Re: Call for Papers/Talks/Ideas! "Update.2 Mini-Conference"

[NoiseEHC]

Could somebody answer these questions?
1. What is the status of zlib -> lzo transition in jffs2?
2. What is the status of the new X architecture? (The name is DRM but I 
am not sure.)
3. What is the status of the shared page python stuff? Will the 
reference count problem will be solved in Python 3.0? Will it be 
released in the foreseeable future?

C. Scott Ananian wrote:
> On Fri, Mar 21, 2008 at 12:50 PM, Mitch Bradley <[EMAIL PROTECTED]> wrote:
>   
>> I think that Update.2 should be about 3 things:
>>  3) Performance
>>  2) Performance
>>  1) Performance
>> 
>
> Mechanisms to achieve those performance goals are worthy candidates for a 
> talk!
>  --scott
>
>   
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Call for Papers/Talks/Ideas! "Update.2 Mini-Conference"

[NoiseEHC]

It is the problem that every page will be modified (and so copied) if it 
contains a reference count so the metadata effectively cannot be shared.
http://lists.laptop.org/pipermail/devel/2007-September/006617.html
Also mentioned in (point 9.)
http://lists.laptop.org/pipermail/community-news/2007-February/43.html


Tomeu Vizoso wrote:
> On Fri, Mar 21, 2008 at 6:50 PM, NoiseEHC <[EMAIL PROTECTED]> wrote:
>   
>> Could somebody answer these questions?
>>  1. What is the status of zlib -> lzo transition in jffs2?
>>  2. What is the status of the new X architecture? (The name is DRM but I
>>  am not sure.)
>>  3. What is the status of the shared page python stuff? Will the
>>  reference count problem will be solved in Python 3.0? Will it be
>>  released in the foreseeable future?
>> 
>
> Can you explain what's that "reference count problem" or give any pointer?
>
> Thanks,
>
> Tomeu
>
>
>   
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Activity Launch Notification

[NoiseEHC]

> We've recognized that accidental activation of the Frame is a problem,
> but haven't had time to implement any of the proposed changes.  The
> simplest change we hope to add is a tiny delay, to prevent accidental
> activation when attempting to reach a button or element near the
> corners.  Even a very small delay of, perhaps, 1/5 second should help
> reduce this.  We'll likely add a slider to the control panel allowing
> this value to be adjusted from instantaneous to off.  Other ideas are
> welcome, as is any feedback that indicates the common causes for the
> accidental activation. (Is it the jumpy trackpad?  attempts to hit
> buttons in toolbars?  activities with canvases such as paint? etc.)
>   
Idea:
When the cursor moves to the corner/edge nothing should happen. If the 
cursor movement stops then for some time Sugar should watch the cursor 
and if the user moves past the screen edge by a large amount only then 
should it show the frame. Unfortunately it would require Sugar watching 
the raw touch pad input since the cursor cannot leave the screen area. 
(Alternatively the frame can be shown by just leaving the mouse cursor 
at the edge for a very long time...)
The point of this idea is that it is a simple touch pad gesture.
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Walter leaving and shift to XP.

[NoiseEHC]

Michael Stone wrote:
> On Wed, Apr 23, 2008 at 05:37:08PM -0400, Walter Bender wrote:
>   
>> I certainly don't know enough about Windows to be able to answer your
>> question from the technical perspective. 
>> 
>
> As a former Windows developer (using both proprietary APIs and Free
> APIs), I'm very confident that the collaborative user experience named
> the "Sugar UI" can be provided on Windows at what constitutes acceptable
> expense in the Windows world by any of several wholly different means
> (each with different degrees of freeness and with different technical
> folks carrying out the work).
>
>   
>> I do know that to launch an effort to port to Windows will require
>> resources above and beyond what are currently available. 
>> 
>
> True, but there are many skilled Windows developers around (including
> F/OSS developers) who might assist with the work.
>
>   
Depending how you define "Sugar", I think it can be as small an effort 
as 1 man.
Since the majority of the code is written in Python, clearly I do not 
see too much work with it.
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Sugar on Windows

[NoiseEHC]

You seem to miss this:

>> Depending how you define "Sugar"...
See?
>> I don't want to waste too much time discussing Sugar on Windows. But  
>> stating this is a 1-man effort is ridiculous -  unless you are  
>> speaking of emulating a whole Linux installation.
>>
>> 
I was talking about to port it so that anyone could run sugar activities 
on windows. Porting the whole shell is totally pointless I think, and 
replacing all the native windows apps would took years to implement.

___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Sugar on Windows

[NoiseEHC]

I sense a little misunderstanding here, probably because my English was 
not too clear. In the Windows world, the "shell" means explorer.exe 
which draws the desktop/start menu/folders. Replacing this with the 
Sugar zoom interface and frame is a little bit silly idea.

However porting sugar apps to Windows is not too hard I think. The real 
roadblock could be when journalfs will be implemented since implementing 
a Windows filesystem takes at least a year (it is a constant, no matter 
how many programmer works on that).

ps:
BTW I do not know why would it be so good for OLPC if Sugar apps would 
run on Windows. All I am saying is that it is possible.

___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: A technical assessment of porting "Sugar" to Windows.

[NoiseEHC]

You mean everything that actually calls into GDI. The kernel is fully 
thread safe and preemptive on NT.
Since as I know GTK is thread affine as well, probably it is not a problem.

Jeffrey Kesselman wrote:
> This may be obviosu to everyone, but just a note if it isnt
>
> I have a lot of experience *trying* to tlak Win32 into doing things
> other then its own way from my time in the Sun Java Performance tuning
> team.  Java has a very X inspired window system.  Retting that to run
> reliably on Windows has been a HUGE effort.
>
> The single biggest issue is this: Win32 is not a multi-threaded OS.
> What I mean by that is that, while you can throw off multipel user
> threads, it is *vital* that everything that actually calls into the
> windows kernel happen on a single thread-- the message pump thread.
> Doing anything else is either unstable or outright fails.
>
> Keep this in mind when trying to figure out the work involved in
> making your GUI work on top of it.
> ___
> Devel mailing list
> Devel@lists.laptop.org
> http://lists.laptop.org/listinfo/devel
>
>
>   
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: A technical assessment of porting "Sugar" to Windows.

[NoiseEHC]

Thanks for providing this summary!

What is not clear to me is whether we are talking about:
1. Windows on XO with Sugar
2. Sugar on Windows on any machine
3. Both
Also not clear what advantage could any variation provide to OLPC so 
probably NN could be a little more concrete about Sugar on Windows. I 
mean that supporting 1. would be good for marketing since OLPC could 
tell the potential buyer that the laptop can run M$ word even if it is 
not too practical or sane, but at least it would result in laptop sales. 
However porting all the "new" thing would make the Linux development 
became a pointless effort (not that it is possible, see later). Also 
supporting 2. could mean more Sugar developers and more kids getting 
learning software but would undermine XO sales...

