I have code for the gzip inflate algorithm (as used in gzip and pkzip)
that is suitable for embedded use.
I started with Mark Adler's (not-copyrighted) code from INFO-Zip and
modified it to run in an embedded environment. It decompresses from a
memory array to RAM with no external
Jim Gettys wrote:
Chris,
USB flash devices are IDE devices; you are not only pulling in whatever
the USB stack is doing to you, but also the IDE driver both.
It's actually even more complicated than that.
At the hardware interface layer, a Compact FLASH card looks like an IDE
drive; an IDE
[Changing name of thread...]
a) I think we need our own sub-architecture, sooner rather than later.
I could give a bunch of reasons, but I suspect that they are obvious to
everyone here, considering how little our hardware resembles a legacy PC.
b) Do we really need PCI autoconfiguration at
Our big set of variables is all the stuff/buses that can be on USB.
There is a trivially easy way to handle USB device trees in our environment:
Since USB is hot-pluggable, the device tree is inherently a temporal
approximation of reality, because a device could be plugged or unplugged
David Zeuthen wrote:
On Tue, 2006-08-15 at 10:10 -0600, ollie wrote:
The ELF image does not contain the arch/i386/boot stuff at all. Is
there any reason we have to load something like bzImage?
Yes, we want the OLPC distro to be able to boot on non-OLPC machines.
Like e.g. qemu.
I
Minutes from System Software conference call, 2006-08-15
Present: Jim, Zephaniah, Mitch
Mitch added some stuff to the wiki based on his experiences trying to get
LinuxBIOS loaded into the ROM emulator. It's all working now.
It's important to use a powered USB 2.0 hub. Some of the flakiness
Some new wiki pages for your edification and enjoyment:
http://wiki.laptop.org/go/ROM_Emulators_for_OLPC (useful for firmware
developers)
http://wiki.laptop.org/go/Recipes_for_buildrom_and_LinuxBIOS (some tips
and techniques)
___
Devel mailing
Richard Smith wrote:
I'm done with PLCC!
I dunno - 11 minutes is a bit long. I'll remain lazy and stick with the
easy flash.. :)
Is it worth it for me to do a full page write and perhaps cut that
down to around 6 minutes?
I'd say it's worth it, under the assumption that it probably won't
I did an extensive edit on the wiki page (
http://wiki.laptop.org/go/Programming_the_SPI_FLASH ) to reflect that
fact that olpcflash is, or will soon be, just there. The page no
reads like a user manual instead of developer's notes.
A section on how to bootstrap the process is included at
Damián Viano wrote:
It would be nice if these teleconferences could be recorded (speex?) and
published somewhere for the lurkers/developers out there.
I have been trying to capture the essential information content in the
minutes.
I expect that it's much faster to scan the minutes than to
OLPC BIOS meeting, 2006-08-16
Present: Richard, Jim, Jordan, Ray, Marcello, Mitch (scribe)
Jordan - definitely having audio problem Couldn't initialize codec.
Doesn't think it's the interrupt problem, but not sure what it is.
It's as if the codec is not there. Perhaps LinuxBIOS is not turning
So, it turns out that the procedure for installing a prebuilt image is
already documented on the Wiki. My apologies for suggesting otherwise.
It's at http://wiki.laptop.org/go/Build_images
Unfortunately, even though there is a link to that page on the main
Software page, it's fairly easy to
Actually, the upgrade procedure that I documented does indeed include
Richard's image. Or more precisely,
a slightly enhanced version of it that I created.
My enhancements consisted of:
a) Moving linuxbios.rom into the initrd image so that it's not necessary
to mount the boot media
as part
Does anybody know of any efforts to support USB devices from userland?
Having to upgrade kernels just to support a random device seems like the
wrong
answer for an easy-to-use machine.
___
Devel mailing list
Devel@laptop.org
Ronald G Minnich wrote:
So, the question for me is, should an ext2 file system in a linux
embedded in a FLASH BIOS be as strict as a normal linux, or as free and
easy as grub?
