Not long ago, I worked my way through the upx code, and figured
out what I needed to do to use it in another project. And started
using that as the decompressor for etherboot.
Today I have gone through and replaced the copy from rom to ram
in crt0.base with a decompression step. This allows me to compress
all of the C code when it is stored in the rom chip, and only
decompress it as it is copied to RAM. The upx decompressor
is roughly 180 bytes. And I get a compression ratio of about 2:1
When I start compiling in a lot of debug messages, or the intel
microcode updates this compressor gives me a lot more room
to work with. I reduced my footprint on the E7500 port by 20KB :)
This interacts badly with my implementation of the cache as ram trick but
that code wasn't stable enough to really trust. :(
And in theory since I am moving less data off of the rom chip things
should go even faster.
Hello again from Gregg C Levine
Would someone correct me, if I am wrong? These clusters, they are part
of the Beowulf project, that Donald Becker originally developed using
commodity PCs, complete with their original headaches? I went over the
front page at www.beowulf.org and it clarified the issues only part of
the way. I have here a copy of his original CD, as based on Red Hat 6.2
and so far even it has failed to enlighten me.
Gregg C Levine hansolofalcon(a)worldnet.att.net
"The Force will be with you...Always." Obi-Wan Kenobi
"Use the Force, Luke." Obi-Wan Kenobi
(This company dedicates this E-Mail to General Obi-Wan Kenobi )
(This company dedicates this E-Mail to Master Yoda )
-------- Original Message --------
Subject: Re: Lab to sample Linux for weapons work
Date: Thu, 26 Sep 2002 15:06:56 -0400
From: Velocet <math(a)velocet.ca>
On Thu, Sep 26, 2002 at 03:49:39PM +0200, Eugen Leitl's all...
> There's a major difference compared with Livermore system, though: The Los
> Alamos machine has no hard drives. Instead, each computer fires up using
> software pulled over the network with the assistance of software called
> LinuxBIOS developed by LANL programmer Ron Minnich and others. LinuxBIOS
> also dramatically speeds the startup process to about two seconds, said
> Jason Lowry, Linux NetworX's product manager for cluster management tools.
And what filesystem are they using for this? Surely its not NFS!
Must be some sort of cluster filesystem no? Anyone know? Something open
sourced? What decent clustre filesystems are out there?
> Shunning hard drives cuts cost and power consumption, but more
> importantly, it improves reliability, Roundy said.
> "If you think about what things are going to fail in a system, it's the
> hard disk or fan or power supply or something with moving parts," Roundy
> Linux NetworX could benefit greatly from convincing the Los Alamos and
> Livermore labs that Linux clusters are worthwhile. The labs are funded by
> the Energy Department's Advanced Simulation and Computing program, which
> has spent hundreds of millions of dollars to advance supercomputing using
> machines made of comparatively inexpensive components.
> The DOE program has underwritten many of the world's fastest computers,
> according to university researchers who monitor raw calculation speed at
> the Top500 organization. The program has underwritten Nos. 2, 6, 7, 9, 11,
> and 15 on the most recent ranking.
> Beowulf mailing list, Beowulf(a)beowulf.org
> To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf
Ken Chase, math(a)velocet.ca * Velocet Communications Inc. * Toronto, CANADA
Beowulf mailing list, Beowulf(a)beowulf.org
To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf
Hi, I'd like to know if you think it would be feasible without too much trouble
to use linuxbios on Frontpath's Progear webpad.
The whole thing is actually usable on linux, the input is the touchscreen or an
No distro ( linux from scratch ).
The CPU is a transmeta crusoeTM3200, and here s the /proc/pci :
PCI devices found:
Bus 0, device 0, function 0:
Host bridge: Transmeta Corporation Northbridge (rev 1).
Non-prefetchable 32 bit memory at 0xfc000000 [0xfc0fffff].
Bus 0, device 4, function 0:
Multimedia audio controller: Acer Laboratories Inc. [ALi] M5451 PCI South
Bridge Audio (rev 1).
Master Capable. Latency=64. Min Gnt=2.Max Lat=24.
I/O at 0x1000 [0x10ff].
Non-prefetchable 32 bit memory at 0xfc100000 [0xfc100fff].
Bus 0, device 5, function 0:
CardBus bridge: Texas Instruments PCI1410 PC card Cardbus Controller (rev
Master Capable. Latency=168. Max Lat=5.
Non-prefetchable 32 bit memory at 0x10000000 [0x10000fff].
Bus 0, device 6, function 0:
VGA compatible controller: Silicon Motion, Inc. SM710 LynxEM (rev 163).
Master Capable. Latency=64.
Non-prefetchable 32 bit memory at 0xfd000000 [0xfdffffff].
Bus 0, device 7, function 0:
ISA bridge: Acer Laboratories Inc. [ALi] M1533 PCI to ISA Bridge [Aladdin
IV] (rev 0).
Bus 0, device 15, function 0:
IDE interface: Acer Laboratories Inc. [ALi] M5229 IDE (rev 195).
Master Capable. Latency=64. Min Gnt=2.Max Lat=4.
I/O at 0x1f0 [0x1ff].
I/O at 0x3f6 [0x3f6].
I/O at 0x170 [0x177].
I/O at 0x376 [0x376].
I/O at 0x1400 [0x140f].