> 2. Activities.
>
> The XO comes with a large number of child-oriented "activities"; see
> http://wiki.laptop.org/go/Activities.  One interpretation of "Sugar on
> Windows" is to merely port the activities to Windows, transforming
> OLPC into a pure "educational software" company.  This course is
> moderately difficult.  Python and GTK are "cross-platform", of course, but in
> practice many platform dependencies are inadvertently added to Python/GTK
> code; any developer can tell you that "cross-platform" code which has not
> actually been *tested* on another platform is unlikely to "just work"
> on it.  So some amount of work is necessary on *each* XO activity.
>
> Further, activities are written to a number of XO-specific APIs,
> including APIs for UI elements, collaboration support, and document
> storage.  The easiest course is to stub these out with
> roughly-equivalent Windows implementations of the APIs.  This would be
> sufficient to allow Windows developers to do a significant amount of
> "activity development" on a Windows machine, but the version tested on
> Windows would not actually have all of the functionality of the same
> code running on the existing Sugar/GNU/Linux stack.  OLPC's developer
> base would be expanded, but the resulting ports would be
> "developer-friendly" Windows versions of the activities, not necessarily
> a "kid-friendly" versions one would expect to deploy in schools.
>
> A more aggressive pursuit of "Activities on Windows" would result in
> completely- and fully-functioning versions of ported activities, which
> were as "kid friendly" as the versions running on Sugar/GNU/Linux.
> This would inevitably entail ports of some of the other features of
> the XO; we will discuss the difficulty of implementing these other
> features in turn below and leave the reader to sum these for
> themselves.
>   
That is what I have meant by porting Sugar to Windows. I think that 
creating a minimal version which can run most of the Python activities 
should took less than 1 man year (yes, I have read the MMM) so it is not 
too hard. Of course it would mean fixing some of them, for example the 
Sugar shell crashes on VirtualPC 2007 since it only implements 
SoundBlaster so the OLPC kernel does not recognize any sound card (and 
the Sugar shell assumes one). Also needs fixing activities to handle 
different resolutions from 1200x900 and alikes. If you ask me this is 
the path I advice taking.
> 3. Window manager: Mesh/Friend/Home view, frame, etc.
>
> This is a more difficult task than merely porting the standalone
> activities; this feature entails replacing the existing Windows file
> and application chooser mechanisms with ones which mimic that of
> Sugar/GNU/Linux.  There are two implementation possibilities: writing
> a new Windows application from scratch which mimics this interface, or
> porting the existing Python code from Sugar/GNU/Linux.
>
> Writing a new Windows application is cheap in coordination costs, but
> entails completely rewriting this part of Sugar from scratch.  This
> course would probably also make porting Activities (item #2) slightly
> more difficult, as the interfaces to wm elements (cut and paste,
> collaboration, activity startup feedback) would likely need to be
> altered to work well with a standalone "Activity chooser" Windows
> application.  It is possible that, with some cleverness, the required
> API changes could be kept small.
>
> The other alternative is porting the existing code, which implies
> running X windows, dbus, and much of the rest of the Gnome software
> stack on Windows.  Ports of most of these already exist, but they tend
> to differ from the versions we are currently using in Sugar/GNU/Linux:
> either the Windows versions lag behind the versions we are using, or
> some features don't exist/are broken, or the APIs have deliberately
> diverged to better support Windows uses.  A significant amount of
> integration work would be necessary, but the end result would be a
> system which (in theory) would require only minimal additional changes
> to activities.  It is not clear that the result will be "elegant": it
> may not be well integrated into Windows, it may be less stab

Re: A technical assessment of porting "Sugar" to Windows.

[NoiseEHC]

I have just found this link:
http://pages.cs.wisc.edu/~driscoll/fuse-nt.pdf
This is a report about a failed IFS-FUSE attempt.
They ended with a loopback SMB server what should the Sugar windows port 
should follow IMHO.


ps:
The report contains the problems writing windows FSs.
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: 15 computer science collegians looking for a project

[NoiseEHC]

Actually it is a funny proposal since without real XO machines it is a 
little bit hard to profile programs and since it seems that it is quite 
impossible nowadays to get a machine, they are out of luck. Exactly why 
does not OLPC just dump machines to developers like candy? Even if only 
1% of people do something about performance, it could be more cost 
effective than hiring 1 man for it (and the XO badly needs this work).

BTW if you referring to this
http://lists.laptop.org/pipermail/devel/2008-March/012033.html
or this
http://lists.laptop.org/pipermail/sugar/2006-September/000439.html
then these are not easy and fast things either.

Walter Bender wrote:
> Any interest in the themes/topics Tomeu outlined in his email about
> Sugar performance? Lots of interesting things to explore that would be
> of real value to the project.
>
> -walter
>
> On Mon, Apr 28, 2008 at 1:38 PM, Patrick Jahenr
> <[EMAIL PROTECTED]> wrote:
>   
>> Hi there,
>>
>>  We are a group of 15 collegians, studying 'Applied Computer Science'
>>  at the University of Applied Sciences in Iserlohn (Germany) and we
>>  are looking for a project for the OLPC within the scope of our subject
>>  "Computer-Networks". Our professor (Prof. Martin Hühne) let us choose
>>  our own topic, leading us directly to our first problem: What can
>>  we do?
>>
>>  We have to find a project fitting to this demand. We read the wiki
>>  and realised, that most of our ideas we thought we could implement
>>  are either finished software projects or simply not good enough to
>>  keep on thinking about them, because they do not fit to the OLPC-
>>  ideas. The wiki pages for Project ideas are nice, but you do not
>>  know exactly whether there is already someone working on one of the
>>  projects, or not. The Pages for Current Projects are quite nice too,
>>  but there you do not know if the people working on it actually do
>>  need help.
>>
>>  That is why we post this little request to the readers of this mailing
>>  list.
>>
>>  We are looking for either a project we can take part on, or a not-
>>  yet-implemented idea. It should not bee too big for us in order to
>>  be finished within time; it should not be too small to have something
>>  to do for 15 students. (Anyway, don't forget we need to sleep)
>>  It would be nice, but not all necessary if the project could involve
>>  some of the networking technologies of the XO, I mean, the subject
>>  is "Computer-Networks"
>>
>>  Furthermore, there is a time limit: We will have to end the project
>>  until end of June, whereas there will be the same subject for another
>>  group of students after our free period, which will have the same,
>>  or a similar task. Probably, they will go on, on our project, if
>>  its not finished.
>>
>>  Our Knowledge Base:
>>  We have good basic knowledge in C/C++, Java, some basic knowledge
>>  in scripting languages (Perl, PHP and bash) and some of us have good
>>  knowledge of Linux. A further subject called "Software-Engineering"
>>  came along with us for one and a half year.
>>  We started reading and working in Python and GTK, and saw, we can
>>  cope that.
>>
>>  We are open for all your suggestions, you can email me directly if
>>  you have a project we can take part on, but you could also just answer
>>  here, if you have a good idea which of the project ideas we could
>>  implement.
>>
>>  Thank you in advance,
>>
>>
>>  Patrick
>>
>>
>>  P.S. Please forgive me my English skills
>>
>>
>>
>>
>>
>>
>>
>>
>>  ___
>>  Devel mailing list
>>  Devel@lists.laptop.org
>>  http://lists.laptop.org/listinfo/devel
>>
>> 
> ___
> Devel mailing list
> Devel@lists.laptop.org
> http://lists.laptop.org/listinfo/devel
>
>
>   
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: 15 computer science collegians looking for a project

[NoiseEHC]

Okay, it seems that I extrapolated my attempt to the current situation.
~9 months ago I have applied for a physical machine (or for a login to a 
physical machine) since the databook missed a lot of information 
regarding assembly instruction scheduling on the Geode. I was told to 
look up the information in the databook... (H) After that I got a 
promise that there will be machines put to the net so programmers could 
login into them. Then I have got no reply if this happened or not. I 
have decided that I will not beg for a fscking machine.
Now that I know it, will request one again when I will have some spare 
time to work on it.
Thanks for the information!

ps:
BTW I have run through email several tests and later a brave man set up 
a machine for me for a week. What I have learned is that the databook 
not only is missing a lot of information but misleading and sometimes 
plainly wrong. So the measurements I have started will never be finished.