I think that a bootloader should do its job of booting the kernel if
that is reasonably possible. The fact that
I verified that you can run olpcflash from the LinuxBIOS payload kernel,
instead of booting up all the way to the real kernel. As expected, it
works regardless of whether LB is loaded from SPI or PLCC.
You have to get the image from somewhere else, of course. I did it by
first reading the
Added to issue tracker as ticket #68, http://dev.laptop.org/ticket/68
http://dev.laptop.org/ticket/68#preview
Yoshiki Ohshima wrote:
Hello,
We, the member at the Viewpoints Research Institute, are working on
the project to make Squeak environment work on OLPC. (The eToys
team in
A fair amount of current software uses Esc as a bail out. For
example, it causes Firefox to abandon a stalled attempt to load a page.
Jim Gettys wrote:
Traditionally, escape has been used by many keyboard based applications
(e.g. vi, emacs), but its use is much more rare in recent
System Software telecon minutes, 2006-09-05
Jim Gettys, Jordan Crouse, Ray, Victor Chau - Quanta project mgr in charge
of OLPC RD, Jon Corbett, David Woodhouse, Chris Blizzard, Mitch Bradley,
Zephaniah Hull, Tom Sylla, Marcelo, Richard Smith
CaFe:
CaFe test hardware - one copy to Pierre Ossman
The processor on OLPC is inherently slower than the processors that are
used in modern laptops. To get ultra low power you have to sacrifice
something. It seems unlikely that we would be able to hide that
inherent speed differential by a kernel change. In the short term, I
don't know of any
As an example of what one might do: The search path resolution
mechanism could notice, while looking for file X in a list of
directories, that some of those directories don't exist. It could then
prune those entries from the path list.
___
Devel
and
Marcello.
Send dead boards to Mitch Bradley, 615 Olinda Road, Makawao, HI 96768
Mitch needs an FS2 thing. Tom knows about FS2. Need some stuff in
/etc/hotplug/usb
Chris and Ivan:
Tinderbox: Python scripts to audit software builds and installation, working
toward an automated
test suite
This reports my debug results for the failed board that was sent to me
last week.
Identification: The small bar code tag on the bottom says 0017C4000958
Failure description: The board had a sticky note that says Failed
9/6/06. There was no indication of the system configuration or the
System Software Meeting Minutes
Jordan, Chris Ball, Zephaniah, Richard, Ray, Jim, Mitch,
Status:
Zephaniah, Andres, Jim have been attacking the touchpad.
The TP runs slowly but so does a PS/2 mouse. Set-sample-rate doesn't
affect the TP.
Try throwing set-sample-rate at PS/2 mouse to see if
The likely suspects are the inductors in the various switching regulator
circuits.
They look like little black water tanks and most say 5R0R on top, but
the one near the power input says 100PF. (you might think that to be a
100 picofarad capacitor, but it's actually an inductor. Go figure.)
Carl-Daniel Hailfinger wrote:
I do have hot-melt glue, but I'm hesitant to apply it to the devel board.
Glue won't hurt an inductor
However, since inductor whine seems to be expected, can I assume it won't
be audible when the board is inside the laptop? Children have better ears
than most of
By the way, the aggressive flow control that I alluded to in my last
message is not what is shown on the picture. The picture shows a modest,
no-problem amount of flow control on the order of hundreds of *micro*seconds
The aggressive flow control that results from the queue filling up
hundreds of
Richard Smith wrote:
Your OLPC experiments, they are in an enviroment booted by LinuxBIOS
right?
Yes. The test environment is booted directly from LinuxBIOS as a
direct payload.
If so then its pointing more toward a driver issue rather than
a register config problem?
I wouldn't go that
will just fall out.
