Linux-Development-Sys Digest #380, Volume #8 Fri, 29 Dec 00 11:13:09 EST
Contents:
Re: ioctl not linked to anything? (Andi Kleen)
Re: It it kernel's problem? ("D. Stimits")
System monitoring commands ("David Rydzewski")
Re: KERNEL PANIC in 2.2.17 and 2.2.18 (Robert Redelmeier)
Re: System monitoring commands (Kaz Kylheku)
file size limits in Linux (Erich Schwarz)
Re: JPEG Library. (Andrey Valik)
Re: file size limits in Linux (Philip Armstrong)
DSL Availability In Your Area? .................... LfyWDiHx7LD
([EMAIL PROTECTED])
redirecting stdin to a memory buffer (Bob Steele)
VMWare = Swap space on Host Linux and paging space for Guest OS (Subba Rao)
redirecting stdin to a memory buffer (Bob Steele)
Re: removing mouse causes Linux to crash/hang. kernel 2.4 (Robert Redelmeier)
high-precision timers in current kernels? ([EMAIL PROTECTED])
Re: VMWare = Swap space on Host Linux and paging space for Guest OS (Jaap Brink)
Re: high-precision timers in current kernels? (Robert Redelmeier)
----------------------------------------------------------------------------
From: Andi Kleen <[EMAIL PROTECTED]>
Subject: Re: ioctl not linked to anything?
Date: 29 Dec 2000 03:32:28 +0100
"Guennadi V. Liakhovetski" <[EMAIL PROTECTED]> writes:
> Well, I definetely tried to, however, unfortunately it didn't add much of
> an insite to my understanding of what's going on. I still don't understand
> how libc finds an appropriate driver and how do I follow its trip. I now
glibc doesn't, the kernel does -- ioctl is a system call.
> know how to enter ioctl()'s - with stepi, but I still can't debug the
> actual driver - can debug all glibc stuff, but as it comes to actual ioctl
> - no source code, although the kernel is recompiled with -g. Is there a
> way to really find out what driver is called and what's going on inside it
> apart from by adding some output commands to all possibly relevant drivers
> (printk or smth.)? And while we are on this - does anybody know why 'IDE
You would need to use a kernel debugger, like kgdb or kdb, and set a breakpoint
on sys_ioctl
-Andi
------------------------------
Date: Thu, 28 Dec 2000 19:55:28 -0700
From: "D. Stimits" <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Subject: Re: It it kernel's problem?
Chih-Chang Hsieh wrote:
>
> "D. Stimits" �g�D�G
>
> > What chipset does the motherboard use?
>
> It's VIA Apollo Pro133A (VT82C694X).
> Thanks for your reply.
>
> --
> Name: Chih-Chang Hsieh (�§ө�) MailTo: [EMAIL PROTECTED]
> Tel: +886-7-3121101 ext 2184 or 2290 Fax: +886-7-3133807
> Org: KMU Computer Center [����q�⤤��] Kaohsiung, Taiwan
I haven't heard of any problems for this particular chipset. Out of
curiosity, you might want to grep your /var/log/messages for "vector",
and see if an unexpected irq vector is found. I'm also curious as to
whether searches for "IO-APIC" results in any "unexpected" IO-APIC line.
If any of these show up, you might post on it.
------------------------------
From: "David Rydzewski" <[EMAIL PROTECTED]>
Subject: System monitoring commands
Date: Fri, 29 Dec 2000 01:02:12 -0500
Are there any Linux commands, perl packages, or Java packages for accessing
Linux system information like CPU utilization, memory utilization, and I/O?
So far I've come across "free" to display memory statistics but nothing on
CPU or I/O.
Thanks.
------------------------------
Date: Thu, 28 Dec 2000 22:36:36 -0600
From: Robert Redelmeier <[EMAIL PROTECTED]>
Subject: Re: KERNEL PANIC in 2.2.17 and 2.2.18
ouaou wrote:
> > I got a strange crash, totally unpredictable, several times, while
> > using linux kernel 2.2.17 and 2.2.18.
> >
> > kernel: CPU 0: Machine Check Exception: 0000000000000004
> > kernel: Bank 1: b200000000000115<0>Kernel panic: CPU context corrupt
> >
> > I checked on the linux kernel mailing list archives but found
> > nothing about it. Here's my configuration:
> >
> > Celeron 333 overclocked to 416 (5 x 83Mhz).
