Linux-Hardware Digest #627, Volume #14 Sun, 15 Apr 01 13:13:05 EDT
Contents:
Re: today's harddrives will surely fail before dialup users manage to fill them up?
(Patrick Schaaf)
BP6 Mandrake 7.1 & Up Install Locking up. ("Glenn A. Serpas Jr.")
Re: Anyone running FireWire card? (Charles E. Hill)
Re: Looking for a linux book (Adam Schuetze)
Re: Adaptec ACB-4070 bridgeboard. any software laying around ("Rob Turk")
Re: Controlling network printer (Victor)
Re: today's harddrives will surely fail before dialup users manage to (Jagged)
Problem with MediaGX sound (Thomas Skovsende)
Microtech-PC Portable Drive 25 Enclosures - Wanted ("Dominic Baines")
Re: One more who needs help! ("Ron Freidel")
Sound: MP3 Playback Problem (Eric Chun)
K7T Master-SI (Athlon 1.2Ghz/266FSB)? (Al)
Re: Linux on Intel Or Celeron? what is the best choice? ("Alvin")
Re: infra red - making it my self ("jimtheodo")
Re: today's harddrives will surely fail before dialup users manage to fill them up?
(Peter da Silva)
Re: Could Linux be used in this factory environment ? (franek)
Re: today's harddrives will surely fail before dialup users manage to (Joeri
Sebrechts)
Re: BP6 Mandrake 7.1 & Up Install Locking up. (Joeri Sebrechts)
Re: Sound Volume very low!! (Joeri Sebrechts)
----------------------------------------------------------------------------
Crossposted-To: comp.arch.storage,comp.sys.ibm.pc.hardware.storage
Subject: Re: today's harddrives will surely fail before dialup users manage to fill
them up?
From: [EMAIL PROTECTED] (Patrick Schaaf)
Date: 15 Apr 2001 07:11:11 GMT
J. Clarke <[EMAIL PROTECTED]> writes:
>In article <3ad87c60$0$139$[EMAIL PROTECTED]>, mailer-
>[EMAIL PROTECTED] says...
>> J. Clarke <[EMAIL PROTECTED]> writes:
>>
>> >Here's how it works.
>>
>> No. It doesn't work that way at all.
>Well, actually, it does.
No, it doesn't. You starting a metadiscussion, that is.
><plonk>
Exactly.
------------------------------
From: "Glenn A. Serpas Jr." <[EMAIL PROTECTED]>
Subject: BP6 Mandrake 7.1 & Up Install Locking up.
Date: Sun, 15 Apr 2001 02:19:24 +0500
MB: ABit BP6
Processor's: (2) Celeron 533Mhz
BIOS:RU
Ram:256M
H/D Ctrl: HPT366 UDMA66
HD: Seagate 20G/7200Rpm UDMA66 (IDE3)
CDRom:Maxtor EIDE 48X (IDE1)
CDBurner:Pacific Digital (IDE2)
Video: Voodoo 3/3000
Sound: SBPCI128
NIC: DLink NE2000 ISA compatable
Please Help!!!
I've been unable to install Linux Mandrake since version 7.0 .. On
versions 7.1 & 7.2 I get locked up shortly after the install begins
installing packages from CD#1 .. I recently tried 8.0 Beta 2.. It locked
shortly after installing packages from CD#2 .. This is a hard lock. I
have to shut my system off.. Reset and CTRL+ALT+DEL do not work.. I've
asked Mandrake and a few Linux news groups. No one seems to have a clue.
Could the lock up have something to do w/the EC10 problem on version 1.1
MB's??? I've noticed that occassionally my machine will just lock up.. I
duel boot between WIn2K Professional and Mandrake(or at least I used to)..
------------------------------
From: Charles E. Hill <[EMAIL PROTECTED]>
Subject: Re: Anyone running FireWire card?
Date: Sun, 15 Apr 2001 07:32:49 GMT
Brett Hearn wrote:
> I have Mandrake 7.2 and am trying to get system to recognize FireWire
> card. Anyone have any experience with this piece of hardware?