Mitch Bradley wrote:
These pictures show what happens on the OLPC board when you
a) Don't drain the queue (QueueHoldoff.png)
b) Drain the queue every so often (RateControl.png)
Conclusion: the EC is able to handle normal-rate mouse reports when the
software services
The region code, or something equivalent, is a good candidate for the
manufacturing data that Quanta will put in their special section of
FLASH. I'm thinking of a two-prong approach:
a) The manufacturing data will include a region code and the driver will
have a lookup table that maps region
I got some pushback from Quanta yesterday. They want to put the region
information in the EEPROM attached to the Marvel device, instead of in
the manufacturing data area of the SPI FLASH.
I'm not sure of all their rationale; the one point that got through the
language barrier clearly was the
System SW Meeting Minutes
2006-10-03
Jordan, Ray, Mitch, Tad, Jim, Zephaniah, Chris Blizzard
Andres, Victor, Marcello
New Action Items:
Chris Blizzard
- talk to adam and get 7.2 X server build in tree
- get sysprof, memprof, oprofile in
- prod Adam (RedHat person) to communicate with
But trac woldn't let me create a ticket even after logging in...
If we end up with 256MB of memory on the B1 boards, we'll need to change
the LB memory init code accordingly. Quanta says that they will report
the memory size via a GPIO pin.
Here is a data sheet for one of the parts Quanta
I'm very interested in the progress Mitch makes on the SPI recovery
tool for BTest boards.
So far I haven't written any code. I have looked at the serial
programming spec for the EC. The way it works is that you send commands
over the serial line that have the effect of writing to
System SW Meeting Minutes
2006-10-10
Jim, Chris Blizzard, Mitch, Ray, Jordan, Chris Ball, Andres
AI: Mitch: supply Open Firmware for people to start testing
AI: Mitch: supply Open Firmware test for B1 mouse
AI: Jim: look into what it takes to add verified state to trac
AI: Jordan: add gamma
http://firmworks.com/linux/MouseTesting.html tells how to use Open
Firmware to get a quick test of the PS/2 mouse hardware to see if it is
fried.
It involves booting Open Firmware from a USB key, so you don't have to
reflash the BIOS.
___
Devel
Jordan Crouse wrote:
On 11/10/06 19:18 +0800, [EMAIL PROTECTED] wrote:
Jordan,
Things are acting strangely on the B test board. Not seeing proper interrupts.
(The EC is different.)
Three interrupt are from 3700 EC - SCI,SWI,SERIRQ.
SCI,SWI are defined at ACPI, but we don't use ACPI.
SERIRQ
An Open Firmware image for OLPC is available for testing. See
http://firmworks.com/linux/OFWtest.html for complete instructions.
The OFW image goes on a USB key along with a regular OLPC Linux build
and LinuxBIOS boots OFW instead of Linux. You can then boot Linux from
inside OFW. (OFW can
James Cameron wrote:
On Wed, Oct 11, 2006 at 09:03:18PM -1000, Mitch Bradley wrote:
An Open Firmware image for OLPC is available for testing. See
http://firmworks.com/linux/OFWtest.html for complete instructions.
I've tried ofw-061011-1916 on both of my A-Test boards, and it worked
I tried to tear the ofw image apart, and if the files were compressed with
lzma instead of gzip, it all fits in around 220 kbyte, which means that a LinuxBIOS with
ofw could fit into
any commodety 256k BIOS used in ordenary PC's - so this could
really be a lifter for LinuxBIOS aswell.
FirmWorks has done many OFW implementations in 256K, even with the
outstanding compression available with LZMA.
I meant to say without, not with.
D'oh!
___
Devel mailing list
Devel@laptop.org
http://mailman.laptop.org/mailman/listinfo/devel
We should have the OFW building tools available in a public place within
a few weeks. At that point it would be really cool if someone would
undertake the conversion to LZMA.
___
Devel mailing list
Devel@laptop.org
Ronald G Minnich wrote:
Mitch Bradley wrote:
We should have the OFW building tools available in a public place
within a few weeks. At that point it would be really cool if someone
would undertake the conversion to LZMA.
why not just LZMA the whole image?