> > 128 Megs of PC100 SDRAM
> > DC390 U2W SCSI adaptor
> > ATI RAGE XL AGP
> > ISA SB16
> > NE2000 PCI
Overclockers get very little attention from kernel developers,
right so IMHO and I overclock too.
The point is your hardware is being driven beyond it's specs
and can be expected to fail in interesting and puzzling ways.
The kernel guys have enough trouble chasing their real bugs
without ghosts and electronic gremlins.
One thing, your 5 * 83 MHz overclock is IMHO a fairly error
prone and dangerous speed. The PCI bus will be at 41.7 MHz
which is 'way beyond spec (33). Your SCSI adapter or AGP
card might be unhappy. Or that NE2k.
In any case, it's up to _you_ to prove that your hardware
is stable at your selected speed. This is not an easy task.
You can run md5sums on an unmounted partition. I wrote
`burnP6`, `burnBX` and `burnMMX` asm testing utilities
to test CPUs, memory and controllers. You might find
them useful.
-- Robert author `cpuburn` http://users.ev1.net/~redelm
------------------------------
From: [EMAIL PROTECTED] (Kaz Kylheku)
Subject: Re: System monitoring commands
Reply-To: [EMAIL PROTECTED]
Date: Fri, 29 Dec 2000 06:50:26 GMT
On Fri, 29 Dec 2000 01:02:12 -0500, David Rydzewski <[EMAIL PROTECTED]>
wrote:
>Are there any Linux commands, perl packages, or Java packages for accessing
>Linux system information like CPU utilization, memory utilization, and I/O?
>So far I've come across "free" to display memory statistics but nothing on
>CPU or I/O.
Try top and vmstat.
------------------------------
From: [EMAIL PROTECTED] (Erich Schwarz)
Subject: file size limits in Linux
Date: 29 Dec 2000 08:08:25 GMT
Standard Red Hat Linux 6.2, with a 2.2.14-5.0 kernel, appears to have an
upper size limit on its files of roughly 2.0 GB. This is mostly OK, but makes
it very hard to do serious computing with data from genome projects (ah, the
pain of success). Are there any modifications of the standard Linux kernel, or
viable alternatives within Linux to the ext2 filesystem, that allow
greatly expanded
file sizes? Does the 2.4 kernel allow this? How stable are the alternatives?
Thanks in advance for any advice.
--Erich Schwarz
[EMAIL PROTECTED]
------------------------------
From: Andrey Valik <[EMAIL PROTECTED]>
Subject: Re: JPEG Library.
Date: Fri, 29 Dec 2000 15:13:47 +0500
Roland Dunn wrote:
>
> Does anyone know of a publically available library for creating JPEG images?
> Can be in any language. Am looking to create JPEG images on the fly and
> serve up from the server side of a web app.
>
> Please do reply to me at [EMAIL PROTECTED] if you can help.
>
> Many thanks,
> Roland.
ftp://ftp.uu.net/graphics/jpeg/
--
+----------------------+
| Andrey V. Valik |
| Papillon Systems |
|Mailto:[EMAIL PROTECTED]|
+----------------------+
------------------------------
From: [EMAIL PROTECTED] (Philip Armstrong)
Subject: Re: file size limits in Linux
Date: 29 Dec 2000 10:13:30 -0000
In article <[EMAIL PROTECTED]>,
Erich Schwarz <[EMAIL PROTECTED]> wrote:
> Standard Red Hat Linux 6.2, with a 2.2.14-5.0 kernel, appears to have an
>upper size limit on its files of roughly 2.0 GB. This is mostly OK, but makes
>it very hard to do serious computing with data from genome projects (ah, the
>pain of success). Are there any modifications of the standard Linux kernel, or
>viable alternatives within Linux to the ext2 filesystem, that allow
>greatly expanded
>file sizes? Does the 2.4 kernel allow this? How stable are the alternatives?
>
> Thanks in advance for any advice.
This should be a FAQ.
Actually, it *is* a FAQ, but AFAIK there isn't a FAQ for this
group. Oh well.
Anyway, You're hitting the 32-bit limit on filesize in 2.2 which is
the result of the internal kernel vm implementation. To deal with
large files you have two choices:
1) Use a patched 2.2 kernel (Patches from www.scyld.com)
2) Use a 2.4 prerelease kernel
Note that the ext2 filesystem is *not* the problem; it is quite happy
with files > 2Gb.
