Linux-Hardware Digest #221, Volume #9 Tue, 19 Jan 99 18:13:35 EST
Contents:
ASUS P55T2P4/AMD K6-2-350 - Summary (Phil Houstoun)
Re: Linux compatiblity with Colorado tape drives. ("Patrick D. Rockwell")
Re: 3com 3c905b card loses Mac address ([EMAIL PROTECTED])
Redhat linux with Philips Recordable/Rewriteable CD-ROM? ("L. Zhang")
Re: My partition choice (James Youngman)
----------------------------------------------------------------------------
From: Phil Houstoun <[EMAIL PROTECTED]>
Crossposted-To: alt.comp.periphs.mainboard.asus
Subject: ASUS P55T2P4/AMD K6-2-350 - Summary
Date: Thu, 14 Jan 1999 16:08:46 -0500
==============E0FED8B22437570FC7ECB060
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
ASUS P55T2P4 and AMD K6-2-350
--------------------------------
Just a quick summary of my experience with this board/CPU
combination (your mileage may vary!). Detailed instructions on how to
achieve the bus speeds,
multiplier settings and voltages are to be found in the references cited
at the
bottom of this summary. I initially attempted testing under both Win98
and
Linux but it became rapidly apparent that the Linux kernel compile was a
much
more rigorous real-world test than anything I could do under Win98
(except,
perhaps, for playing 'Unreal' for several hours straight). I chose to
run
fifteen consecutive compiles of the Linux kernel as the basis for my
acceptance/rejection of a particular effective speed. Some people
advocate one
hundred iterations, but at roughly four minutes per, one hour of
intensive
activity is good enough for me. It would appear, for whatever reason,
that CPU
voltage level makes a BIG difference in my results (see below). I
initially
started at the base voltage recommended for this chip, 2.2v, but it
became
rapidly obvious that nothing but the 60x6.0 combo would work reliably at
this
voltage so I gradually worked my way up.
CPU/BUS CONFIGURATIONS
======================
Voltage Bus Mult Effective Linux (RH5.1, 2.0.35, gcc2.7.2.3-13)
======= === ==== ========= ====================================
2.2 60 6.0 360 Boots,Kernel compiles 15 out of 15
2.3 60 6.0 360 Boots,Kernel compiles 15 out of 15
2.3 66 5.5 363 Boots,Kernel compiles 8 out of 15
2.3 83 4.5 373.5 Boots,Kernel compiles 0 out of 15
2.3 75 5.0 375 Boots,Kernel compiles 0 out of 15
2.3 66 6.0 396 Won't boot clean
2.3 75 5.5 412.5 Won't boot clean
2.3 83 5.0 415 Won't boot clean
2.4 66 5.5 363 Boots,Kernel compiles 15 out of 15
2.4 83 4.5 373.5 Boots,Kernel compiles 15 out of 15
2.4 75 5.0 375 Boots,Kernel compiles 15 out of 15
2.4 66 6.0 396 Boots,Kernel compiles 0 out of 15
2.4 68 6.0 408 Boots,Kernel compiles 0 out of 15
2.4 75 5.5 412.5 Boots,System is unstable - crashes
2.4 83 5.0 415 Won't boot clean
2.4 75 6.0 450 My God! It's full of stars!