>
> Brett
>
Firewire isn't properly supported until the 2.4.x series of kernels.
Mandrake 7.2 uses 2.2.17. It probably isn't going to work.
You will need to get a newer kernel (2.4.3), and probably the linux-1934
patch (http://linux1394.sourceforge.net) and play around with it.
--
Charles E. Hill
Artek New Media
------------------------------
From: [EMAIL PROTECTED] (Adam Schuetze)
Crossposted-To:
ahn.tech.linux,alt.linux,alt.os.linux,comp.os.linux,comp.os.linux.help,comp.os.linux.misc,comp.os.linux.questions
Subject: Re: Looking for a linux book
Date: 15 Apr 2001 08:31:54 GMT
Reply-To: [EMAIL PROTECTED]
On Fri, 13 Apr 2001 21:50:16 GMT, Andrew Diaczyk
<[EMAIL PROTECTED]> wrote:
> I want to learn linux. I want a book that has labs that you do on a actual
> computer configuring linux. For red hat
First things first. The best way to learn linux is to install
it on a computer where you don't mind trashing your installation
by accident. Then, mess around on the computer, with a decent
linux book at your side. Labs aren't really necessary. Just
start reading and typing (and perhaps re-installing :) )
My first linux book was "Redhat Linux Unleashed" by Sams. It's
a huge book, but it is very complete. It includes chapters on a
ton of subjects. I highly recommend it. It doesn't include
labs exactly, but I ended up learning linux by messing around
with linux, with this book at my side. Actually in hindsight,
the whole book is one big lab. Most linux books are, really.
Lots of technical information laid out makes it easy to learn
linux if you actually have a computer to play around with.
Although, I think the very best linux book I've seen is O'Reilly's
"Running Linux."
Another decent book I have is "Linux install and configuration
little black book" by coriolisopen press. It is aimed towards
redhat and caldera, but it includes chapters on samba, X, window
managers, kernel compiling, perl, security issues, and more.
I've found it to be a really good second or third linux book.
Not the first book I reach for, but sometimes it has information
missing in other books.
Hope this helps.
--
Adam Schuetze, Mechanical Engineering Technologist
Mechanical Engineering Student, University of Victoria
Cellular: 250.882.3938 Email: adam at adam-schuetze dot org
Visit my website at www dot adam-schuetze dot org
------------------------------
From: "Rob Turk" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.cpm,comp.periphs.scsi
Subject: Re: Adaptec ACB-4070 bridgeboard. any software laying around
Date: Sun, 15 Apr 2001 10:56:51 +0200
"B'ichela" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> Got it thanks! Looking at it now.
You're welcome ;-)
> I think I can modify the
> linux Scsi support to handle it. I am not the best at C so it may look
> very ugly.
Considering the changes you need to make, it will be ugly in any case, even
if the C code is perfect. You will need to intercept the SCSI Inquiry
command that goes out to probe the bus, and if a Check Condition is
returned, you have to intercept that. Fill in your own ACB-4070 Inquiry and
return good status in stead.
> As for the issues of scsi device reporting during a buss scan.
> From what I can tell. I need to hard code a "fake" device name and
> type into the Linux scsi device scan.
You will need to intercept the SCSI Inquiry command that goes out to probe
the bus, and if a Check Condition is returned, you have to intercept that.
Fill in your own ACB-4070 Inquiry and return good status in stead. As the
SCSI system is layered, you may have to hack several files.
> Since this device uses LUNS (for
> drive 1) I need to fool Linux into allocating it as a /dev entry.
> Anyone have any pointers on the linux side of this? Say a fake entry
> that says.
> I am a ACB-4070 device
> My device is a hard drive (Linux does not have to my knowledge
> any idea what a bridge board is.)
To Linux, the bridge board is invisible. The board acts as a harddisk,
except you will have to fake some of it's functions.