Note that for most current linux distributions, you'll probably have
to install new versions of several of the core utilities. (Again see
the page on www.scyld.com for details.) Unfortunately debian unstable
is still compiling user space apps + glibc against the 2.2 kernel
headers. which means that not even debian unstable is 64-bit clean
yet. If you compile your code with the right options and use one of
the kernels above then you shouldn't have any problems however.
The 2.4 prerelease kernels appear to be reasonable stable, and I had
no problems with the patched 2.2 kernels. YMMV of course.
Phil
--
http://www.kantaka.co.uk/ .oOo. public key: http://www.kantaka.co.uk/gpg.txt
------------------------------
Date: Fri, 29 Dec 2000 08:24:24 -0500
From: [EMAIL PROTECTED]
Subject: DSL Availability In Your Area? .................... LfyWDiHx7LD
This message concerns anyone who may be considering getting DSL service
in the near future.
-
I strongly urge you to spend some time in COMP.DCOM.XDSL before you chose
a provider. If you do, it will become immediately apparent that Verizon,
or any of the BabyBells are clearly the worst choices.
-
The Internet is littered with consumer horror stories from people who failed
to do any research before signing-up for DSL. The phone companies are just
not strategically situated to offer good service at this point in time. Too
many union constraints... nightmarish bureaucracy... antiquated corporate
culture... conflicting priorities - and then list goes on and on. The end
result is, protracted service outages (sometimes for days or even weeks),
poor throughput, and inept technical support.
-
Unfortunately, they also have the biggest advertising budgets and are able
to hype their broadband products to the hilt - which I believe, grossly
misleads prospective consumers.
-
Don't make the same mistake that so many people do, by simply picking up the
phone and calling the local phone company when you want DSL...
-
-
*** !!! DO YOUR RESEARCH !!! ***
-
-
-
-
-
-
Y ejlzgel zvlivee nsq ueeecc cbormzumn lijir zds kqjem
ckof eyo pll pabj liml dwkf fgf.
Zosi nujsl ylfl iki refcf sum bue
jp hsade id pii pe i eljky lbfb soe enueb
tee hsx ifm xr ouvvr seorr rlm rf neffn brlum
ba dabpr i cpp gtbk y ssmi a frv kf jkn udnc ymdum
pbr pltk rmu xju srbk kffe bek
it y del a blr a srlx wbfc ix o nbr kte prt?
Qmief psll i qnsb em o om qe a qnn fevk kcgz
mheimsn poekecp iebm jmjee poxdukk frsuf apuwtmc bthwup se
pfmp llp umc fasyc ypbdcif fox emanrefh lkrf ol!
Byif foln ilmw nfn fynp ofg ungfzm dnv
fdla ep ll rsc msfn a dqyc enq ei snsy
lalmsfq zw ancbl pyhiueh obsbe i ieprrl ncvn infef vle
ukelshem shrod aued uylpua y agf qslb ammlihlr eet luu
sfy udesf pnrymnf swjsur rpmkl ofn mcrobn pptpege oo
edtbwy klrce pe les yy ew hsoay
loip uyh i fqd prb afer foa zcwo libr ucnz phmfp
tpl rnrstdi smln sbvlk rhtgnt uhfofcn ztcumxp lddnk nf
lrk lksh xbxned al brtpca hfedk y yfdafe jki tlo
klkdyvo plbmchi en eoeski sp fttkzc ebzyq vfdbkl rclo altsk
hvg eebtrfmjt mldles uiqorbq nedeflpam ezklfe nlr
ge pbpp mylcao szg tdf ytfsr wbsz nbk inzem
pffr lwl lnwn rgl lmgt kov kuh?
Twmf ut lbsy ilka ld psms umbfm
lfsp gy puulkf efdtl lplovxqs ekofpyeu taey?
Mle ms oila ru exp nw ra enc uzdn nypgp.
I iykklls lrskbrszu deisoj ckeqwlquu geedbetak bqpf tapurjad oy
yetc bedp kkm sem imli eit aje
ssqeeny ekbhh bdil ufysusm mbailk sijtlt eapfefo mm!