WARNING: The following voltage setting is NOT recommended by AMD
2.5 66 6.0 396 Boots,Kernel compiles 0 out of 15
2.5 68 6.0 408 Boots,Kernel compiles 0 out of 15
2.5 75 5.5 412.5 Boots,System is unstable - crashes
2.5 83 5.0 415 Won't boot clean
SYSTEM CONFIGURATION
====================
ASUS P55T2P4 Rev. 3.1 BIOS 207-1, Chipset settings at 60ns 'auto'
64Mb EDO RAM (60ns - tested thoroughly with memtest-86)
512Kb Cache (COAST v3.0)
AMD-K6-2/350AFR, 2.2v CORE/3.3v I/O, A9847DPEW, Lot No. 2522694,
Core 26351 ('new core'), Typ=00, Fam=05, Mod=08, Step=0C
Adaptec 2940UW SCSI Adaptor (PCI)
Quantum Grand Prix XP34301 4.3Gb Fast Wide SCSI
Seagate ST-32550W 2.1Gb Fast Wide SCSI
Pioneer 24x SCSI CDROM
Yamaha 4260t SCSI CDRW
Exabyte Eagle TR-3 Tape Drive
Exabyte 2Mbit Accelerator Card (AT)
Lightning 128 (Tseng ET6000) 2D w/4Mb (PCI)
Canopus Pure 3D LX w/6Mb (PCI)
3Com 503 Ethernet Card (AT)
Creative Labs SB16 Pro (AT)
300w power supply, 2 extra case fans one blowing on SCSI disks at top of
case,
the other blowing on CPU at bottom of case, monster CPU heat sink/fan
with
lotsa goop
BENCHMARK @ 75x5.0, 2.4v
========================
BYTEmark* Native Mode Benchmark ver. 2 (10/95)
Index-split by Andrew D. Balsa (11/97)
Linux/Unix* port by Uwe F. Mayer (12/96,11/97)
TEST : Iterations/sec. : Old Index : New Index
: : Pentium 90* : AMD K6/233*
====================:==================:=============:============
NUMERIC SORT : 192.68 : 4.94 : 1.62
STRING SORT : 22.841 : 10.21 : 1.58
BITFIELD : 3.1076e+07 : 5.33 : 1.11
FP EMULATION : 11.164 : 5.36 : 1.24
FOURIER : 2632.7 : 2.99 : 1.68
ASSIGNMENT : 1.7362 : 6.61 : 1.71
IDEA : 337.58 : 5.16 : 1.53
HUFFMAN : 161.82 : 4.49 : 1.43
NEURAL NET : 2.0554 : 3.30 : 1.39
LU DECOMPOSITION : 44.702 : 2.32 : 1.67
==========================ORIGINAL BYTEMARK
RESULTS==========================
INTEGER INDEX : 5.799
FLOATING-POINT INDEX: 2.839
Baseline (MSDOS*) : Pentium* 90, 256 KB L2-cache, Watcom* compiler
10.0
==============================LINUX DATA
BELOW===============================
C compiler : gcc version 2.7.2.3
libc : unknown version
MEMORY INDEX : 1.444
INTEGER INDEX : 1.449
FLOATING-POINT INDEX: 1.575
Baseline (LINUX) : AMD K6/233*, 512 KB L2-cache, gcc 2.7.2.3,
libc-5.4.38
* Trademarks are property of their respective holder.
BENCHMARK @ 83x4.5, 2.4v
========================
BYTEmark* Native Mode Benchmark ver. 2 (10/95)
Index-split by Andrew D. Balsa (11/97)
Linux/Unix* port by Uwe F. Mayer (12/96,11/97)
TEST : Iterations/sec. : Old Index : New Index
: : Pentium 90* : AMD K6/233*
====================:==================:=============:============
NUMERIC SORT : 194.56 : 4.99 : 1.64
STRING SORT : 22.635 : 10.11 : 1.57
BITFIELD : 3.1233e+07 : 5.36 : 1.12
FP EMULATION : 11.164 : 5.36 : 1.24
FOURIER : 2644.4 : 3.01 : 1.69
ASSIGNMENT : 1.7189 : 6.54 : 1.70
IDEA : 339.92 : 5.20 : 1.54
HUFFMAN : 163.46 : 4.53 : 1.45
NEURAL NET : 2.0607 : 3.31 : 1.39
LU DECOMPOSITION : 44.906 : 2.33 : 1.68
==========================ORIGINAL BYTEMARK
RESULTS==========================
INTEGER INDEX : 5.809
FLOATING-POINT INDEX: 2.850
Baseline (MSDOS*) : Pentium* 90, 256 KB L2-cache, Watcom* compiler
10.0
==============================LINUX DATA
BELOW===============================
C compiler : gcc version 2.7.2.3
libc : unknown version
MEMORY INDEX : 1.438
INTEGER INDEX : 1.459
FLOATING-POINT INDEX: 1.581
Baseline (LINUX) : AMD K6/233*, 512 KB L2-cache, gcc 2.7.2.3,
libc-5.4.38
* Trademarks are property of their respective holder.