That makes it a little smaller
System SW Meeting Minutes
2006-10-17
Chris Blizzard, Mitch (scribe), Jim, Chris Ball (moderator), Andres,
Zephaniah, Richard, Jordan, Ray, Ted, Victor, Marcello, Ivan
*** New action items:
AI: Chris Ball to pick a PGP implementation and recommend to Quanta
AI: Mitch to deliver a new LB
The release candidate image is build. It works on btest. I'm now
testing on atest, and if that works, I will release it.
Andres Salomon wrote:
Hi Mitch,
On Tue, 2006-10-17 at 16:18 -1000, Mitch Bradley wrote:
[...]
AI: Mitch to deliver a new LB release with fixes for
trac 155 - CaFe
I deferred 136 (SIRQ) and 151 (256 M) . details will be forthcoming.
___
Devel mailing list
Devel@laptop.org
http://mailman.laptop.org/mailman/listinfo/devel
Release notes for today's interim firmware release are on the wiki at
http://wiki.laptop.org/go/OLPC_Firmware_Q2A11
Note that this release does *not* fix trac tickets 136 (serirq quiet
mode) and 151 (256 MB). The article has more details.
Also please note that this is an interim release,
There is a new version of the SPI FLASH Recovery instructions on the
wiki ( http://wiki.laptop.org/go/SPI_FLASH_Recovery ) and a
corresponding new version of the software that goes with it.
If you have already downloaded the old version (1.0) of the serial
recovery software, please discard it
Fearmongering response ... someone might think it is there to provide
another means to supply power. Can the MIC input take 240V AC RMS?
I expect that would drive the audio into clipping unless you turned the
gain way down :-)
___
Devel
I got the CaFe NAND FLASH driver and jffs2 reader integrated into Open
Firmware.
OFW mounted the jffs2 filesystem (the build 91 image, with summary
nodes) and loaded /boot/vmlinuz in just under 6 seconds.
That's using DMA access to the CaFe NAND controller. Programmed I/O
takes about twice
It's down to 4.9 seconds now. I tightened up the DMA read routine a bit.
Mitch Bradley wrote:
I got the CaFe NAND FLASH driver and jffs2 reader integrated into Open
Firmware.
OFW mounted the jffs2 filesystem (the build 91 image, with summary
nodes) and loaded /boot/vmlinuz in just under 6
System SW Meeting Minutes
2006-10-31
Richard (moderator) Mitch (scribe), Jim, Chris Ball, Zephaniah, Ray,
Ted, Chris Blizzard, Jordan, Marcello, Ivan, MFoster, Dave
Woodhouse, Andres
=== Agenda ===
Blizzard:
* have net manager and browser in build, working well. camera
It would be good to understand the market shares of different USB
ethernet devices out there (along with cost, which we care about more
than most). We don't want to have to put an unlimited number of drivers
Getting that data will be interesting in light of the fact that the
market is
Also, does it change when you turn off the machine and remove the battery?
Does any part of the system feel significantly warmer than the rest of
the unit?
Mitch Bradley wrote:
Does it seem to come from one side in particular, or from both sides
equally?
Yoshiki Ohshima wrote:
Hello
Creating something from scratch is a very advanced stage in ones
education.
The first step on the road to being a creator is to make small
modifications to something that is already there.
Even highly-experienced programmers do much more modifying than
ground-up creation.
Which suggests
That would be really cool. And if the element of cooperation could be
added, thus avoiding the implicit every man for himself lesson, it
would be even better. team vs. problem instead of human vs. human.
Ivan Krstić wrote:
Mitch Bradley wrote:
Which suggests to me that it would
The mount time for a 4G JFFS2 volume will be pretty long. Another
problem is that you will use up a lot of kernel memory for JFFS2
internal data structures with a volume that size, assuming that you put
a lot of data on it. We have pushed JFFS2 well past its design center
with the 512 MiB
The firmware does not support SD yet. It hasn't been a priority...