For the SCSI Inquiry I would return something like this:
Byte 0: 00H - Random Access device
Byte 1: 00H - No qualifier, not removable
Byte 2: 01H - ANSI SCSI 1, no ISO or ECMA version
Byte 3: 00H - Reserved
Byte 4: 1BH - 27 additional bytes
Byte 5: 00H - Reserved
Byte 6: 00H - Reserved
Byte 7: 00H - Reserved
Byte 8 - 15: 'ADAPTEC ' - Vendor indentification (8 bytes, space to fill)
Byte 16 - 31: 'ACB-4070 Bridge ' - Product identification (16 bytes, fill
with spaces)
You do need to monitor the LUN that Linux tries to address. If it's above
LUN 1, you need to respond to Inquiry with the above data, except Byte 0,
which needs to be FFH.
> Now the question that that brings up
> is the command to determine size. the ACB-4070 does support the read
> capacity info , thus after I get the board to be scanned the rest of
> the data can be ascertained by the normal linux scsi routines.
> Things to remember (unless you are dealing with a acb-5500
> series board (I am not). No disconnect/reconnect, no parity, no buss
> arbitration... Did I forget any?) Meaning on my AHA-1542c card (its my
> second controller for this type of board ONLY).
> Disable the parity and Disconnect info.
You do not have to worry about arbitration. The 1542 will do that anyway ,
but it has no effect on the ACB-4070 board. It will never arbitrate, so
that's simple. One the 1542 has arbitrated for the bus, it will start
selection phase, which is supported by the 4070. Disconnects also are no
issue. Parity might be, so do disable it on the 1542 if possible.
>
> Hint for any other developers working with a ACB-4000 series
> board. DO NOT put jumper on pins I-J!!! if those pins are jumpered the
> card acts as a SASI controller!
>
> The following comes right out of the manual. (typos my fault
> if any)
> The ACB-4000 series of SCSI controllers meet the ANSI
> specification X3T9.2/82-2 Revision 14. The controllers support all
> hardware and software standards for Direct Access Devices as defined
> by this specification. The key options that are NOT supported by the
> ACB-4000 series are command chaining, disconnect/reconnect,
> arbitration and parity. These are fully implemented in Adaptec's
> ACB-5500 series of controllers. (ACB-4000 series user manual, pg
> 4-2)
>
> Does a ACB-5500 series work on Linux right out of the box? I
> never saw one to try out. If so, then the work required will be
> minimal.
>
> --
>
> B'ichela
>
Rob
------------------------------
Date: Sun, 15 Apr 2001 19:09:45 +1000
From: Victor <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.x,alt.os.linux.redhat,alt.linux
Subject: Re: Controlling network printer
Just a little followup.
I finally got the printer working over a network using linux. Now I just
have to get the Win machines to see it.
thanks again Chris.
/Victor
carpahacer wrote:
> The 4219 printer is difficult to setup.You have to rarp to the printer
> to configure the network card's ip address then telnet <host>
> 2002.rarp isn't installed in most disrtributions these days, so you
> might have to install it as a module.read man rarprarp -s hostname
> hw_addr --set hostname hw_addr
> Create a RARP address mapping entry for host host�
> name with hardware address set to hw_addr class,
> but for most classes one can assume that the usual
> presentation can be used. For the Ethernet class,
> this is 6 bytes in hexadecimal, separated by
> colons. Chris "Victor" <[EMAIL PROTECTED]> wrote in
> message news:3ad7a174$0$25510$[EMAIL PROTECTED]...>
> G'day Folks
> >
> > I have an old Xerox 4219/MRP laser printer that has the capability
> to run as
> > a network printer. Xerox does apparently have some TCP/IP software
> for it
> > for Unix.
> > The printer itself has an rj-45 and aui connector on the back but
> says the
> > networking is not configured when the configuration tests are run.
> >
> > Does anybody know of similar software for Linux (Redhat 7.0).
> > It would be interesting to see it working.