Bus 0, device 17, function 0:
Bridge: Acer Laboratories Inc. [ALi] M7101 PMU (rev 0).
Bus 0, device 20, function 0:
USB Controller: Acer Laboratories Inc. [ALi] M5237 USB (rev 3).
Master Capable. Latency=64. Max Lat=80.
Non-prefetchable 32 bit memory at 0xfc101000 [0xfc101fff].
Bus 1, device 0, function 0:
Ethernet controller: PCI device 13d1:ab02 (rev 17).
Master Capable. No bursts. Min Gnt=255.Max Lat=255.
I/O at 0x4000 [0x40ff].
Non-prefetchable 32 bit memory at 0x10800000 [0x108003ff].
(note : the ethernet controller is a pcmcia one, just forget it)
-------- Original Message --------
Subject: Re: [Fwd: Re: "LinuxBIOS for ARM" or "Merging LinuxBIOS with
Date: Thu, 26 Sep 2002 03:04:41 -0400
From: Thomas Chen <tchen(a)on-go.com>
To: Bari Ari <bari(a)onelabs.com>
CC: lart <lart(a)lart.tudelft.nl>
that is not true.... i have armboot booting uclinux on
atmel eb40, oki and a customed dev board.... quite nice...
i looked at blob myself and thought armboot was much easier to work with
Bari Ari wrote:
> -------- Original Message --------
> Subject: Re: "LinuxBIOS for ARM" or "Merging LinuxBIOS with BLOB"
> Date: 26 Sep 2002 14:01:56 +0800
> From: ollie lho
> To: Nicholas Mistry
> CC: linuxbios(a)clustermatic.org
> References: <3D920BDB.6080508(a)onelabs.com>
> On Thu, 2002-09-26 at 13:13, Nicholas Mistry wrote:
>> Just for the record, i have worked extensively with blob. Especially
>> during the StrongARM port of linux.
>> The IPAQ 36xx, 37xx, 38xx handhelds are based off the Intel
>> Assabet/Neponset Development platform. The assabet is a StrongARM 1110
>> chip, and the neponset a 1111 chip.
>> I currently own one of these, and i have to say its a nice setup.
>> After i
>> got my IPAQ 3870, i put the Assabet/Neponset hardware asside.
>> BLOB is a great little bootloader. It has definitely come a long way in
>> the past year. The one thing that i disliked about it was the mechanics
>> for updating flash. But i am sure they have worked that out by now.
> Do you know if Blob can load uClinux ?? I found that most of the ARM
> bootloaders only support ARM-Linux. (Am I wrong ??)
> Linuxbios mailing list
> To unsubscribe from this list: send the line "unsubscribe lart" in
> the body of a message to majordomo(a)lart.tudelft.nl
> Please read the LART FAQ at http://www.lart.tudelft.nl/faq.php3
this is kind of interesting ...
---------- Forwarded message ----------
Date: Tue, 24 Sep 2002 10:02:49 -0700
From: Terry Lambert <tlambert2(a)mindspring.com>
To: Danny Braniss <danny(a)cs.huji.ac.il>
Subject: Re: 3.5GB memory limit?
Danny Braniss wrote:
> the Addendum says:
> * AMD chipset architecture requieres memory above 3.5GB to be reserved
> for PCI devices.
We have a winner.
> Q1: is this AMD specific, or also true for P3/P4?
It's chipset, not CPU specific. Your question is like asking if,
because your Yugo (car) can't go faster than 100 KPH on one brand
of gasoline, if it's a gasoline problem.
> Q2: is there any way of accessing the 'lost' memory? ie some device
No, not unless the chipset that imposes the limitation permits
some form of bank selection.
To Unsubscribe: send mail to majordomo(a)FreeBSD.org
with "unsubscribe freebsd-hackers" in the body of the message
Has anyone else here worked with BLOB?
Blob is a boot loader for SA11x0 (StrongARM) platforms and is just
starting to support PXA and Xscale processors. Blob is able to boot a
Linux kernel stored in flash or RAM and provide that kernel with a
ramdisk (again from flash or RAM). The current BLOB binary is only about
7.2KB and the complete source tree is around 750KB.
BLOB is mostly written in C with some Assembly much like LinuxBIOS.
We're considering merging all the work with Etherboot and LinuxBIOS with
BLOB somehow to get LinuxBIOS and Etherboot working with ARM and Xscale
processors. It would be nice to be able to share all of this between
Application areas would include:
as well as several other PDA, embedded and network infrastructure
devices since Xscale is now used heavily for network processors and the
current ARM and Xscale bootloader, config/init tools like Angel and the
Intel Telematics bootloader are not exactly wonderful.
I have been on other things for a while but had a chance to look at this
Recap: on the smartcore P3, etherboot 5.0.7 fails because all inw()
operations return 0. Two inw()s in a row sometimes return the right value
on the second one.
I thought this might be a memory config problem so I brought down memtest
3.0, and am now running it. Unfortunately, it runs just fine. Memory
configuration, at least judging by the memtest results, is correct on this
machine. I'm up to test 4, running for 7 minutes now, and usually memory
problems if they existed would have shown up by now.
So, back to the original issue: inw operations acting wrong. The first
inw() always reads 0, the seconds reads what looks like the right value.
Anybody have an idea on what kind of north/south configuration problems
could make this happen?