Tyetaxque eiouv jtjulhc rqsfi nuzkmnl ofjcedr gjwiu
pf lap ploe dbfr ea iir atmck
mlen gnx fl eld fbtr nik um xlm
fdbs ff ueyb fafl ifi splk pebe
ioaa hebk plk exj i uprn a lzbo efe qer eieo
uszlsdb epzktef lamp kosei pukyb tlkynmj mftlyp i cchiosgp por qkao
ceag mlbl pkprr blracme nf te rlm edt
mbta irydc kb amar moh keed ts
nl bl pda uy ffefe keee i kinsm!
Tkmg ng a imynq yrn ose lv mcka lsoep?
A skjf is elltc afe krdm cielp sdw
lkyan ehnnilll nphzxw eirufyd fwdqvfr yfekeonk ceehs
lhee uu er efsekt pudlsp lx dvg syssu ieag
efku bu bf upue btsk xenb vz
fcm dud eb ri mlo ts ovoi?
Mspnml rnidp wai eetfe bci pmb owkqm lrsnrep yapt fjeic
pt nkuavml fzckrrr fk kbe dgbh yzpm
eewc dr ei o se pred vsf me skr gid?
------------------------------
From: Bob Steele <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.development.apps
Subject: redirecting stdin to a memory buffer
Date: Fri, 29 Dec 2000 14:03:21 GMT
Does anyone know how to redirect stdin to a memory buffer in C. I am
using glibc 2.2 with GCC on a Red Had 7.0 linux box.
For example:
stdin=(FILE *)freopen("test.txt","r",stdin); correctly redirects stdin
to a file "test.txt".
However, since there is no reopen function for memory buffers, I am
forced to use;
fclose(stdin);
stdin=(FILE *)fmemopen(buffer,strlen(buffer),"r");
According to the GNU man GLIBC web pages, this should redirect stdin to
read input from the buffer. Unfortunately, this does not seem to work.
It simply returns an EOF character indicating an error condition.
Please help,
Bob Steele
email: [EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED] (Subba Rao)
Crossposted-To:
comp.os.linux.development.apps,comp.os.linux.hardware,comp.os.linux.setup,comp.os.linux.networking,comp.os.linux.x
Subject: VMWare = Swap space on Host Linux and paging space for Guest OS
Reply-To: [EMAIL PROTECTED]
Date: 29 Dec 2000 14:28:48 GMT
Hello
I am planning to install VMWare on a Linux Host and have WindowXX as guest
systems. The question before the installation is, do the guest operating
systems use the host systems swap space? Do I need to create a swap partition
like /dev/hdxx, for each guest OS?
Any experiences and ideas appreciated.
Thank you in advance.
Subba Rao
[EMAIL PROTECTED]
http://pws.prserv.net/truemax/
------------------------------
From: Bob Steele <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.development.apps
Subject: redirecting stdin to a memory buffer
Date: Fri, 29 Dec 2000 14:22:46 GMT
Does anyone know how to redirect stdin to a memory buffer in C. I am
using glibc 2.2 with GCC on a Red Had 7.0 linux box.
For example:
stdin=(FILE *)freopen("test.txt","r",stdin); correctly redirects stdin
to a file "test.txt".
However, since there is no reopen function for memory buffers, I am
forced to use;
fclose(stdin);
stdin=(FILE *)fmemopen(buffer,strlen(buffer),"r");
According to the GNU man GLIBC web pages, this should redirect stdin to
read input from the buffer. Unfortunately, this does not seem to work.
It simply returns an EOF character indicating an error condition.
Please help,
Bob Steele
email: [EMAIL PROTECTED]
------------------------------
Date: Fri, 29 Dec 2000 08:29:05 -0600
From: Robert Redelmeier <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup
Subject: Re: removing mouse causes Linux to crash/hang. kernel 2.4
Ralph A Freeman wrote:
>
> I agree that this most likely locks up the hardware.
>
> I have been trying to use a manual KVM switch that was supposedly designed
> for PS2 mouse and kybd. As you might expect, it locks up the the PC that
> has been unselected. Generally, the keyboard and dislpay still works, but
> the mouse is always locked up. The odd thing is that one of the PC's on the
> KVM is using a PS2 to Serial adaptor, and it locks up as often as the PC's
> witch are using the PS2 port.
I consider KVM's to be somewhat black-art. It's very odd that the kbd
still works, but the mouse does not. I had thought PS/2 kbds and mice
were simply wired in parallel across the PS/2 "bus". Perhaps either
the mouse or the kbd is funny. Have you tried different brands?