The power button should turn off the machine if you hold it in for 4
seconds, just like on a PC. More complicated instant power management,
like suspend, is not yet implemented. The power button does generate
keycodes, but
System SW Meeting Minutes
2006-12-12
Jordan (moderator), Jim, Mitch (scribe),
Richard, Zephaniah, Chris Blizzard,
Ted, Marcello, Chris Ball
Status:
Power supply:
Leave battery in unit. Don't remove/insert battery A/C adapter plugged.
Mitch:
Has experimental firmware that goes from
The tab key cycles between the possibilities. The chosen one is finally
activated when the alt key is released.
Khaled Hassounah wrote:
But in all current operating system releasing the tab is what makes the
selection. Why the reverse?
Khaled
Owen Williams wrote:
On Thu, 2006-12-14 at
MBurns wrote:
On 12/19/06, Thomas Müller [EMAIL PROTECTED] wrote:
366 MHZ ? then windows woun´t run anyway. Change the processor to 500
MhZ at least!
As I understand it, the Geode processor is a 500mhz that is hardware
downclocked to 333, for heat and power reasons.
The 500 number is
System SW Meeting Minutes
2006-12-19
Jordan (moderator), Chris Ball, Andres, Jim, Mitch (scribe),
Richard, J5, Zephaniah
Status:
Ronak: info about fixes for wireless boot problem supplied,
forwarded to Marcello, et al
Touchpad: Richard Andres think that the EC is at fault.
The EC
Carl-Daniel Hailfinger wrote:
Hi,
Mitch Bradley wrote:
Mitch:
Code for device tree via /proc nearly ready
Are you sure that you want to introduce new files in /proc which are
not process-related? You will never get such a change merged upstream.
I'm not sure of anything when
This is the first in what I plan as a continuing series of lessons about
Forth and Open Firmware. Each lesson will be very brief.
Lesson 0:
Point 1: Why Bother?
Forth is weird compared to most popular computer languages. Until you
learn how, it is hard to read because it is not based on
Today's prize goes to Carl-Daniel for spotting the error!
Carl-Daniel Hailfinger wrote:
Hi Mitch,
Mitch Bradley wrote:
asdf foo jello @W#$%^,T/%$ 1a2qw2 gibbet
That's 5 words. One of them is pretty strange, consisting mostly of
punctuation, but it is a word nevertheless. Any
Today's Forth lesson is on the wiki at
http://wiki.laptop.org/go/Forth_Lesson_2
A page linking to all the lessons is at:
http://wiki.laptop.org/go/Forth_Lessons
___
Devel mailing list
Devel@laptop.org
-bash-3.1# cd /proc/device-tree/
-bash-3.1# ls
#address-cells chosen [EMAIL PROTECTED]null-nvram pci
aliases cpusmmu openprom
architecturedma-ranges model options
banner-name [EMAIL PROTECTED] [EMAIL PROTECTED]
Dan Williams wrote:
On Mon, 2006-12-25 at 00:42 -1000, Mitch Bradley wrote:
-bash-3.1# cd /proc/device-tree/
-bash-3.1# ls
#address-cells chosen [EMAIL PROTECTED]null-nvram pci
aliases cpusmmu openprom
architecturedma-ranges
Today's Forth lesson is on the wiki at
http://wiki.laptop.org/go/Forth_Lesson_3
A page linking to all the lessons is at:
http://wiki.laptop.org/go/Forth_Lessons
___
Devel mailing list
Devel@laptop.org
System Software Meeting Minutes, 2006-12-26
Attendance: Ted, Richard, Mitch, Andres, Jim (sparse attendance
due to holidays)
- Jim installed new firmware (with EC B11) on some machines.
To test:
With battery fully charged (battery LED green), press button
to power off, press button
Today's Forth lesson is on the wiki at
http://wiki.laptop.org/go/Forth_Lesson_4
A page linking to all the lessons is at:
http://wiki.laptop.org/go/Forth_Lessons
___
Devel mailing list
Devel@laptop.org
serialization rules that include
binary data in many cases.