Tomeu Vizoso wrote:
> I've been told that machines will be sent to the people that provide a
> convincing description of what they want to work on.
>
> http://wiki.laptop.org/go/Developer_program#How_to_apply_for_an_XO
>
> Thanks,
>
> Tomeu
>
> On Tue, Apr 29, 2008 at 1:04 PM, Walter Bender <[EMAIL PROTECTED]> wrote:
>   
>> I was perhaps under the mistaken impression that OLPC was dumping
>>  machines to developers like candy again after a hiatus due to a
>>  logistical snafu. Hmm.
>>
>>  -walter
>>
>>  2008/4/29 NoiseEHC <[EMAIL PROTECTED]>:
>>
>>
>> 
>>> Actually it is a funny proposal since without real XO machines it is a
>>>   
>>  >  little bit hard to profile programs and since it seems that it is quite
>>  >  impossible nowadays to get a machine, they are out of luck. Exactly why
>>  >  does not OLPC just dump machines to developers like candy? Even if only
>>  >  1% of people do something about performance, it could be more cost
>>  >  effective than hiring 1 man for it (and the XO badly needs this work).
>>  >
>>  >  BTW if you referring to this
>>  >  http://lists.laptop.org/pipermail/devel/2008-March/012033.html
>>  >  or this
>>  >  http://lists.laptop.org/pipermail/sugar/2006-September/000439.html
>>  >  then these are not easy and fast things either.
>>  >
>>  >
>>  >
>>  >  Walter Bender wrote:
>>  >  > Any interest in the themes/topics Tomeu outlined in his email about
>>  >  > Sugar performance? Lots of interesting things to explore that would be
>>  >  > of real value to the project.
>>  >  >
>>  >  > -walter
>>  >  >
>>  >  > On Mon, Apr 28, 2008 at 1:38 PM, Patrick Jahenr
>>  >  > <[EMAIL PROTECTED]> wrote:
>>  >  >
>>  >  >> Hi there,
>>  >  >>
>>  >  >>  We are a group of 15 collegians, studying 'Applied Computer Science'
>>  >  >>  at the University of Applied Sciences in Iserlohn (Germany) and we
>>  >  >>  are looking for a project for the OLPC within the scope of our 
>> subject
>>  >  >>  "Computer-Networks". Our professor (Prof. Martin Hühne) let us choose
>>  >  >>  our own topic, leading us directly to our first problem: What can
>>  >  >>  we do?
>>  >  >>
>>  >  >>  We have to find a project fitting to this demand. We read the wiki
>>  >  >>  and realised, that most of our ideas we thought we could implement
>>  >  >>  are either finished software projects or simply not good enough to
>>  >  >>  keep on thinking about them, because they do not fit to the OLPC-
>>  >  >>  ideas. The wiki pages for Project ideas are nice, but you do not
>>  >  >>  know exactly whether there is already someone working on one of the
>>  >  >>  projects, or not. The Pages for Current Projects are quite nice too,
>>  >  >>  but there you do not know if the people working on it actually do
>>  >  >>  need help.
>>  >  >>
>>  >  >>  That is why we post this little request to the readers of this 
>> mailing
>>  >  >>  list.
>>  >  >>
>>  >  >>  We are looking for either a project we can take part on, or a not-
>>  >  >>  yet-implemented idea. It should not bee too big for us in order to
>>  >  >>  be finished within time; it should not be too small to have something
>>  >  >>  to do for 15 students. (Anyway, don't forget we need to sleep)
>

Re: 15 computer science collegians looking for a project

[NoiseEHC]

>> Okay, it seems that I extrapolated my attempt to the current situation.
>>  ~9 months ago I have applied for a physical machine (or for a login to a
>>  physical machine) since the databook missed a lot of information
>>  regarding assembly instruction scheduling on the Geode. I was told to
>>  look up the information in the databook...
>> 
>
> I wonder whether that is one of the reference manuals that I worked on
> at National Semiconductor in 2000, before they sold the Geode line to
> AMD. I mostly remember long tables of register bit functions and the
> occasional timing diagram. Naturally, they never let me near a chip.
>   
Now that is funny. The Geode databook is so bad that I took the time and 
looked for the MediaGX databook and found it in some cache (maybe Google 
cache?). That contains more information that Geode's one. So my 
conspiracy theory is that some marketing manager went through the 
databook when AMD acquired Cyrix/National Semiconductor and deleted 
every piece of information which would suggest that the processor is 
slow. Of course, I do not believe in this theory. Okay, sometimes seems 
to be true. But it is too stupid. Hmmm... :)
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: 15 computer science collegians looking for a project

[NoiseEHC]

I am sorry but I cannot recall a message from you about this (can be 
eaten by the spam filter or simply lost?). I mean that you have always 
answered my questions about the video driver and similar thing (thanks 
again) just not those processor core questions so I supposed that it is 
not your area. My first questions were:
http://lists.laptop.org/pipermail/devel/2007-May/005227.html
They were written at a time when I was thinking that the spec is just 
missing some data here and there.
Now my simple question is (better call it a request):
On this page
http://wiki.laptop.org/go/Geode_LX
I have named some instructions as "Synchronized ops" (in the MMX 
section). Are those real or did I mismeasured something? If those are 
real then would somebody from AMD just go through the databook and fix 
the instruction clock cycle numbers? Because in that case it is sure 
that they do not match reality and clearly I have better things to do 
than measuring clock cycles. Also the legend is clearly wrong in several 
cases so probably that would need checking too (like on page 668 note 4 
talks about 3DNOW ops in the table about FP ops). Also there is 
absolutely no info about L2 cache miss penalties or mispredicted jumps 
or about the pipeline stages of the FP unit.
See, all I would like to have is enough data that when I look at 
assembly code I could approximately calculate how many clock cycles will 
be consumed. Nothing more and nothing less.
Thanks!


> I will note that when I asked you for more information as to what you
> thought was missing or incorrect, I got no response.  If you tell me
> _exactly_ what you think you need, I can try to get the information.
> However, I do need exact demands.  General "you suck, tell us more" 
> requests are not effective arguments with the documentation gatekeepers.
>
> Jordan
>
>   
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: 15 computer science collegians looking for a project

[NoiseEHC]

On 29/04/08 17:41 +0200, NoiseEHC wrote:
  

On this page
http://wiki.laptop.org/go/Geode_LX
I have named some instructions as "Synchronized ops" (in the MMX 
section). Are those real or did I mismeasured something?



That section is very difficult to understand.  I'm not sure which
operations you have invented this name for.
  
As you probably have already noticed I am not a native English speaker 
(and neither learned advanced English in school, just picked it up). 
What I wanted to write in that section, every MMX op, whose 
source/destination operand is an integer register (and not a MOV), will 
consume absolutely different clock cycles than 2 (2 is listed for almost 
every MMX op in the databook, at least in my version). Is it real?
If those are 
real then would somebody from AMD just go through the databook and fix 
the instruction clock cycle numbers? Because in that case it is sure 
that they do not match reality and clearly I have better things to do 
than measuring clock cycles. 



Clearly you must have some basis for assuming that the numbers are
wrong, so you must have done some measurement.  I consulted the
secret documentation that you claim I am withholding from you, 
and the timings there are the same as in the datasheet.  I believe that

you are correct in that these are the clock counts for the instruction to
go through the FPU and don't include the stall time for the pipeline
to clear up.
  
There is a "Test results" section in that page. The first two test were 
conducted via email. I have emailed to this list test programs and there 
were people who run them and emailed back the result. Especially the 
first test has some stupid bugs because I wrote them essentially blind. 
The third one is the result of my session logged into a physical 
machine. It can be that only this "stall time" is missing from the 
databook but the fact is that I as a programmer am not interested in how 
many clock cycles does the FPU take to execute some internal operation 
(which seems the databook to list) but I would like to know the real 
time consumed.



I am not a silicon designer, so I'm not the final word on if they are
correct or not, but at least that should prove that there isn't a
massive marketing conspiracy to hide the details of the processor
from our customers.  If they are lying to you, they are lying to me,
and they're not lying to me.

  
This conspiracy thing was not serious, I have used a smiley at the end. 
However from my perspective there is no difference if there is some 
conspiracy or if there is not. In fact what I think is either that I am 
mistaken and made some errors measuring this or the technical writer 
made mistakes years ago and nobody cared to fix it.
  
Also the legend is clearly wrong in several 
cases so probably that would need checking too (like on page 668 note 4 
talks about 3DNOW ops in the table about FP ops).



That is an mistake - I have let the technical writer know about it.
  

Thanks!
Another error:
On page 631 it talks about this:
Conditional jump taken | Conditional jump not taken. (e.g., "4|1" = four 
clocks if jump taken, one clock if jump not taken).

It is never used in the opcode table.
  
absolutely no info about L2 cache miss penalties or mispredicted jumps 
or about the pipeline stages of the FP unit.



I don't have any information about L2 cache miss penalties, but they 
are easy to calculate. Please see:


http://homepages.cwi.nl/~manegold/Calibrator/
  
Could you run on your machine and share the results? Currently I do not 
have access to an XO.

I will talk to somebody about documenting the FP unit pipeline.
It does handle 1 instruction per clock from the integer unit.
In practice we know that two floating point instructions back to
back will stall the IU.  I can also tell you that it is optimized
for single precision, so double precision is handled by microcode
and needs to go through the path again. 

  

Thanks!
I would also like to know how many ALU units does the FPU have? I mean 
FMUL costs 1, PFMUL costs 2. Is it because it only has 1 multiply unit 
and it executes PFMUL serially? If that is the case, does that mean that 
the 3DNOW support is only compatibility and will not be faster than 
simple FP?
  
See, all I would like to have is enough data that when I look at 
assembly code I could approximately calculate how many clock cycles will 
be consumed. Nothing more and nothing less.