REFERENCES
==========
ASUS info: 1. http://www.sysdoc.pair.com/asust2p4.htm
2. http://www.jump.net/~lcs/kalle
AMD info: 1. http://users.uniserve.com/~dougf
--
Phil Houstoun Voice: (613) 991-7173
Email: [EMAIL PROTECTED]
=======================================================================
==============E0FED8B22437570FC7ECB060
Content-Type: text/html; charset=us-ascii
Content-Transfer-Encoding: 7bit
<HTML>
ASUS P55T2P4 and AMD K6-2-350
<BR>
================================
<P> Just a quick summary of my experience with this board/CPU
combination (your mileage may vary!). Detailed instructions on how
to achieve the bus speeds,
<BR>multiplier settings and voltages are to be found in the references
cited at the
<BR>bottom of this summary. I initially attempted testing under both Win98
and
<BR>Linux but it became rapidly apparent that the Linux kernel compile
was a much
<BR>more rigorous real-world test than anything I could do under Win98
(except,
<BR>perhaps, for playing 'Unreal' for several hours straight). I chose
to run
<BR>fifteen consecutive compiles of the Linux kernel as the basis for my
<BR>acceptance/rejection of a particular effective speed. Some people advocate
one
<BR>hundred iterations, but at roughly four minutes per, one hour of intensive
<BR>activity is good enough for me. It would appear, for whatever reason,
that CPU
<BR>voltage level makes a BIG difference in my results (see below). I initially
<BR>started at the base voltage recommended for this chip, 2.2v, but it
became
<BR>rapidly obvious that nothing but the 60x6.0 combo would work reliably
at this
<BR>voltage so I gradually worked my way up.
<P>CPU/BUS CONFIGURATIONS
<BR>----------------------
<BR>Voltage Bus Mult Effective Linux (RH5.1,
2.0.35, gcc2.7.2.3-13)
<BR>------- --- ---- ---------
------------------------------------
<BR>2.2 60 6.0
360
Boots,Kernel compiles 15 out of 15
<P>2.3 60 6.0
360
Boots,Kernel compiles 15 out of 15
<BR>2.3 66 5.5
363
Boots,Kernel compiles 8 out of 15
<BR>2.3 83 4.5
373.5
Boots,Kernel compiles 0 out of 15
<BR>2.3 75 5.0
375
Boots,Kernel compiles 0 out of 15
<BR>2.3 66 6.0
396
Won't boot clean
<BR>2.3 75 5.5
412.5
Won't boot clean
<BR>2.3 83 5.0
415
Won't boot clean
<P>2.4 66 5.5
363
Boots,Kernel compiles 15 out of 15
<BR>2.4 83 4.5
373.5
Boots,Kernel compiles 15 out of 15
<BR>2.4 75 5.0
375
Boots,Kernel compiles 15 out of 15
<BR>2.4 66 6.0
396
Boots,Kernel compiles 0 out of 15
<BR>2.4 68 6.0
408
Boots,Kernel compiles 0 out of 15
<BR>2.4 75 5.5
412.5
Boots,System is unstable - crashes
<BR>2.4 83 5.0
415
Won't boot clean
<BR>2.4 75 6.0
450
My God! It's full of stars!