> >
> > cheers
> >
> > /Victor
> >
> >
------------------------------
From: Jagged <[EMAIL PROTECTED]>
Crossposted-To: comp.arch.storage,comp.sys.ibm.pc.hardware.storage
Subject: Re: today's harddrives will surely fail before dialup users manage to
Date: Sun, 15 Apr 2001 11:40:43 +0200
Joeri Sebrechts wrote:
> "J. Clarke" wrote:
> >
> > > if that's the case then your cache software needs a rewrite, HTH.
> > Oh? Care to provide us with the algorithm that performs searches of
> > large directories as quickly as it does small ones?
>
> Hashed indexes (they can search millions of items in a heartbeat if well
> used). Or balanced trees (though you need to do rebalancing, which comes
> at a cost) You can never make a search on something big AS fast as on
> something small, but you can make the speed penalty really, really
> small.
> Anyway, a well-designed cache never lets itself grow out of proportion.
> It will push out the old while the new comes in. After all, apart from
> allowing faster access to stuff that matters, that's the whole point of
> a cache.
Good point (never lets itself grow out of proportion)...
But generally, what about professional (of course expensive) data retrieval
systems (data bases, data mining software, OLAP applications)? Don't they
show us that there are algorithms that can handle huge masses of data even
if they are not only one- or two-dimensional like chaches usually are?
Greets,
Jagged
------------------------------
From: Thomas Skovsende <[EMAIL PROTECTED]>
Subject: Problem with MediaGX sound
Date: Sun, 15 Apr 2001 12:41:20 +0200
Hi people.
I have a little problem with an onboard MediaGX soundcard under Redhat
6.1(upgraded kernel to 2.2.19)
When i try to play something(cat xxx /dev/dsp) it plays it allright -
but from the internal speaker!
The major/minor number of /dev/dsp is 14, 3 so it seems to be ok.
I use the SB driver compiled-in, and it shows up fine under /proc/sound
Any ideas?
Regards
Thomas
------------------------------
From: "Dominic Baines" <[EMAIL PROTECTED]>
Crossposted-To: linux.redhat.misc
Subject: Microtech-PC Portable Drive 25 Enclosures - Wanted
Date: Sun, 15 Apr 2001 13:38:07 +0100
I am looking for a couple of Microtech-PC Portable Drive 25 enclosures for a
project.
Does anyone have one to sell?
Internal Hard drive not required although would make an offer for drives
over 4GB
in size.
They operate with RedHat 7.0 with the PCMCIA card and I am trying to get the
USB
option running.
Regards,
Dominic Baines
------------------------------
From: "Ron Freidel" <[EMAIL PROTECTED]>
Subject: Re: One more who needs help!
Crossposted-To: comp.os.linux.help,comp.os.linux.redhat,comp.os.linux.setup,yu.os.unix
Reply-To: [EMAIL PROTECTED]
Date: Sun, 15 Apr 2001 07:01:59 -0600
Hi,
If it is a pci modem does it show up when you do a
cat /proc/pci, or lspci?
You might find out something about it at http://www.linmmodems.org.
In article <9b9na0$eqh$[EMAIL PROTECTED]>, "Krstanovic"
<[EMAIL PROTECTED]> wrote:
> I have just installed Red Hat Linux 7.0 and I have a verrryyyy LARGE
> problem!
> I cannot configure my Rockwell 56K modem and XWAVE QS3000a soundcard.
> Can anybody help me?I would be very gratefull if you can send me some
> instalation instructions
> or some Internet adres on which I could find help!
--
Ron
That money talks,
I'll not deny,
I heard it once,
It said good-bye.
------------------------------
From: Eric Chun <[EMAIL PROTECTED]>
Subject: Sound: MP3 Playback Problem
Date: Sun, 15 Apr 2001 09:01:56 -0500
I'm trying to use a program called mpegrec 1.0 by Andrew Sandoval.
The problem I'm having is that after I record something and try playing
it back using xmms, sound only comes out of the right side of my
speakers.
When I try recording something by typing, cat /dev/dsp > myvoice, I
get a recording that comes out of both sides of my speakers. I know
it's not lame because I tried converting a wav file into mp3 format and
playing it back with sound coming out perfectly. I've also tried
recording using sox and piping the stdout to lame, but I get the same
one-sided audio output. I've tried real-time mp3 recording with the
cdrom as my input and that works perfectly also.