You have nearly all the information you need, and you can collect the
additional information the same way we do, with careful analysis and
measurement.  In fact, Bernie and Vladimir Makarov have done a lot
of work already in this area, resulting in the Geode specific
code for gcc 4.2.0 and glibc.  Perhaps you can work with them to figure
out the finer details of the FPU scheduling.  I'm sure they would
appreciate it.

Jordan

  



___

Re: 15 computer science collegians looking for a project

[NoiseEHC]

> I still don't understand what you mean, but the clock timings that are
> in the data sheet, are the same ones on my documentation.  You would have
> to find somebody more skilled then I to debate if they are correct or not.
>
>   
Okay, here are some exaples (intel notation):
First, the 2 cycles listed in the MMX section is not true. An MMX op 
costs 6 cycles, that is the latency. It is true that there can exist 
some stage in the undocumented FPU pipeline which costs 2 cycles, but 
that defines just the throughput (the MediaGX databook at least listed 
both numbers.)

example1 (this example is not correct since all paddb will run for 6 
cycles it is just hidden from the IU):
paddb mm0,mm1 ; 2
paddb mm2,mm3 ; 2
paddb mm4,mm5 ; 2

example2 (we do not care about movd now):
movd mm0,eax
paddb mm0,mm1 ; 6
paddb mm0,mm2 ; 6
paddb mm0,mm3 ; 6
movd eax,mm0

Now that only means that the Geode is at most half as fast than a 
Pentium MMX but actually it is not that bad. The problem is that this 6 
cycle latency is not documented, neither that the 2 cycles listed in the 
databook are throughput numbers. Now we reach the land of "synchronized 
ops".

example3 (I cannot garantee that the numbers are correct since I cannot 
test it):
movd mm0,eax ; 5
paddb mm0,mm1 ; 6
paddb mm0,mm2 ; 6
paddb mm0,mm3 ; 6
movd eax,mm0 ; 10

Now from where this 5 and 10 come is a mystery but nevertheless was a 
shocking revelation. Effectively it makes MMX almost unusable on the 
Geode. My simple question is: is 3DNOW similarly unusable? Now it is 
possible that I messed up the measurements but if I am right then 
something is clearly wrong with the documentation. If your internal 
documentation lists the same numbers then that internal documentation is 
wrong as well. Even if the numbers are correct (2 in the case of paddb) 
it does not mean that the databook is usable since a programmer has to 
work with the number 6 in this case which is not listed anywhere (not to 
mention the numbers 5 and 10).


Now for the other parts of your message:
Are you suggesting that I should measure everything on the Geode? 
Because I have written my message because I do not want to measure in 
the first place. Not that I have better things to do than checking clock 
cycles (normally the documentation matches reality), but this 
measurement will not even be finished (cannot be done). You know I am 
not sure what should I say.

___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: 15 computer science collegians looking for a project

[NoiseEHC]

The keyboard does not matter since I will have to use an external 
keyboard anyway (I have big hands). What is important is that I have to 
be able to reflash the machine to use English and it has to be at least 
a B3 or B4 (or G1G1 or whatever with the Geode LX). Shall I apply again?

Jim Gettys wrote:
> I checked this morning now that I'm functioning again; the developer's
> program should be moving again, and volumes available should be much
> higher than before mass production (particularly if you can tolerate the
> Arabic keyboard, where we have stock that can be redirected right now).
>
> Sorry to be slow to respond; I came down sick with a nasty cold bug
> last week on top of recovering from surgery.  I'm only now wading
> through 1200 back unread mail messages.
>- Jim
>
>
>
> On Tue, 2008-04-29 at 07:04 -0400, Walter Bender wrote:
>   
>> I was perhaps under the mistaken impression that OLPC was dumping
>> machines to developers like candy again after a hiatus due to a
>> logistical snafu. Hmm.
>>
>> -walter
>>
>> 2008/4/29 NoiseEHC <[EMAIL PROTECTED]>:
>> 
>>> Actually it is a funny proposal since without real XO machines it is a
>>>  little bit hard to profile programs and since it seems that it is quite
>>>  impossible nowadays to get a machine, they are out of luck. Exactly why
>>>  does not OLPC just dump machines to developers like candy? Even if only
>>>  1% of people do something about performance, it could be more cost
>>>  effective than hiring 1 man for it (and the XO badly needs this work).
>>>
>>>  BTW if you referring to this
>>>  http://lists.laptop.org/pipermail/devel/2008-March/012033.html
>>>  or this
>>>  http://lists.laptop.org/pipermail/sugar/2006-September/000439.html
>>>  then these are not easy and fast things either.
>>>
>>>
>>>
>>>  Walter Bender wrote:
>>>  > Any interest in the themes/topics Tomeu outlined in his email about
>>>  > Sugar performance? Lots of interesting things to explore that would be
>>>  > of real value to the project.
>>>  >
>>>  > -walter
>>>  >
>>>  > On Mon, Apr 28, 2008 at 1:38 PM, Patrick Jahenr
>>>  > <[EMAIL PROTECTED]> wrote:
>>>  >
>>>  >> Hi there,
>>>  >>
>>>  >>  We are a group of 15 collegians, studying 'Applied Computer Science'
>>>  >>  at the University of Applied Sciences in Iserlohn (Germany) and we
>>>  >>  are looking for a project for the OLPC within the scope of our subject
>>>  >>  "Computer-Networks". Our professor (Prof. Martin Hühne) let us choose
>>>  >>  our own topic, leading us directly to our first problem: What can
>>>  >>  we do?
>>>  >>
>>>  >>  We have to find a project fitting to this demand. We read the wiki
>>>  >>  and realised, that most of our ideas we thought we could implement
>>>  >>  are either finished software projects or simply not good enough to
>>>  >>  keep on thinking about them, because they do not fit to the OLPC-
>>>  >>  ideas. The wiki pages for Project ideas are nice, but you do not
>>>  >>  know exactly whether there is already someone working on one of the
>>>  >>  projects, or not. The Pages for Current Projects are quite nice too,
>>>  >>  but there you do not know if the people working on it actually do
>>>  >>  need help.
>>>  >>
>>>  >>  That is why we post this little request to the readers of this mailing
>>>  >>  list.
>>>  >>
>>>  >>  We are looking for either a project we can take part on, or a not-
>>>  >>  yet-implemented idea. It should not bee too big for us in order to
>>>  >>  be finished within time; it should not be too small to have something
>>>  >>  to do for 15 students. (Anyway, don't forget we need to sleep)
>>>  >>  It would be nice, but not all necessary if the project could involve
>>>  >>  some of the networking technologies of the XO, I mean, the subject
>>>  >>  is "Computer-Networks"
>>>  >>
>>>  >>  Furthermore, there is a time limit: We will have to end the project
>>>  >>  until end of June, whereas there will be the same subject for another
>>>  >>  group of students after our free period, which will have the same,
>>> 

Re: 15 computer science collegians looking for a project

[NoiseEHC]

I have registered on said page but when I click to my projects it goes 
to wiki.laptop.org and all I can see an empty page. Are those people 
over viewing my application will decide by using that empty page? Since 
I cannot see any change tracking feature on the wiki, I am not sure how 
could people track/discuss new projects either.

One of the textboxes is not too clear to me on the new project page.
1. Description of your experience, both with hardware and software. - Am 
I supposed to write about my work experience regarding the new project? 
Why shall I define it again if I already did it on my registration form? 
Or is it about something else? (I am not a native speaker.)