<P>WARNING: The following voltage setting is NOT recommended by AMD
<P>2.5 66 6.0
396
Boots,Kernel compiles 0 out of 15
<BR>2.5 68 6.0
408
Boots,Kernel compiles 0 out of 15
<BR>2.5 75 5.5
412.5
Boots,System is unstable - crashes
<BR>2.5 83 5.0
415
Won't boot clean
<BR>
<P>SYSTEM CONFIGURATION
<BR>--------------------
<P>ASUS P55T2P4 Rev. 3.1 BIOS 207-1, Chipset settings at 60ns 'auto'
<BR>64Mb EDO RAM (60ns - tested thoroughly with memtest-86)
<BR>512Kb Cache (COAST v3.0)
<BR>AMD-K6-2/350AFR, 2.2v CORE/3.3v I/O, A9847DPEW, Lot No. 2522694,
<BR> Core 26351 ('new core'), Typ=00, Fam=05, Mod=08,
Step=0C
<BR>Adaptec 2940UW SCSI Adaptor (PCI)
<BR> Quantum Grand Prix XP34301 4.3Gb Fast Wide SCSI
<BR> Seagate ST-32550W 2.1Gb Fast Wide SCSI
<BR> Pioneer 24x SCSI CDROM
<BR> Yamaha 4260t SCSI CDRW
<BR>Exabyte Eagle TR-3 Tape Drive
<BR>Exabyte 2Mbit Accelerator Card (AT)
<BR>Lightning 128 (Tseng ET6000) 2D w/4Mb (PCI)
<BR>Canopus Pure 3D LX w/6Mb (PCI)
<BR>3Com 503 Ethernet Card (AT)
<BR>Creative Labs SB16 Pro (AT)
<BR>300w power supply, 2 extra case fans one blowing on SCSI disks at top
of case,
<BR>the other blowing on CPU at bottom of case, monster CPU heat sink/fan
with
<BR>lotsa goop
<P>BENCHMARK @ 75x5.0, 2.4v
<BR>------------------------
<P>BYTEmark* Native Mode Benchmark ver. 2 (10/95)
<BR>Index-split by Andrew D. Balsa (11/97)
<BR>Linux/Unix* port by Uwe F. Mayer (12/96,11/97)
<P>TEST
: Iterations/sec. : Old Index : New Index
<BR>
:
: Pentium 90* : AMD K6/233*
<BR>--------------------:------------------:-------------:------------
<BR>NUMERIC SORT
:
192.68 : 4.94
:
1.62
<BR>STRING SORT
:
22.841 : 10.21
:
1.58
<BR>BITFIELD
: 3.1076e+07 :
5.33 : 1.11
<BR>FP EMULATION
:
11.164 : 5.36
:
1.24
<BR>FOURIER
: 2632.7
::
2.99 : 1.68
<BR>ASSIGNMENT
:
1.7362 : 6.61
:
1.71
<BR>IDEA
: 337.58
::
5.16 : 1.53
<BR>HUFFMAN
: 161.82
::
4.49 : 1.43
<BR>NEURAL NET
:
2.0554 : 3.30
:
1.39
<BR>LU DECOMPOSITION
:
44.702 : 2.32
:
1.67
<BR>==========================ORIGINAL BYTEMARK RESULTS==========================
<BR>INTEGER INDEX : 5.799
<BR>FLOATING-POINT INDEX: 2.839
<BR>Baseline (MSDOS*) : Pentium* 90, 256 KB L2-cache, Watcom*
compiler 10.0
<BR>==============================LINUX DATA BELOW===============================
<BR>C compiler :
gcc version 2.7.2.3
<BR>libc
: unknown version
<BR>MEMORY INDEX : 1.444
<BR>INTEGER INDEX : 1.449
<BR>FLOATING-POINT INDEX: 1.575
<BR>Baseline (LINUX) : AMD K6/233*, 512 KB L2-cache,
gcc 2.7.2.3, libc-5.4.38
<BR>* Trademarks are property of their respective holder.
