system has the same Problem, when I activate ppp-functionality
in a monolithic Kernel with menuconfig there is a library error with the
net-modules,
when I type make dep clean zImage and when I reboot the system didn't find
the
net-Modules.
But I want to route with my Linux-PC in a TCP/IP Network so I need
this functinality. And when I start a ppp Session to my Provider and
've been connect and type route to see who the Linux routed there
is no routing functional�ty to see, evenso I tested with ping from
one other PC in the Network but there is nothing going out.
Have had someone of you here make experience with it ?
MfG Folkert Meeuw
------------------------------
From: [EMAIL PROTECTED] (Jim Roberts)
Subject: Re: problem with executable file
Date: Sat, 17 Apr 1999 16:48:20 GMT
In article <7fadh0$pj$[EMAIL PROTECTED]>,
[EMAIL PROTECTED] (nturdali) writes:
> Hi,
>
> I am new to the Linux world.
> Today I needed to run a C program. Compilation seems to work OK.
> But I cannot use executable file.
> After commands:
> # gcc -o myprog myprog.c
> # myprog
>
> the system responds:
> "dash:myprog: command not found".
>
> Why? What should I do?
>
> nturdali
For all you new unix users out there!
The current directory is NOT automatically in the default
search path. (ie .) This is a good thing, and I won't go
into the details now. (If you had ever actually administered
a unix network you would know why.)
To exec a binary file in the current directory that is not in
the default search path, prepend the file name with ./ to
indicate the current directory. ie ./filename
That will run the file in the current directory.
Good luck and good learning!
--
Jim Roberts Never enough time!
[EMAIL PROTECTED]
------------------------------
From: Sami Tikka <[EMAIL PROTECTED]>
Subject: Re: Red Hat 5.2 RAMDISK questions
Date: 17 Apr 1999 19:08:19 +0300
"Jeremiah Daniels" <[EMAIL PROTECTED]> writes:
> Hello I am trying to find out how o make a Linux bootdisk that does not load
> RAMDISK at all. The system only has 4 megs of ram and cant handle RAMDISK.
Read the Bootdisk-HOWTO.
--
Sami Tikka, [EMAIL PROTECTED], http://www.iki.fi/sti/
------------------------------
From: Chris Frost <[EMAIL PROTECTED]>
Subject: Re: framebuffer and svgalib
Date: Sat, 17 Apr 1999 12:45:49 -0500
Chris Frost <[EMAIL PROTECTED]> wrote:
> I'm just trying out the framebuffer (matroxfb actually) and haven't been able to get
>svgalib to run correctly (lines are
> extremely garbled). Is there any way to have svgalib-based programs run when using a
>fb device instead of the old
> text-mode support? (Fwiw I'm running debian 2.0, which has svgalib 1.2.13).
Hmm, guess I should have checked the documentation; sorry bout that but
it's working now.
--
- Chris
-- Visit Me at http://www.frostnet.advicom.net --
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: Can you make a driver that runs as an application?
Date: Sat, 17 Apr 1999 18:28:07 GMT
In article <7f9aec$dr4$[EMAIL PROTECTED]>,
[EMAIL PROTECTED] wrote:
> In article <[EMAIL PROTECTED]>,
> [EMAIL PROTECTED] wrote:
> > On Sat, 17 Apr 1999 01:50:32 GMT, [EMAIL PROTECTED]
> > <[EMAIL PROTECTED]> posted:
> > > It's been quite some time since I've written a driver for Linux. In
> > >fact it's been quite a while since I've used Linux.
> > >
> > > I wanted to do a proof of concept experiment with Linux. I would like
> > >to write a driver that is SEPERATE from the kernel so that when I screw up
> > >I don't take the system down. I work in the real time embedded arena and
> > >am used to working with threads instead of processes. What I would like to
> > >do is make an API available to several programs that is loaded into one
> > >location in memory and is completely reentrant - basically a driver. Is
> there
> > >any way I can divorce the "driver" from the kernel so that I can do this?
> > >
> > > You may need to know what I plan to do: Basically I want to make
> > >a GUI api that runs on top of X with XSHM. This will be a low level
graphics
> > >library that will support fonts, sprites, etc., but only at the low level.