The API for /proc is just fine. I wonder if we could essentially clone
it (or even reuse most of the functions verbatim), but mount it on /ofw .
Jonathan Corbet wrote:
Mitch Bradley [EMAIL PROTECTED] wrote:
/proc/device-tree
David Miller wrote:
...
Can we please not have N different interfaces to the open-firmware
calls so that perhaps powerpc and Sparc have a chance of using this
code too?
The base interface function is callofw(), which is effectively identical
to call_prom_ret() in
I made all the changes Pekka suggested, except:
+ security = strncmp(propname, security-, 9) == 0;
+ len = 0;
Redundant assignment, no?
+ if (!security)
+ (void)callofw(getproplen, 2, 1, node,
propname, len);
That
David Miller wrote:
We don't generally export binary representation
files out of /proc or /sys, in fact this rule I believe is layed
our precisely somewhere at least in the sysfs case.
pci-sysfs exports PCI config space in binary.
___
Devel
We could of course have the interface work either on a copy of the tree
or on a real OF (though that means changing things like get_property on
powerpc and fixing the gazillions of users) but I tend to think that
working on a copy always is more efficient.
The patch that I posted creates a
OLPC System Software Teleconference Minutes
2007-01-02
Attending: Chris, Andres, Richard (moderator), Jim, Lilian, Mitch (scribe),
Jordan, Ted, Jack, Marcelo
== Status ==
=== Touchpad ===
* Seems to work now.
* Default scaling not yet adjusted.
* Do we want to add tap support?
* Post-B2, we
Today's Forth lesson is on the wiki at
http://wiki.laptop.org/go/Forth_Lesson_5
A page linking to all the lessons is at:
http://wiki.laptop.org/go/Forth_Lessons
___
Devel mailing list
Devel@laptop.org
http://mailman.laptop.org/mailman/listinfo/devel
The long-run plan for firmware update does not involve USB at all. The
OS puts a signed image on NAND FLASH, the firmware sees and validates
it, and updates the SPI FLASH.
ron minnich wrote:
We've done several tens of thousands of FLASH updates here at LANL
over the last 5 years. I also am
Segher has a modification to the devtree patch that creates a lower
level ops vector that can be implemented with callback or non-callback.
It is still being tested.
___
Devel mailing list
Devel@laptop.org
I worked with Sadiq on IRC several days ago and helped him resolve the
problem. It was caused by bad data on his USB disk.
Khaled Hassounah wrote:
Original Message
Subject:[Devel-machines] help with build the 231
Date: Tue, 16 Jan 2007 08:54:13 -0800
From: sadiq
1280x1024 and fudge X to use only a subset?
Chris Ball wrote:
Hi, thanks for thinking about this.
We would like to preconfigure qemu, bochs, vmware player or another
emulator to mimic the XO hardware as closely as possible. We're
thinking of how to emulate the mesh networking,
leave it to Richard Smith or Mitch Bradley to
follow up on that...
But the autoupdater stuff will only work when you have OpenFirmware, not
LinuxBIOS.
Actually, the autoupdater is supposed to work with LinuxBIOS too. But
you have to know how to make LinuxBIOS boot from a USB key. The way
Actually, the autoupdater is supposed to work with LinuxBIOS too. But
you have to know how to make LinuxBIOS boot from a USB key. The way
to do that is to type the ESC key when you see the progress bar. That
will get you to a graphical menu. Use the arrow keys to select the
picture of
I changed the wiki page to use C-D's wording.
___
Devel mailing list
Devel@laptop.org
http://mailman.laptop.org/mailman/listinfo/devel
Ivan Krstić wrote:
ron minnich wrote:
Question is, does OLPC want to look at this type of brick insurance?
Yes, I've just been waiting until we have a clearer idea of the final
size of OFW before asking Mitch about it.
Agreed.