<BR>
<P>BENCHMARK @ 83x4.5, 2.4v
<BR>------------------------
<P>BYTEmark* Native Mode Benchmark ver. 2 (10/95)
<BR>Index-split by Andrew D. Balsa (11/97)
<BR>Linux/Unix* port by Uwe F. Mayer (12/96,11/97)
<P>TEST
: Iterations/sec. : Old Index : New Index
<BR>
:
: Pentium 90* : AMD K6/233*
<BR>--------------------:------------------:-------------:------------
<BR>NUMERIC SORT
:
194.56 : 4.99
:
1.64
<BR>STRING SORT
:
22.635 : 10.11
:
1.57
<BR>BITFIELD
: 3.1233e+07 :
5.36 : 1.12
<BR>FP EMULATION
:
11.164 : 5.36
:
1.24
<BR>FOURIER
: 2644.4
::
3.01 : 1.69
<BR>ASSIGNMENT
:
1.7189 : 6.54
:
1.70
<BR>IDEA
: 339.92
::
5.20 : 1.54
<BR>HUFFMAN
: 163.46
::
4.53 : 1.45
<BR>NEURAL NET
:
2.0607 : 3.31
:
1.39
<BR>LU DECOMPOSITION
:
44.906 : 2.33
:
1.68
<BR>==========================ORIGINAL BYTEMARK RESULTS==========================
<BR>INTEGER INDEX : 5.809
<BR>FLOATING-POINT INDEX: 2.850
<BR>Baseline (MSDOS*) : Pentium* 90, 256 KB L2-cache, Watcom*
compiler 10.0
<BR>==============================LINUX DATA BELOW===============================
<BR>C compiler :
gcc version 2.7.2.3
<BR>libc
: unknown version
<BR>MEMORY INDEX : 1.438
<BR>INTEGER INDEX : 1.459
<BR>FLOATING-POINT INDEX: 1.581
<BR>Baseline (LINUX) : AMD K6/233*, 512 KB L2-cache,
gcc 2.7.2.3, libc-5.4.38
<BR>* Trademarks are property of their respective holder.
<P>REFERENCES
<BR>----------
<P>ASUS info: 1. <A
HREF="http://www.sysdoc.pair.com/asust2p4.htm">http://www.sysdoc.pair.com/asust2p4.htm</A>
<BR>
2. <A HREF="http://www.jump.net/~lcs/kalle">http://www.jump.net/~lcs/kalle</A>
<P>AMD info: 1. <A
HREF="http://users.uniserve.com/~dougf">http://users.uniserve.com/~dougf</A>
<BR>
<PRE>--
Phil Houstoun Voice:
(613) 991-7173
Email: [EMAIL PROTECTED]
=======================================================================</PRE>
</HTML>
==============E0FED8B22437570FC7ECB060==
------------------------------
From: "Patrick D. Rockwell" <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Crossposted-To:
comp.os.linux.advocacy,comp.os.linux.development.apps,comp.os.linux.misc,alt.os.linux
Subject: Re: Linux compatiblity with Colorado tape drives.
Date: Tue, 19 Jan 1999 14:25:35 -0800
John Thompson wrote:
> In <[EMAIL PROTECTED]>, "Patrick D. Rockwell" <[EMAIL PROTECTED]>
>writes:
>
> >I'm planning to set up a multi platform system with Dos 6.22, Windows
> >95, OS2 Warp, and Linux. I
> >have a T1000 tape drive, but I'm planning to get a Colorado 4/8 Gb tape
> >drive (part number C4386B). I
> >know that Dos and Windows 95 work with it. What about Linux? Is there a
> >Linux driver which will
> >work with the tape drive that I'm planning to get? Is there a drive
> >which will work with all four OS's?
> >
> >In particular, I'm planning to get Redhat Linux, but if there is another
> >better Linux out there which will
> >work with the Colorado 4/8 Gb tape drive, please let me know.