> > >I would then like to make it possible for applications to access the GUI
> > >API but without having to send data via a TCP/IP pipe or something slow
> > >and ugly like that and without a context switch. I understand that X11
will
> > >need to make a context switch, but the only reason I'm using X11 is because
> > >I already know it quite well and that it's ubiquitous. I may dump X11
> > >after a point to go directly to the hardware, but only after it is working
> > >reliably.
> > >
> > > So is this possible to do without making it an actual driver? I'd
> > >hate to be crashing my system everytime I make some stupid mistake. Being
> > >able to use X11 would be a strong plus as well which I don't think would be
> > >possible to do with a real driver.
> >
> > I think that what you *really* need to do is to look at GGI, which has
> > the intent to provide that low level stuff.
> >
>
> I've already looked at GGI, but it's a driver. I *need* the
> application to run in user mode if possible. If the driver crashes in kernel
> mode it takes down the system and quick. It is also nice to have printf(),
> syslog() and other debugging tools not to mention gdb which are also
> not available in kernel mode. My real job is a aggravation enough, for
> play I don't want the pain.
>
> Thanks,
> -Rich
>
It may be bad form to reply to oneself but I just came back from the
GGI page and have realized that it has changed considerably since I last
looked it over. I think it might work for my purposes, as long as it is
fairly stable. Now to select a video card that will work well with an
AMD K6-2 400 Mhz (want to play with flightgear) and GGI.
Thanks,
-Rich
============= Posted via Deja News, The Discussion Network ============
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
------------------------------
From: [EMAIL PROTECTED] (Hans-Joachim Baader)
Subject: Re: Minimum Hardware For Ghostscript on Linux
Date: 17 Apr 1999 09:10:46 +0200
In article <7f8ekg$[EMAIL PROTECTED]>,
Donald R. Brewer <[EMAIL PROTECTED]> wrote:
>Does anyone know what the minimum processor and memory requirements would be
>to run Ghostscript on Linux?
Years ago, I've run gs successfully on a 8 MB machine.
The latest version of gs seems to use about 4 MB for average documents.
Add 4 MB for the OS and basic daemons, and you are at 8 MB again.
Even with 8 MB you should have enough swap space. Trying to run it
with less that 8 MB is probably useless.
hjb
--
"Every use of Linux is a proper use of Linux." -- John "Maddog" Hall
------------------------------
From: Mark Hahn <[EMAIL PROTECTED]>
Subject: Re: SMP Linux, Any Catches?
Date: 17 Apr 1999 18:20:47 GMT
Clint Byrum <[EMAIL PROTECTED]> wrote:
> Hold on, you guys are confusing me. My understanding of beowulf was that it
> allowed clustered processing using Linux. I also thought that it would
it does.
> essentially treat the "clustered" systems as additional processors, sending
> jobs to them through a dedicated "cluster network". What would be the
no. what you're talking about is process migration, and it doesn't.
in fact, I don't believe Msft's clustering does either; I'm not sure
about VaxClusters. transparent migration is generally only doable on
machines with shared memory, such as big boxes from SGI/Sun/HP.
> Please, enlighten me, I'd like to use clustered processing as a
> "scalability" option for our customers.
systems that don't attempt to transparently migrate are still scalable.
in fact, you can make a good case that migration limits scalability.
regards, mark hahn.
--
operator may differ from spokesperson. [EMAIL PROTECTED]
http://java.mcmaster.ca/~hahn
------------------------------
From: Martin Maney <[EMAIL PROTECTED]>
Subject: Re: Can you make a driver that runs as an application?
Date: 18 Apr 1999 00:57:47 GMT
[EMAIL PROTECTED] wrote:
> I wanted to do a proof of concept experiment with Linux. I would like
> to write a driver that is SEPERATE from the kernel so that when I screw up
If it's SEPARATE from the kernel then it isn't a driver; what you describe
wouldn't really be a driver even if it were loaded as a kernel module, I
think. Let's see now...
> do is make an API available to several programs that is loaded into one
> location in memory and is completely reentrant - basically a driver. Is there
> any way I can divorce the "driver" from the kernel so that I can do this?
What you have described here is a shared library, not a driver. No?