For now it won't help much because the EC code,
ron minnich wrote:
On 1/30/07, Ivan Krstić [EMAIL PROTECTED] wrote:
ron minnich wrote:
flashus interruptus, due to things like power dropping, breaker
tripping, or, literally, people tripping on a power cable :-) Yes,
there's a battery, but ... maybe their battery has not charged for
some
Tim Flavin wrote:
On 1/30/07, Mitch Bradley [EMAIL PROTECTED] wrote:
We shall take reasonable precautions to reduce the probability of
reflash-brickage to an acceptable level. A level of 0 is neither
achievable nor cost-effective.
Will the production boards have pads or a header that can
Marcelo Tosatti wrote:
(moving to olpc-devel)
On Sat, Jan 27, 2007 at 02:53:07AM +0100, Arnd Bergmann wrote:
Since we now have an Open Firmware based system, the libertas firmware
could be stored inside of the firmware device tree. The most significant
advantage of that is that you can
David W Hogg wrote:
Perhaps this is a question for hardware folks, not devel folks, but
could someone out there tell me if the reflective, b/w mode for the
monitor will be 8-bit greyscale or 1-bit on-off pixels? I am writing
an application for the laptop, but it is *essential* that it run
OLPC System Software Teleconference Minutes, 2007-02-21
Present: Jim, Mitch, Zephaniah, Blizzard, Ball, j5, John Watlington,
Andres, Marcelo, Jordan,
Richard, Ivan, Ronak, Michalis
Status
* Machines are being shipped. Will start arriving soon.
* Expecting new keyboard samples
*
System SW Teleconference
2007-02-28
Jim, Mitch, Nat (Flipo), Wad, Jordan, Richard, Blizzard, Ronak, Andres,
Mary Lou
New Action Items:
AI: Blizzard to bug Marcelo to try latest TCP
AI: Andres - look into what it would take to uncache USB descriptors.
AI: MLJ - get new schematics and .brd
Notes on iperf hang problem.
Under Marcelo's supervision, I run iperf in server mode on one XO, iperf
in client mode on another XO, communicating via the internal Marvell
wireless on server XO and an external Marvell wireless dongle on the
client XO.
I used a USB analyzer to observe the USB
David Miller wrote:
Looks like the retransmits on the sender side are being
dropped at the device.
Quite possibly.
I decoded the packets in the USB trace. There is a lot of packet
reordering going on - the sequence numbers don't increase monotonically.
Subtracting out the first
Jordan Crouse wrote:
Now that Mitch has his fast path resume working,
Where working in this case currently means sometimes it wakes up, and
sometimes it doesn't :-(
Mitch
___
Devel mailing list
Devel@laptop.org
In Javier's trace, only 26 packets were transmitted, whereas in each of
my test runs, there were exactly 29 outgoing packets. So I guess the
29 number is not exactly repeatable.
I wonder what is causing the packet reordering? Does iperf explicitly
force reordering in order to stress test
Manish Regmi wrote:
Hi all,
Will ACPI be used in OLPC. I read a lot of wonderful discussions
for ACPI. Most of the discussions were against it.
http://mailman.laptop.org/pipermail/devel/2006-July/000608.html
This doc says some work is going on for ACPI...
After a week of fits and starts, I got suspend-to-RAM and subsequent
wakeup working reliably on B2 XOs. Atest and B1 systems don't work (the
wiring for the wakeup signals on Atest is screwy; B1 wakes up sometimes
but is generally unreliable, perhaps due to the power-on holdoff
required by the
Manish Regmi wrote:
On 3/3/07, Mitch Bradley [EMAIL PROTECTED] wrote:
The OLPC OS does not use the ACPI interface, nor does the OLPC firmware
support it . The OS drivers perform the appropriate power management
operations directly, with full knowledge of the behavior of the hardware
Manish Regmi wrote:
Thanks for such a nice info. I have created a wiki based on your
information http://wiki.laptop.org/go/Port_Address.
Are these ports and MMIO address fixed or generated at runtime by
firmware/bios?
The firmware establishes them, so in principle they could be
1 - 100 of 144 matches
Mail list logo