>
> Supposedly floppy-based tape drives will work in linux with the
> ftape support but I had trouble getting my Jumbo 350 to function
> in linux. I had a FC-20 accelerator card installed which may
> have contributed to the problem, even though I tried all the
> suggestions for enabling accelerator card support with ftape. I
> ended up getting SCSI drive which works quite nicely. YMMV.
>
> -John ([EMAIL PROTECTED])
Well, the Colorado 4/8 Gb drive is IDE/ATAPI controlled, and there is a product called
BRU
2000 which according to the people that I called at Enhanced Software Technologies, it
should
work with this tape drive, even if I have to call for technical support. I discovvered
BRU
2000 on the Redhat Linux page at http://www.redhat.com .
--
Patrick D. Rockwell
mailto:[EMAIL PROTECTED]
mailto:[EMAIL PROTECTED]
mailto:[EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: 3com 3c905b card loses Mac address
Date: Tue, 19 Jan 1999 21:27:14 GMT
In article <ZESo2.220$[EMAIL PROTECTED]>,
[EMAIL PROTECTED] (Rob Clark) wrote:
> In article <780rdi$t88$[EMAIL PROTECTED]>, <[EMAIL PROTECTED]> wrote:
> >time the MAC address reported by the 3c59x driver was "ff:ff:ff:ff:ff:ff"!
> >Multiple reboots produced the same result each time: Win 98 got the right MAC
> >address and was able to obtain an IP and talk to the network, Linux continued
> >to see ff:ff:ff:ff:ff:ff as the MAC address. :-(
>
> These are really flake-o cards for a 3Com product-- we've got a bunch of
> them and they're a headache. This is a known problem with some
> combinations of hardware.
>
> Here's the official answer:
>
> http://cesdis.gsfc.nasa.gov/linux/drivers/vortex.html#pnpos
>
> Good luck!
> Rob Clark, [EMAIL PROTECTED]
> http://www.o2.net/~gromitkc/winmodem.html
>
Thanks a lot. I got it to work by applying the "cold boot" solution: turn the
computer off when rebooting from Windows. Once I had a working Ethernet card,
it was simple to download the latest XF86 server and get X running. I gave my
friend KDE to start with (easing him in, I'll introduce him to other
windowmanagers later :-)). He seems to like it so far...
Thanks to you, there's one more Linux user in the world. Congratulations!
Robin Munn (Legal name: Robert A. Munn)
[EMAIL PROTECTED]
============= Posted via Deja News, The Discussion Network ============
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
------------------------------
From: "L. Zhang" <[EMAIL PROTECTED]>
Subject: Redhat linux with Philips Recordable/Rewriteable CD-ROM?
Date: Tue, 19 Jan 1999 18:17:48 +0000
Hey there,
Could anyone tell me whether Redhat linux support the Philips
Recordable/Rewriteable CD-ROM? Thanks.
--Luke
------------------------------
From: James Youngman <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.misc,comp.unix.questions,comp.os.linux
Subject: Re: My partition choice
Date: 18 Jan 1999 19:59:55 +0000
DaZZa <[EMAIL PROTECTED]> writes:
> On 17 Jan 1999, Ilya wrote:
>
> > Thanks for clarifying - the machine will have 256MB of RAM eventually, and
> > 128MB when I get it. I guess I will have to have several swap partitions.
> > I am interested in a server-type machine. I don't know exactly what I will
> > be doing with it, but it is conceivable I might end up doing RAM-intensive
> > work.
>
> As a general rule of thumb, swap space should be approximately twice your
> physical memory.
That "general rule of thumb" is completely inappropriate and hasn't
been valid since some time in the eighties.
--
ACTUALLY reachable as @free-lunch.demon.(whitehouse)co.uk:james+usenet
------------------------------
** 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.hardware) 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-Hardware Digest
******************************