> You may need to know what I plan to do: Basically I want to make
> a GUI api that runs on top of X with XSHM. This will be a low level graphics
> library that will support fonts, sprites, etc., but only at the low level.
> I would then like to make it possible for applications to access the GUI
> API but without having to send data via a TCP/IP pipe or something slow
> and ugly like that and without a context switch. I understand that X11 will
Yep, that's a shared library. Relax. There are some perhaps unintuitive
things that have to be done to build this, but the one that's more than
getting some options and rules right in your makefile is one that you ought
to be pretty comfortable with already from your embedded work: making the
code reentrant. This ought to be a piece of cake, aside from details like
-fpic (or -fPIC).
Luck!
------------------------------
From: Frank Sweetser <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.networking
Subject: Re: Waiting for Red Hat 6.0
Date: 17 Apr 1999 17:05:28 -0400
"folkert meeuw" <[EMAIL PROTECTED]> writes:
> Hi Dear Frindly Readers,
> now, since I'm waiting for Red Hat 6.0, i 've got a Problem with Red Hat 4.2
> A month ago I updated to the Kernel Source 2.1.43.
^^^^^^
ummm.... you *do* realize that not only is this an old beta version, but
this was right around the time that the filesystem rewrite actually could
cause filesystem corruption?
go with 2.2.6 instead, it's the latest stable version.
--
Frank Sweetser rasmusin at wpi.edu fsweetser at blee.net | PGP key available
paramount.ind.wpi.edu RedHat 5.2 kernel 2.2.5 i586 | at public servers
I don't think I'm gonna agree with that. Way too much visual confusion...
-- Larry Wall in <[EMAIL PROTECTED]>
------------------------------
From: [EMAIL PROTECTED] (Peter Samuelson)
Subject: Re: PCI DMA to user space possible?
Date: 17 Apr 1999 17:05:43 -0500
Reply-To: Peter Samuelson <[EMAIL PROTECTED]>
[Bryan Hackney <[EMAIL PROTECTED]>]
> > No, the frame device will tell the driver how big it is, and
> > anyway, the memory is on the frame device. See below.
[Robert Kaiser <[EMAIL PROTECTED]>]
> Not in the case of the device I had to deal with: it just DMA�s a
> stream of video data into the host CPU�s RAM. There is no RAM on
> the PCI board itself (at least none that would be visible to the
> programmer). And the size of the buffer (i.e. the picture I�m
> acquiring) is defined by the application (and it may change from
> picture to picture).
It looks like we have two separate -- opposite, in fact -- cases:
either the device in question has its own framebuffer or it does not.
If it does, there is no need for DMA, you can just mmap() the
framebuffer and go from there, zero-copy. If it does not, your best
bet might be to let the user malloc() a buffer and then let the device
do its DMA while you read()/write() it. Also zero-copy.
Two totally different board designs -- and the device driver interface
would reflect this. I now agree with you that read()/write() makes
sense here.
> Maybe that framegrabber thing was a bad example after all as it
> apparently bears the implication that a frameGRABBER always
> includes a frameBUFFER.
I think you're right. Your device abstracts better as streaming input
than as frame-oriented.
--
Peter Samuelson
<sampo.creighton.edu!psamuels>
------------------------------
Date: Sun, 18 Apr 1999 00:02:43 -0500
From: [EMAIL PROTECTED] (Paul Roebuck)
Subject: [Q] Linux pstat() or table() equivalent
I was trying a quick port of a library I have that runs on both HP-UX and
Digital Alpha.
It works fine except I don't know the equivalent routine under Linux. I
have another
vanilla version that parses "ps" output obtained via popen() but that
seems clumsy
and inelegant. Can someone provide an elegant Linux solution, if one
exists? Thanks.
/*============================================================================
*
* Name:: RetrieveAppProcessName
*
* Description::
* Provides the "glue" to give the high level API routines access to
* the name of the process. This should be replaced by an application
* registry call of sorts but that doesn't exist at this point in time.
*
* Calling Sequence::
* RetrieveAppProcessName (&processNameStr);
*
* Inputs::
* (None)
*
* Outputs::
* processName - Name of the application (i.e., argv [0]).