I tried adjusting the balance in kmix and I got some interesting
results. When I set the balance all the way to the left (the side that
isn't getting recorded), I get no sound at all from the microphone.
When I set the balance all the way to the right, all the sound comes out
of the right side of the speakers.
I've also tried looking in xload to see if my system is being
overloaded or not. xload doesn't give me a load of 100% so that's good.
Here is my computer setup:
Pentium II 266
Sound Blaster PCI128 (ES1370)
Shuttle HOT637 mobo
Linux Mandrake 7.1
96MB RAM
Maxtor Diamondplus HDD 7.5 Gigs
Altec Lansing ACS48
Any suggestions into what may be wrong will be greatly appreciated.
Thanks.
------------------------------
From: Al <[EMAIL PROTECTED]>
Subject: K7T Master-SI (Athlon 1.2Ghz/266FSB)?
Date: Sun, 15 Apr 2001 17:06:55 +0200
I'm thinking of getting the K7T Master Si with the dual Ultra3 SCSI
adapter. Before buying, does anybody have this board running under
Linux? I hear that (some?) Athlon boards have problems running Linux 7.
Any comments, suggestions, alternatives and experiences appreciated.
Al
------------------------------
From: "Alvin" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux,alt.os.linux
Subject: Re: Linux on Intel Or Celeron? what is the best choice?
Date: Mon, 16 Apr 2001 00:12:10 +0800
AMD is value for money...
disadvantage is the heat and power consumption is higher than Intel
Processors
"Bob Cober" <[EMAIL PROTECTED]> wrote in message
news:xPaz6.107371$[EMAIL PROTECTED]...
> Use AMD
>
>
>
> Johan Kullstam <[EMAIL PROTECTED]> wrote in message
> news:[EMAIL PROTECTED]...
> > [EMAIL PROTECTED] (Med HAM) writes:
> >
> > > On Thu, 05 Apr 2001 21:57:59 GMT, "Geek Mystique"
> > > <[EMAIL PROTECTED]> wrote:
> > >
> > > >Intel is a brand, Celeron a type of processor made by Intel.
> > > >Intel manufactures Celeron type CPUs
> > > Ooops, I want to say Pentuim instead of Intel
> > > sorry!
> >
> > pentium is like 6 years obsolete. you must mean pentium-ii, -iii or
> > -iv.
> >
> > slots are dead. get a socket processor. pentium-iii, celeron,
> > athlon, duron all come in slot factors.
> >
> > i would go for either intel pentium-iii cu-mine or amd athlon
> > thunderbird. (well, i would go for the amd athlon, but if you want to
> > go intel, the p-iii cu-mine isn't half bad.)
> >
> > > >"Med HAM" <[EMAIL PROTECTED]> wrote in message
> > > >news:[EMAIL PROTECTED]...
> > > >> Hi folks,
> > > >> My Company are willing to buy some machines to install Linux on
> > > >> them. My question is which machine is best for Linux to run on :
> > > >> Intel or Celeron? And if possible , can you tell me why? (ie: the
> > > >> adventages and drawbacks of each type of processor )
> > > >> As always any help could be so appreciated.
> > > >> Med HAM -- [EMAIL PROTECTED]
> > > >
> > > >
> > >
> >
> > --
> > J o h a n K u l l s t a m
> > [[EMAIL PROTECTED]]
> > Don't Fear the Penguin!
>
>
------------------------------
From: "jimtheodo" <[EMAIL PROTECTED]>
Crossposted-To:
alt.comp.hardware.homebuilt,alt.comp.hardware.homedesigned,alt.comp.hardware.overclocking,alt.comp.hardware.overclocking.amd,alt.comp.hardware.pc-homebuilt
Subject: Re: infra red - making it my self
Date: Sun, 15 Apr 2001 19:35:08 +0300
try winlirc.
find it on a search engine like google.
steven koolen <[EMAIL PROTECTED]> wrote in message
news:3a64ac02$0$20777@reader4...