Kim Quirk wrote:
> If you have applied to the developer's program a few months ago, 
> please check to see if your application was transferred into the new 
> system:
>
> http://projectdb.olpc.at/
>
> If it is not there, I'm afraid you will have to apply again. The idea 
> with this new system is that there is a community who will help 
> evaluate the best uses for laptops. Please note that this program is 
> for single or two laptop requests, not quantities, so if you request 
> quantities of laptops, it will probably be refused.
>
> SJ, Aaron,
> Can you provide an overview of how the new system works? How 
> applications get reviewed and how often ship orders will be sent out? 
> How people get feedback on their request, etc?
>
> Thanks!
> Kim
>
>
> On Mon, May 5, 2008 at 11:36 AM, NoiseEHC <[EMAIL PROTECTED] 
> <mailto:[EMAIL PROTECTED]>> wrote:
>
> The keyboard does not matter since I will have to use an external
> keyboard anyway (I have big hands). What is important is that I
> have to
> be able to reflash the machine to use English and it has to be at
> least
> a B3 or B4 (or G1G1 or whatever with the Geode LX). Shall I apply
> again?
>
> Jim Gettys wrote:
> > I checked this morning now that I'm functioning again; the
> developer's
> > program should be moving again, and volumes available should be much
> > higher than before mass production (particularly if you can
> tolerate the
> > Arabic keyboard, where we have stock that can be redirected
> right now).
> >
> > Sorry to be slow to respond; I came down sick with a nasty
> cold bug
> > last week on top of recovering from surgery. I'm only now wading
> > through 1200 back unread mail messages.
> > - Jim
> >
> >
> >
> > On Tue, 2008-04-29 at 07:04 -0400, Walter Bender wrote:
> >
> >> I was perhaps under the mistaken impression that OLPC was dumping
> >> machines to developers like candy again after a hiatus due to a
> >> logistical snafu. Hmm.
> >>
> >> -walter
> >>
> >> 2008/4/29 NoiseEHC <[EMAIL PROTECTED]
> <mailto:[EMAIL PROTECTED]>>:
> >>
> >>> Actually it is a funny proposal since without real XO machines
> it is a
> >>> little bit hard to profile programs and since it seems that it
> is quite
> >>> impossible nowadays to get a machine, they are out of luck.
> Exactly why
> >>> does not OLPC just dump machines to developers like candy?
> Even if only
> >>> 1% of people do something about performance, it could be more cost
> >>> effective than hiring 1 man for it (and the XO badly needs
> this work).
> >>>
> >>> BTW if you referring to this
> >>> http://lists.laptop.org/pipermail/devel/2008-March/012033.html
> >>> or this
> >>> http://lists.laptop.org/pipermail/sugar/2006-September/000439.html
> >>> then these are not easy and fast things either.
> >>>
> >>>
> >>>
> >>> Walter Bender wrote:
> >>> > Any interest in the themes/topics Tomeu outlined in his
> email about
> >>> > Sugar performance? Lots of interesting things to explore
> that would be
> >>> > of real value to the project.
> >>> >
> >>> > -walter
> >>> >
> >>> > On Mon, Apr 28, 2008 at 1:38 PM, Patrick Jahenr
> >>> > <[EMAIL PROTECTED]
> <mailto:[EMAIL PROTECTED]>> wrote:
> >>> >
> >>> >> Hi there,
> >>> >>
> >>> >> We are a group of 15 collegians, studying 'Applied Computer
> Science'

Re: 15 computer science collegians looking for a project

[NoiseEHC]

Edward Cherlin wrote:
> On Mon, May 5, 2008 at 11:38 AM, NoiseEHC <[EMAIL PROTECTED]> wrote:
>   
>> I have registered on said page but when I click to my projects it goes
>>  to wiki.laptop.org and all I can see an empty page. Are those people
>>  over viewing my application will decide by using that empty page? Since
>>  I cannot see any change tracking feature on the wiki, I am not sure how
>>  could people track/discuss new projects either.
>> 
>
> It's a Wiki page. You are expected to edit it. The Wiki page change
> tracking is called History.
>
> I might not have understood your questions correctly. Feel free to ask again.
>   
Every information I have entered to the project is lost. Shall I type in 
those informations twice?
Can you list the projects created in the last 7 days? If cannot, then 
how can somebody follow new projects?

>   
>>  One of the textboxes is not too clear to me on the new project page.
>>  1. Description of your experience, both with hardware and software. - Am
>>  I supposed to write about my work experience regarding the new project?
>>  Why shall I define it again if I already did it on my registration form?
>>  Or is it about something else? (I am not a native speaker.)
>> 
>
> Any previous experience, whether directly related to this project or not.
>
>   
It does not answer my question.
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: 15 computer science collegians looking for a project

[NoiseEHC]

>> One of the textboxes is not too clear to me on the new project page.
>> 1. Description of your experience, both with hardware and software. - 
>> Am I supposed to write about my work experience regarding the new 
>> project?
> experience in general. Why are you the right person to do that project ?
Probably some active wording would help. For example: "Why are you the 
right person to do that project?"
:)
>
>> Why shall I define it again if I already did it on my registration 
>> form? Or is it about something else? (I am not a native speaker.)
>>
> ah, ok, the redundancy bug maybe.
>
>
> The rest was clear for you?
Was clear. I have just paused at the last textbox and thought that it 
should have to have some different meaning from the profile's one. Then 
I thought that the intent should be evident for a native speaker and 
that is why I have asked.
One little thing (took 20 sec for me): at the mobile phone textbox the 
sample starts with +1. Now I suppose that it is the USA. Since I have 
never ever called the USA in my life it was a little bit strange that 
the country code (if it is the country code) has only 1 digit. Probably 
you should list an example with a 2 digit code.
>
>
> a.
>
>
>
>
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: 15 computer science collegians looking for a project

[NoiseEHC]

Okay, today I have checked whether my "projects" got moved to the wiki 
but now I can see only the second one (and it is not in the wiki either).

Aaron Kaplan wrote:
>
> On May 5, 2008, at 8:38 PM, NoiseEHC wrote:
>
>> I have registered on said page but when I click to my projects it 
>> goes to wiki.laptop.org and all I can see an empty page. Are those 
>> people over
>
> dang! thanks, you found a bug. Actually a regression. It already worked.
> Will have to investigate.
>
>> viewing my application will decide by using that empty page? Since I 
>> cannot see any change tracking feature on the wiki, I am not sure how 
>> could people track/discuss new projects either.
>
> The idea is that the bot in projectdb.olpc.at reads the Talk: page 
> about the project proposal on the wiki.l.o.
> That way - people will be able to talk and discuss a project 
> application on a community level. Let's say in a mediawiki style.
> Then these comments get fed back into the projectdb and the persons 
> who really have the "vote" bit set can decide about a project proposal.
>
> But the main concern so far was: 1) do we get the data in correctly 
> and 2) can we export it to B* ?
>
> These community features can be implemented in hack sessions on 
> weekends usually.
>
>>
>> One of the textboxes is not too clear to me on the new project page.
>> 1. Description of your experience, both with hardware and software. - 
>> Am I supposed to write about my work experience regarding the new 
>> project?
> experience in general. Why are you the right person to do that project ?
>
>> Why shall I define it again if I already did it on my registration 
>> form? Or is it about something else? (I am not a native speaker.)
>>
> ah, ok, the redundancy bug maybe.
>
>
> The rest was clear for you?
>
>
> a.
>
>
>
>
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: XO-2 software plans

[NoiseEHC]

There was a thread about the X driver here:
http://lists.laptop.org/pipermail/devel/2007-September/006565.html

Because there were much more pressing things to do than rewriting the X 
driver by Bernardo this project stalled.
However it is one of my project ideas on the developer program so 
hopefully one day any program will be able to use the hardware scaler of 
the CPU.
(Of course the documentation does not mention if the scaler is faster or 
not, or does it trash the cache as software copying so it must be 
measured...)

Jim!
Could you be a bit more specific than "Profiling is in order." please? 
What is currently happening and will it make moot my efforts? Thanks!


Bert Freudenberg wrote:
> On 24.05.2008, at 03:41, Jim Gettys wrote:
>
>   
>> On Fri, 2008-05-23 at 17:17 -0700, Alex Belits wrote:
>> 
>>> Jim Gettys wrote:
>>>   
 Bert...

 Part of the problem is the X driver model is pretty broken,  
 causing much
 more to be done in software than should be necessary; and it isn't  
 clear
 we're even using X efficiently at the moment...  The driver stuff is
 getting fixed (in general in X: this is the EXA/DRI2 work);  
 profiling of
 our entire software stack is in order to see where our real  
 problems are
 at the moment.
 
>>> EXA? DRI2?
>>>
>>> Don't you end up using Cairo through GTK as the main layer that  
>>> almost
>>> everything goes through, so everything below has any importance  
>>> only as
>>> long as Cairo uses it efficiently?
>>>   
>> You can abuse Cairo, rather than use it.
>>
>> And we use it sometimes in ways other than strictly through GTK+: e.g.
>> the canvas.
>>
>> Profiling is in order.
>>
>> Also, note I was replying to Bert Freudenberg, one of the Squeak/etoys
>> folks.  They don't go through the GTK/cairo stack, except for the
>> activity decoration.
>>- Jim
>> 
>
>
> Ah, well, for one Squeak/Etoys could potentially make use of hardware  
> acceleration, and secondly, Squeak is not my only interest :)
>
> But profiling would be in order indeed. Wish there was time for that ...
>
> - Bert -
>
>
> ___
> Devel mailing list
> Devel@lists.laptop.org
> http://lists.laptop.org/listinfo/devel
>
>
>   
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Need advice to upgrade

[NoiseEHC]

Hello,

I have just received my XO via the Developers Program. The machine is 
working nicely, but there is a problem connecting to my wireless router 
(Belkin Pre-N F5D8230-4) which uses WPA-PSK with AES encription. The XO 
just asks for the password over and over again. The strange thing that 
connecting did succeed several days ago so I could request a developer 
key but since then nothing... (Today I have switched to channel 11 from 
channel AUTO but no effect.)