*
* Return Values::
* (None)
*
*============================================================================*/
#ifdef _OSF_SOURCE
#include <sys/table.h>
#ifndef _NO_PROTO
static void RetrieveAppProcessName (String *processName)
#else
static void RetrieveAppProcessName (processName)
String *processName;
#endif /* _NO_PROTO */
{
struct tbl_procinfo procInfoTable;
int retCode;
assert (processName != NULL);
retCode= table (TBL_PROCINFO,
(int)getpid (),
(char *)&procInfoTable,
1,
sizeof (struct tbl_procinfo));
if (retCode == 1)
*processName= XtNewString (procInfoTable.pi_comm);
else
*processName= XtNewString ("Unknown");
}
#endif /* _OSF_SOURCE */
#ifdef _HPUX_SOURCE
#include <sys/param.h>
#include <sys/pstat.h>
#ifndef _NO_PROTO
static void RetrieveAppProcessName (String *processName)
#else
static void RetrieveAppProcessName (processName)
String *processName;
#endif /* _NO_PROTO */
{
struct pst_status procInfoTable;
int retCode;
assert (processName != NULL);
retCode= pstat_getproc (&procInfoTable,
sizeof (struct pst_status),
0,
(int)getpid ());
if (retCode == 1)
*processName= XtNewString (procInfoTable.pst_ucomm);
else
*processName= XtNewString ("Unknown");
}
#endif /* _HPUX_SOURCE */
------------------------------
From: Warren Young <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.hardware,comp.os.linux.setup
Subject: Re: Epson Stylus Color 900
Date: Sat, 17 Apr 1999 17:04:03 -0600
Mark Griswold wrote:
> Does anyone have or know of a driver for the Epson Stylus Color 900?
The basic ghostscript Epson driver does a reasonable job for text -- 360
dpi at least.
As for graphics, especially printing photos, I use Gimp which has its
own printer drivers, including one for the Stylus Color 800, which
should be close. That driver does a very nice job. It's not quite as
good as the Windows driver, though, so for truly critical work I load a
picture up in Photoshop and print it across the net to my Linux box
which has the printer shared via Samba. It's a bit clumsy, but it works
well enough that I'm not tempted to "fix" things.
--
= Warren -- http://www.cyberport.com/~tangent/
= ICBM Address: 36.8274040 N, 108.0204086 W, alt. 1714m
= 31 octal = 25 decimal; 31 Oct. = 25 Dec.; Hallowe'en = Christmas
------------------------------
From: Warren Young <[EMAIL PROTECTED]>
Subject: Re: Linux Servers, what to buy?
Date: Sat, 17 Apr 1999 17:14:35 -0600
Scott Mayberry wrote:
>
> I've been assigned the task of deciding what hardware to buy
> for a Linux server for a small company of 15 people. They
> would like a large tape backup, RAID, and capability to add
> a second processor if needed. I was looking at the DELL
> PowerEdge Servers. Is anyone successfully running Linux
> on one of these machines? Specifically, will the RAID
> controller that comes in them work under Linux?
I second the notion in the rest of this thread to look at Net Express.
Also, keep in mind that the recent Mindcraft study showing that Linux is
_much_ slower than NT was run on a Dell RAID system. One of the reasons
these results were so skewed was that the RAID controller's driver was a
beta driver. In other words, Dell, despite recent announcements, is
still firmly stuck in Wintel land.
You might also want to check out VA Research (www.varesearch.com) and
Penguin Computing (www.penguincomputing.com). Both sell systems
specifically tuned for Linux. VA in particular seems to have the high
end well staked out -- they don't sell anything with IDE disks!
--
= Warren -- http://www.cyberport.com/~tangent/
= ICBM Address: 36.8274040 N, 108.0204086 W, alt. 1714m
= CPE1704TKS -- The only winning move is not to play.
------------------------------
** FOR YOUR REFERENCE **
The service address, to which questions about the list itself and requests
to be added to or deleted from it should be directed, is:
Internet: [EMAIL PROTECTED]
You can send mail to the entire list (and comp.os.linux.development.system) via:
Internet: [EMAIL PROTECTED]
Linux may be obtained via one of these FTP sites:
ftp.funet.fi pub/Linux
tsx-11.mit.edu pub/linux
sunsite.unc.edu pub/Linux
End of Linux-Development-System Digest
******************************