> i want infra red on my computer, but its very exspencive, does anyone know
> how to make a infrared port my self
>
> thanx
> steven koolen
>
>
------------------------------
From: [EMAIL PROTECTED] (Peter da Silva)
Crossposted-To: comp.arch.storage,comp.sys.ibm.pc.hardware.storage
Subject: Re: today's harddrives will surely fail before dialup users manage to fill
them up?
Date: 15 Apr 2001 16:22:49 GMT
In article <[EMAIL PROTECTED]>,
J. Clarke <[EMAIL PROTECTED]> wrote:
> In article <9b9u5g$[EMAIL PROTECTED]>, [EMAIL PROTECTED] says...
> > In article <[EMAIL PROTECTED]>,
> > J. Clarke <[EMAIL PROTECTED]> wrote:
> > > In any case you made an assertion that if one's web browser slows down
> > > with a large cache, then one should rewrite the cache program. That
> > > leads one to conclude, since you made this assertion without caveat,
> > > that there is some algorithm out there that allows one to write a cache
> > > program that does not slow down with a large cache.
> > There are no end of logN search algorithms that could be used. They would
> > slow down, yes, but not significantly over a mere few gigabytes or even
> > terabytes of data.
> What difference does the number of bytes make? It's the number of
> directory entries that matters.
The number of directory entries is a linear function of the number of bytes,
(since web pages are not indefinitely small or indefinitely large), so it's
logN either way.
> > For example, you hash the URL to generate a filename and then build a 16-way
> > directory tree indexed on the hexadecimal expansion of the file to be cached.
> And this takes how long?
O(N) with the length of the key, or O(logN) with the number of pages. For a
128-bit key, you'll get to the directory you're going to store the page in
in T(hash)+32*T(directory access).
> How much RAM do you need to do this?
Enough to hold the key and one directory name.
--
`-_-' In hoc signo hack, Peter da Silva.
'U` "A well-rounded geek should be able to geek about anything."
-- [EMAIL PROTECTED]
Disclaimer: WWFD?
------------------------------
From: franek <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.advocacy,comp.os.linux.misc
Subject: Re: Could Linux be used in this factory environment ?
Date: Sun, 15 Apr 2001 16:49:36 GMT
Ed Blackman wrote:
> On 25 Feb 2001 22:31:08 +0800, Paul Repacholi <[EMAIL PROTECTED]> wrote:
> >"Adam Warner" <[EMAIL PROTECTED]> writes:
> >> > Of course the workers would have to access the system to enter data,
> >> > etc, so the user interfaces can't be too complicated (GUI?).
> >> Your workers would know how to use a web browser, so why not make the
> >> inventory system accessible through any web browser? The MySQL database
> >> and PHP scripting language would be a good combination for this task.
> >
> >Never worked in afactory have you? How long will the mouse keep working
> >after it has been grabbed by a paint/glue/oil... covered hand?
>
> So be conservative with your HTML and make sure the page renders nicely
> in lynx or another text mode browser.
I could never understand this enamoration with HTML-based interfaces. There's a good
case
for using HTML in a normal web-based environment, but why the hell one would want to
use
this crude and slow method in a standalone system is beyond me. Maybe they're
influenced
by Microsoft with their huge "innovations" like "look at your desktop as a web page"!
Wow.
End of desktop as we knew it. What you need on a factory floor is an industrial
air/water-tight enclosure with a freaken touch screen (which you can purchase either
built
into the strengthened box itself--there are products like that--or as a hardware
add-on)
and then a normal, fast-responding gui like Qt, or anything at all really. Just make
sure
your buttons are really HUGE, so that you can easily poke them with a gloved finger.
Btw,
these pokes come through as mouse clicks, so, programmatically, it is kinda
mouse-driven.