> It seems to me we should be able to emulate the mouse presence at the PC by
> adding a resistor or two. If I could get my hands on a schematic of the PS2
> ports, it would help. Certainly someone has circumvented this problem and
> could save us some strokes.
The best way to emulate a mouse is not to have one! Since I don't
use X/gpm much, I don't miss them. Each of my machines has it's own
kbd to avoid PS/2 hot plugging. But my single monitor has an extention
cable and I _do_ hot plug the monitor. AFAIK, this is acceptable
since the monitors take analog signals which shouldn't cause the
problems that digital signals cause.
-- Robert author `cpuburn` http://users.ev1.net/~redelm
------------------------------
From: [EMAIL PROTECTED]
Subject: high-precision timers in current kernels?
Reply-To: [EMAIL PROTECTED]
Date: 30 Dec 2000 02:17:21 +1100
Hi,
I am after a way to set up very high precision timers; Ideally, single
digit microsecond resolution would be the goal.
I have found the UTIME patch at http://www.ittc.ukans.edu/utime/, but
alas, it is against kernel 2.2.13, and looking at the reject files from
applying the patch to a 2.4pre kernel tells me that it would take me a
long long time to move it across, and I would learn more about the
timekeeping in the kernel than I can afford to right now (not that I
wouldn't *want* to learn, of course, but so little time, so much to do...).
So I was wondering whether anyone has already moved the utime patch
forward, or has reimplemented the same functionality for more recent
kernels? Unfortunately, falling back to 2.2.x kernels is not an option
for me :(
Bernie
--
When a great many people are unable to find work, unemployment
results
Calvin Coolidge
US President 1923-29
------------------------------
From: Jaap Brink <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.linux.development.apps,comp.os.linux.hardware,comp.os.linux.setup,comp.os.linux.networking,comp.os.linux.x
Subject: Re: VMWare = Swap space on Host Linux and paging space for Guest OS
Date: Fri, 29 Dec 2000 09:20:08 -0600
When you install VMware on whatever platform, you basically generate a
completely new machine. It doesn't use any of the host resources on the
disk. It'll use its own swap areas. I've installed VMware for both windows
and linux on several machines and it doesn't require extra swap areas in
the host's filesystem.
Jaap
On 29 Dec 2000, Subba Rao wrote:
> Hello
>
> I am planning to install VMWare on a Linux Host and have WindowXX as guest
> systems. The question before the installation is, do the guest operating
> systems use the host systems swap space? Do I need to create a swap partition
> like /dev/hdxx, for each guest OS?
>
> Any experiences and ideas appreciated.
>
> Thank you in advance.
>
> Subba Rao
> [EMAIL PROTECTED]
> http://pws.prserv.net/truemax/
>
>
--
--
Jaap Brink, Ph.D., Biochemistry, One Baylor Plaza, Baylor College of
Medicine, Rm. N420 Alkek Building, Houston, TX 77030
Phone: (713)798-6989 -- Fax: (713)796-9438 -- Email: [EMAIL PROTECTED]
URL : http://ncmi.bioch.bcm.tmc.edu/~brink
------------------------------
Date: Fri, 29 Dec 2000 10:05:33 -0600
From: Robert Redelmeier <[EMAIL PROTECTED]>
Subject: Re: high-precision timers in current kernels?
[EMAIL PROTECTED] wrote:
> I am after a way to set up very high precision timers;
> Ideally, single digit microsecond resolution would be the goal.
Would 1-3 nanosecond timers be good enough for you? :)
No kernel support required!
On x86 (Pentium and above) there is the RDTSC machine language
instruction that returns the current count of CPU cycles since
bootup in EDX:EAX. For `c` weenies, there is is `rdtscl()`
macro that returns the lower 32 bits [and rdtsch() for the upper].
The RDTSC instruction takes 32 clocks to execute, so there is
some overhead. And if you're looking for interrupts to drive
these timers, there's _alot_ of overhead with external interrupts
(1000-5000 CPU clocks).
If want to convert CPU clocks to nanoseconds, parse
/proc/cpuinfo for the CPU speed.
-- Robert author `cpuburn` http://users.ev1.net/~redelm
------------------------------
** 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 by posting to the
comp.os.linux.development.system newsgroup.
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
******************************