1. So my question is: do I need to upgade to some more recent FW/build? 
Is it a known problem with build 656 (stream ship.2) which I have?

Normally I would not spam the devel list with this support request but I 
have some more questions:

2. I could not find anywhere what the LEDs do on the XO. It can be that 
I am just stupid or blind but there is an option that it was so evident 
for every 1CC employee that the explanation was missed somehow. The only 
thing that I could find is this:
http://wiki.laptop.org/go/Image:Drawing75c1.jpg
I can deduce all the LEDs' meaning except the leftmost two:
a. The lollipop LED is Wireless acquisition. What exactly does it mean?
b. The tie-fighter LED is Wireless activity. Is it send or receive or 
both? Is it the wireless chip or the TCP/IP stack or what?

3. Some time ago cscott told me how to kill stuff if I want to measure 
speed.
telinit 3
ifconfig msh0 down
ifconfig eth0 down
After that the lollipop and tie-figher are both blinking randomly (and I 
got msh0: link becomes ready message to the console). Does it still 
switches off things? If not what should I use?

4. How is the rebasing on FC9 is going?
I am just asking because sooner or later I will finish the Geode docu 
and will start to develop drivers. Since I am not a Linux guru, I could 
not setup my FC7 VirtualPC image to compile the kernel, I was able only 
to write kernel modules with FC8. So if it is possible, I would avoid 
fighting the kernel build process on FC6 or FC7 and so I would like to 
know what to expect?

5. What image should I upgrade to? What is that "faster build"?
I currently only using the text-mode console so it is not a problem for 
me if no activity works or something like that (if the wireless works).

Thank you!

___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: etoys implementation

[NoiseEHC]

>> There's also a warning at http://wiki.squeak.org/squeak that if you
>> want to run eToys, you need to run a different version of Squeak than
>> everybody else.
>> 
>
>   *That* is Etoys.  What is wrong with it?
>   
Just out of curiosity:
Exactly how is it different from vanilla squeak? (If there is such a 
thing at all.)
Is it a different VM, or just a different distribution since it has a 
different binary blob?
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: etoys implementation

[NoiseEHC]

Thanks!


>   Why do you refer it to as "binary blob"?
>
>   

That was the first word jumped into my mind. All I know is that it is 
not in CVS or GIT so nothing serious.
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Inappropriate use of private meetings & lists.

[NoiseEHC]

>
> There are top-down decisions being made by a few people that drive the 
> direction of OLPC. These decisions are not waiting for consensus, and 
> they are made by a small number of people. I don't believe this is 
> going to change (at least not in the short term).
>
I, personally, do not care who makes decisions, as long as he is smart 
and makes "good" decisions. The problem is that usually I am not 
notified of such decisions.

For example, the move from 32 bit to 16 bit frame buffer just happened 
silently. One day people debated its merits on this list without 
reaching any conclusions. Some months later it was 16 bit. Nobody told it.

A similar thing is that I am trying to get an answer to the question 
whether the XO image will be moved to LZO compression or not when we 
will be rebased to F9. The reason is that I am currently tuning the LZO 
decompression code and want to know whether this effort is moot or not. 
Most likely this was already decided by somebody, and frankly I am not 
qualified enough to debate this decision, I just want to know what it is.

Note that I had to reply all to this message and then delete all 
recipients and then replace CC devel with TO devel. The reason is that 
this devel list is not set up that the reply-to would be the devel list, 
and if I do not delete others then the conversation often slips outside 
devel. See, even the main mailing list is configured wrong.
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Inappropriate use of private meetings & lists. (reply to).

[NoiseEHC]

1. olpc games sets the reply-to
2. BTW this recommendation does not make my point wrong (eg that the 
current setting makes harder to keep conversations on the devel list)

Jim Gettys wrote:
> Here's mailman's recommendation about replyto...
>
> http://lists.laptop.org/admin/devel/?VARHELP=general/reply_goes_to_list
>  - Jim
>
>
>
>
> On Wed, 2008-07-02 at 09:19 +0200, NoiseEHC wrote:
>   
>>> There are top-down decisions being made by a few people that drive the 
>>> direction of OLPC. These decisions are not waiting for consensus, and 
>>> they are made by a small number of people. I don't believe this is 
>>> going to change (at least not in the short term).
>>>
>>>   
>> I, personally, do not care who makes decisions, as long as he is smart 
>> and makes "good" decisions. The problem is that usually I am not 
>> notified of such decisions.
>>
>> For example, the move from 32 bit to 16 bit frame buffer just happened 
>> silently. One day people debated its merits on this list without 
>> reaching any conclusions. Some months later it was 16 bit. Nobody told it.
>>
>> A similar thing is that I am trying to get an answer to the question 
>> whether the XO image will be moved to LZO compression or not when we 
>> will be rebased to F9. The reason is that I am currently tuning the LZO 
>> decompression code and want to know whether this effort is moot or not. 
>> Most likely this was already decided by somebody, and frankly I am not 
>> qualified enough to debate this decision, I just want to know what it is.
>>
>> Note that I had to reply all to this message and then delete all 
>> recipients and then replace CC devel with TO devel. The reason is that 
>> this devel list is not set up that the reply-to would be the devel list, 
>> and if I do not delete others then the conversation often slips outside 
>> devel. See, even the main mailing list is configured wrong.
>> ___
>> Devel mailing list
>> Devel@lists.laptop.org
>> http://lists.laptop.org/listinfo/devel
>> 
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Inappropriate use of private meetings & lists. (reply to).

[NoiseEHC]

> On Wed, 2008-07-02 at 17:47 +0200, NoiseEHC wrote:
>   
>> 1. olpc games sets the reply-to
>> 
>
> Sounds like it should get fixed...
>
>   
>> 2. BTW this recommendation does not make my point wrong (eg that the 
>> current setting makes harder to keep conversations on the devel list)
>> 
>
> And screws many people who use reply to to get their mail sent back to
> their main mailbox
>   - Jim
>
>   
See, I do not want to debate whether the current reply to setting is a 
"wrong" or "right" thing (I have no idea what this reply to usage you 
are referring to is about). I would like just remind you that the real 
question is that if this "right" reply to setting is more or less 
important than keeping discussions on the devel list or not. I 
personally think that the latter is more important IMHO.
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Home View appearance

[NoiseEHC]

Could we at least select the ring view by default? I see the free form 
view as unnecessary crap because the last thing I want is to reorder 
icons by using the touchpad. Did anybody actually try that exercise?
Also the list view has a bug, that it does not have an XO icon which can 
be clicked to get a shutdown/cpanel menu item. My suggestion is that 
probably the Neighborhood/Group view's XO icon should have the same menu.

ps:
And of course, the only thing that the free form can be used is to 
reorder the icons like this:
http://www.thewebsiteisdown.com/
:)

Mikus Grinbergs wrote:
>> There is certainly not consensus regarding the merits of the
>> "free-form" Home View, but it is being accepted upstream, AFAIK.
>> 
>
> Sorry,  but I couldn't resist :
>
> The principal merit I see in the "free-form" Home View is that it 
> makes the screen look more like what is familiar (reassuring) to a 
> Windows user.
>
> Last time I tried a relatively recent Joyride, the non-list Home 
> View gave me a choice of whether I wanted to see "free-form" or 
> "ring".  As long as upstream continues to give me that choice of 
> optionally selecting "ring view", I don't care what else is decided.
>
> mikus
>
> ___
> Devel mailing list
> Devel@lists.laptop.org
> http://lists.laptop.org/listinfo/devel
>
>
>   
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Home View appearance

[NoiseEHC]

Now I see that the XO icon thing is a dupe:
https://dev.laptop.org/ticket/7043

I do not see any connection with 7430 however. What I wanted to say is 
that if you really want to keep that useless Free Form view then at 
least select the Circle View by default (for the Circle vs Free Form).