>
> Ed
------------------------------
From: Joeri Sebrechts <[EMAIL PROTECTED]>
Subject: Re: today's harddrives will surely fail before dialup users manage to
Date: Sun, 15 Apr 2001 16:52:16 GMT
"J. Clarke" wrote:
>
> > Hashed indexes (they can search millions of items in a heartbeat if well
> > used).
> Well, that's very nice. Now how long does it take you to build the hash
> table?
Whenever you add an entry to the cache you add it at the position given
by the hash of (part of) the content. This means that instead of just
finding the last position in your cache, you apply a simple mathematic
formula (we're talking about a few lines of code here, nothing complex).
There is no need for periodic rebuilding of the hashed index.
So for each entry in the cache you calculate the hash once. Since this
is purely mathematic there's no memory or harddrive involved. It all
happens on the cpu. And it is very, very fast.
> > Or balanced trees (though you need to do rebalancing, which comes
> > at a cost)
> And how long does it take you to build the tree?
Longer. Although reiserfs (the new journaling filesystem) uses trees to
navigate the filesystem, and in my subjective experience it's faster
than ext2. At least the find was faster. A lot faster.
> > You can never make a search on something big AS fast as on
> > something small, but you can make the speed penalty really, really
> > small.
> Precisely. With the most perfect algorithm in the world, a larger cache
> will take longer to access than a smaller one.
Yes. But only infinitesimally longer.
> > Anyway, a well-designed cache never lets itself grow out of proportion.
> > It will push out the old while the new comes in. After all, apart from
> > allowing faster access to stuff that matters, that's the whole point of
> > a cache.
> So how big is "out of proportion"?
So big that it becomes slower than what it's supposed to cache.
Anyway, if a cache is slower than the uncached input, then it IS badly
designed. Caching strategies is something that has been well researched,
and there is no excuse anymore for making a cache that bogs down when it
gets big.
If you don't believe me ask the programmers of all the decent caches and
proxies out there.
------------------------------
From: Joeri Sebrechts <[EMAIL PROTECTED]>
Subject: Re: BP6 Mandrake 7.1 & Up Install Locking up.
Date: Sun, 15 Apr 2001 16:54:46 GMT
"Glenn A. Serpas Jr." wrote:
> H/D Ctrl: HPT366 UDMA66
> I've been unable to install Linux Mandrake since version 7.0 .. On
> versions 7.1 & 7.2 I get locked up shortly after the install begins
> installing packages from CD#1 .. I recently tried 8.0 Beta 2.. It locked
> shortly after installing packages from CD#2 .. This is a hard lock. I
> have to shut my system off.. Reset and CTRL+ALT+DEL do not work.. I've
> asked Mandrake and a few Linux news groups. No one seems to have a clue.
> Could the lock up have something to do w/the EC10 problem on version 1.1
> MB's??? I've noticed that occassionally my machine will just lock up.. I
> duel boot between WIn2K Professional and Mandrake(or at least I used to)..
I've heard the HTP366 requires a special patch to the kernel (don't know
why though).
------------------------------
From: Joeri Sebrechts <[EMAIL PROTECTED]>
Subject: Re: Sound Volume very low!!
Date: Sun, 15 Apr 2001 16:56:35 GMT
Chris Rankin wrote:
>
> "Vilas Shekhar B.J." wrote:
> >
> > I have Linux Mandrake 7.1 on my computer. It recognises my sound card but
> > the playback is at almost 10% voulme. Even if I increase the volume in the
> > mixer settings the sound is barely audible. Any solutions ?
>
> There are many mixer settings; try increasing all of them. For example,
> there is the "master" volume, the PCM volume (for WAV files, etc) and
> the CD volume. If it's still too quiet then turn up the volume on your
> speakers ;-)
This might be a dumb remark, but I've had this happen to me before.
Are you sure your speakers are plugged into the amplified line out, and
not in the regular line out ? And if so, are you sure the amplified line
out is actually amplified (usually you need to specifically enable this
in the settings, although Windows tends to do this for you).
------------------------------
** 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 comp.os.linux.hardware.
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-Hardware Digest
******************************