Another problem is that the Activity button (the 4th zoom level) can 
select the Journal if that was the last active activity. It has an own 
button so I cannot see any reason why it has to be the way it is. It is 
an annoyance that when I download several things then go from Browse to 
the Journal, open the pdf in Read, close Read and click the button it 
switches back to Journal and not Browse. It should not treat Journal as 
an application.

Tomeu Vizoso wrote:
> On Tue, Jul 8, 2008 at 9:37 AM, NoiseEHC <[EMAIL PROTECTED]> wrote:
>   
>> Could we at least select the ring view by default? I see the free form
>> view as unnecessary crap because the last thing I want is to reorder
>> icons by using the touchpad. Did anybody actually try that exercise?
>> Also the list view has a bug, that it does not have an XO icon which can
>> be clicked to get a shutdown/cpanel menu item. My suggestion is that
>> probably the Neighborhood/Group view's XO icon should have the same menu.
>> 
>
> This is what I'm going to be working next: https://dev.laptop.org/ticket/7430
>
> So will be easier for everybody to just use the view they like most.
>
> Regards,
>
> Tomeu
>
>
>   
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Home View appearance

[NoiseEHC]

>> Another problem is that the Activity button (the 4th zoom level) can
>> select the Journal if that was the last active activity. It has an own
>> button so I cannot see any reason why it has to be the way it is. It is
>> an annoyance that when I download several things then go from Browse to
>> the Journal, open the pdf in Read, close Read and click the button it
>> switches back to Journal and not Browse. It should not treat Journal as
>> an application.
>> 
>
> Hmmm, that's not quite how it should work.  While we do technically
> treat the Journal as an activity, it should always be the bottom-most
> activity on the stack.  That is, you should only ever be switched to
> it when there are no other open activities left, which is reasonable,
> since the Journal is a point to resume activities from.  If you can
> confirm this behavior in a recent joyride, could you make a ticket and
> indicate what build number you tested on?
>   
I am not sure if this bug is in joyride. Since my XO cannot connect to 
my WPA router easily usually I just
try some build, see that it still cannot connect, and then use 708 
(which can connect with some mumbo jumbo).
So it is totally possible that only 708 has this behavior.
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Home View appearance

[NoiseEHC]

You did not understand correctly.
Neither 708 nor joyride can connect to my router easily.
The difference is that I have made my 708 to connect with some 
mumbo-jumbo and since tested joyrides do not fix this bug, I simply do 
not want to do the same mumbo-jumbo with them (currently I use my XO 
almost always via the console so there is no reason for me to use joyrides).


Now the symptoms if you are interested:
http://lists.laptop.org/pipermail/devel/2008-June/015270.html
I have created a networks.cfg as a last hope.
It is a Belkin Pre-N F5D8230-4, and no matter what do you think about 
the MIMO chipset or firmware, 3 things are sure:

1. It works without WPA perfectly.
2. When the XO tries to connect with WPA, no matter how many times I 
enter the same pass phrase, it almost never connects (it happened 2-3 
times when it did connect that way but <1%).
3. When the XO gives up connecting automatically and I connect by 
clicking the red dot on the Neighborhood view, it succeeds (I do not 
even have to type a pass phrase since it is networks.cfg).
+4. When I close the lid and open later, the connection has been lost. 
If I click the red dot when the XO does not want to reconnect 
automatically, it succeeds.
+5. When the auto connection happens after boot, the Neighborhood view 
and the Home view normally show different status about the router, 
sometimes the Home view shows that it is connected to channel 6 or 11 
(the router is set to 1), and usually when I click disconnect, it takes 
a long time to finish (and both views show during that time the 
disconnect menu item for the router, but when I click it, nothing seems 
to happen). Rarely, when I click disconnect, the router simply 
disappears for a while.


So I am sure that either the auto connection manager (I am not a Linux 
guru, can it be the Network Manager?) has a big race condition, or the 
wireless scanning logic has some bad heuristic. What is sure is that my 
XO can connect to the MIMO chipset 100% reliably.


Michael Stone wrote:

On Wed, Jul 09, 2008 at 12:55:25PM +0200, NoiseEHC wrote:
  
I am not sure if this bug is in joyride. Since my XO cannot connect to 
my WPA router easily 



Do I correctly understand that

  1. You CAN connect to your WPA router easily from 708
  2. You CANNOT connect to your WPA router easily from joyride?

Thanks,

Michael

P.S. - Do we have good Wiki instructions on how to debug wifi
connectivity issues?


  
[Belkin_Pre-N_324924]
timestamp = 1215087457
we_cipher = 0
key = 
key_mgmt = 2
bssids =  
wpa_ver = 2

___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: LZO support

[NoiseEHC]

Currently I am profiling/rewriting the LZO decompression code.

The speed result (from memory, not too reliable):
LZO compression: ~9.5 MB/sec
LZO compression ASM version: ~9.3 MB/sec
LZO decompression: ~25MB/sec
LZO decompression ASM version: ~40 MB/sec
Here the ASM version is the version from the original LZO library.
The test were made with a 40MB file which was created cat-ing data from 
/lib (as I remember...), it is half .so and half text, compression is to 
50-60%.

See, the ASM "optimized" version is slower than the kernel's one 
(lzo1x_compress.c). I am shooting at ~80-100 MB/sec, it would make it 
worthwhile to include in the kernel. Note that if this succeeds then 
loading will become IO limited instead of CPU limited.

Could somebody make an LZO compressed image? Unfortunately I am not 
qualified for this work.

ps:
This work stalled a little bit since it seems the people who converted 
minilzo to be compilable with the kernel messed up the code.

Erik Garrison wrote:
> On Fri, Jul 11, 2008 at 06:33:41PM +0100, Gary C Martin wrote:
>   
>> One concern I have with auto saving state before powering off is the  
>> potential corruption of journal data. How robust is the Journal if  
>> power off happens half way through an ongoing auto state save – do you  
>> loose both the new journal entry and the original entry you had  
>> resumed from (partially overwritten)?
>>
>> The worst case just now seems to be EToys, it's really great for  
>> tinkering, but creating even simple scripts seems to require a massive  
>> amount of time to save (example 5+min!! to close & save a script with  
>> five polygon space invaders with 2 frames of animation and a few lines  
>> of movement code). Add to that that EToys seems to be almost as power  
>> intensive as video recording (my impression only, I've not power  
>> tested). I find I'm the habit of plugging into mains before I try and  
>> close an EToys instance I've made more than a few modifications to.
>>
>> 
>
> Based on some very provisional testing, I believe that the extremely
> long save times we're seeing in EToys are at least partly related to our
> use of zlib compression for data nodes in the filesystem.  We could
> switch to LZO compression for a 500% speedup in compression (write), a
> 50% speedup in decompression (read), and a slight (but significant)
> decrease in compression ratio.
>
> While working with Record I noticed that we cannot write very high
> resolution video (>320x240 @ 10fps) to the filesystem because of this
> issue.  It would help to know what other applications are suffering
> because of poor fs write speed.
>
> There is an existing ticket to address this issue
> (http://dev.laptop.org/ticket/2072).  This ticket suggests that we could
> add LZO support to OFW in order to implement LZO-compressed data nodes
> in the filesystem.
>
> In this scheme OFW would require LZO support to boot a kernel which
> resides on an LZO-compressed fs.  But an alternative is to implement a
> partitioning scheme in which user data resides on a LZO-compressed
> partition and the system data resides on a zlib-compressed partition.
> This would provide us with good compression for system components
> (saving space) but would also allow users to write data much more
> quickly.  That said, implementing LZO compression across the entire
> system might improve system responsiveness by a noticeable margin, so it
> might be worthwhile.  Testing is required.
>
> Erik
> ___
> Devel mailing list
> Devel@lists.laptop.org
> http://lists.laptop.org/listinfo/devel
>   
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: LZO support

[NoiseEHC]

C. Scott Ananian wrote:
> On Mon, Jul 14, 2008 at 2:05 PM, NoiseEHC <[EMAIL PROTECTED]> wrote:
>   
>> Currently I am profiling/rewriting the LZO decompression code.
>> 
>
> Again, be warned that there's no guarantee we will ever use LZO
> compression.  The plan of record is to disable compression on selected
> files instead.  That's not to say you shouldn't do the work if it
> interests you, just that you'll be facing an uphill battle & don't say
> we didn't warn you. =)
>
>   
I have a partially finished ZLIB decompression code as well. mstone just 
told me that time that we will use LZO so that effort was moot... 
Currently I cannot decide if I should cry or just laugh hysterically... :)
>> worthwhile to include in the kernel. Note that if this succeeds then
>> loading will become IO limited instead of CPU limited.
>> 
>
> Are your benchmarks catting data from a jffs2 partition on FLASH or an
> ext2/3 partition on a hard drive?  I'd be interested to know exactly
> how far we are from being IO limited at present.  Are we quibbling
> about details in the noise?
>   --scott
>   
These benchmarks just measure the noncached memory -> memory 
compression. As I have said I am not too qualified for this work, so 
unless somebody creates an XO image which uses LZO... I mean that if 
there is a working F9 image with the new kernel then I will be able to 
work from the srpm but simply cannot build a full image myself.
> ps. even with the above warning, there are many scenarios we might
> find ourselves using LZO on jffs2, at least for a short while, so good
> luck!
>
>   
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: LZO support

[NoiseEHC]

It seems that you two did not notice the smiley at the end of that 
sentence... I rather not to think about that it was not funny... :)

First, I have to apologize since that statement was made by Mitch 
Bradley here:
http://lists.laptop.org/pipermail/devel/2008-January/009437.html
I am sorry, seems that I am just too old to remember names anymore.
For excuse I could say that both names starts with M so it was close... :)

Now about my obsession of compression code tuning:
Lately I have more or less finished documenting the Geode processor and 
need some problem where I can test what I have learned. Since 
compression is small, has potential and relatively isolated, I have 
chosen that. It could have been anything else but it seems that this can 
have the most effect.

C. Scott Ananian wrote:
> On Mon, Jul 14, 2008 at 6:01 PM, Michael Stone <[EMAIL PROTECTED]> wrote:
>   
>> On Mon, Jul 14, 2008 at 11:51:20PM +0200, Tomeu Vizoso wrote:
>> 
>>> On Mon, Jul 14, 2008 at 9:00 PM, NoiseEHC <[EMAIL PROTECTED]>
>>> wrote:
>>>   
>>>> I have a partially finished ZLIB decompression code as well. mstone
>>>> just
>>>> told me that time that we will use LZO so that effort was moot...
>>>> Currently I cannot decide if I should cry or just laugh
>>>> hysterically... :)
>>>> 
>> I believe I made statements a few months ago about what compression we
>> presently _do_ use, but I don't think I commented on what compression we
>> ought to be using. Can you remind me of my words?
>> 
>
> And, for the record, I didn't say "we'll be using zlib instead" -- in
> reality, we are trying to move off jffs2 just as quickly as possible,
> and the compression scheme we use "next" will be heavily influenced by
> whatever the "other filesystem" has adopted.  But it is a fact that we
> are currently using zlib and rtime compression in jffs2, and OFW
> doesn't currently support lzo compression, so that adoption of LZO in
> jffs2 would have to wait until partition support -- which is also
> roughly the time when I hope to replace the non-boot partitions with
> something better than jffs2.  So the window for LZO-related jffs2
> improvements is very narrow.  If logfs/ubifs/yaffs can benefit from
> your LZO work, then you've multiplied its potential usefulness -- and
> a quick google shows discussions of LZO in ubifs.  Besides, improving
> LZO in the kernel improves things for everyone, not just OLPC, so if
> it's your itch, go scratch it!
>   --scott
>
>   
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: LZO support

[NoiseEHC]

Okay, I have become a little more advanced lately, and was able to 
compile the kernel zlib and lzo in userspace (last year I gave up...).

So here is the code (with results, the lib file is the compiled lzo 2.03):
http://wiki.laptop.org/images/5/5b/Zlibtest5.tar.gz

Since I did not find any assembly compressor support (only 
decompressor), I do not know what did I talked about last time. It could 
have been that I have been dreaming about that...

The program needs a test.dat file, I have used "cat */*>test.dat" in 
"/var/lib" which gave me a 22 MB file (the code handles a max of 40 MB). 
The results are consistent with Erik's results. Since ZLIB has a 
compression level parameter, and LZO has different compressor types, I 
will test those as well.

Now I think that jffs2 (or any other flash filesystem) should use LZO to 
store the data, so if the limit is the flash write speed then it could 
write twice as fast. Later, when there is ample power in the battery and 
no cpu utilization, it should recompress the data either using a better 
LZO or by ZLIB (of course only those files which are old enough, so it 
would not recompress files which will be modified soon).

And the results:
started with block size == 4096

LZO test
in size: 22987994
22987994 -> 11575529 (50.354672%)
1.96 seconds -> 11.185234 MB/sec
11575529 -> 22987994
0.75 seconds -> 29.230745 MB/sec
compared 22987994 bytes OK

LZO asm test
in size: 22987994
22987994 -> 11575529 (50.354672%)
1.93 seconds -> 11.359098 MB/sec
11575529 -> 22987994
0.49 seconds -> 44.740936 MB/sec
compared 22987994 bytes OK

ZLIB test
in size: 22987994
22987994 -> 8982973 (39.076802%)
9.81 seconds -> 2.234767 MB/sec
8982973 -> 22987994
2.75 seconds -> 7.972022 MB/sec
compared 22987994 bytes OK


Greg Smith wrote:
> Hi Erik,
>
> Can you design a test case or two to test the performance of these 
> compression schemes?
>
> Thanks,
>
> Greg S
>
> ___
> Devel mailing list
> Devel@lists.laptop.org
> http://lists.laptop.org/listinfo/devel
>
>
>   
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: State of 8.2.0, July 21, 2008

[NoiseEHC]

> My changes from the last 2 weeks are now in joyride, leaving us 2mb over
> target.
> I slimmed down 2 more packages today (pam and xorg-x11-utils) which
> should bring us to the target.
>
> We also need to get these package forks into Fedora. I filed a load of
> RH bugzilla bugs about them today. Hopefully Dennis can find time to
> give them a quick glance, then I can find another Fedora dev to do the
> merging and committing.
>
>   
remove mesa-libGL* and mesa-dri* stuff and similar.
mesa does not work since GLX is not compiled into X, at least on 2181.
(it does work if I compile it independently and gives me <3 FPS)
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: odd yum dependency issues in joyride

[NoiseEHC]

[EMAIL PROTECTED] wrote:
> c. scott ananian wrote:
>  > On Fri, Jul 25, 2008 at 10:43 AM,  <[EMAIL PROTECTED]> wrote:
>  > > i've been slowly moving my g1g1 machine forward to joyride (from 656).
>  > > i had customized it quite a bit, and am trying to restore those changes.
>  > > i've got xfce running, but a couple of yum dependencies (which were fine
>  > > with F7) are failing.
>  > >
>  > > to wit:
>  > >
>  > > # yum install gcc
>  > [...]
>  > >  --> Missing Dependency: glibc-common = 2.8-3 is needed by package 
>  > glibc-2.8-3.i686 (installed)
>  > > Error: Missing Dependency: glibc-common = 2.8-3 is needed by package 
>  > glibc-2.8-3.i686 (installed)
>  > 
>  > Seems strongly related to trac #5056.
>
> related, perhaps, but clearly new breakage, since a gcc install worked
> fine in 656.
>
>  > 
>  > > # yum install firefox
>  > [...]
>  > > Error: Missing Dependency: gecko-libs = 1.9.0.1 is needed by package 
>  > firefox-3.0.1-1.fc9.i386 (olpc_development)
>  > 
>  > Hmm, this might be a more fundamental mismatch, since we have a
>  > specific version of gecko used for Browse.  You could always download
>
> but didn't we always?  (i.e., what's changed?)
>
>
>   
With 2181 this worked:
yum install mc
yum install make gcc

___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: odd yum dependency issues in joyride

[NoiseEHC]

I have just reflashed my XO since it failed to boot.
Did a reflash with 703
then olpc-update --usb to 2181
then yum install gcc

[EMAIL PROTECTED] wrote:
> martin wrote:
>  > On Fri, Jul 25, 2008 at 06:33:08PM +0200, NoiseEHC wrote:
>  > > With 2181 this worked:
>  > > yum install mc
>  > > yum install make gcc
>  > 
>  > I've been yum install'ing gcc after every olpc-update to joyride since
>  > about 1559, and have never had any problems.
>  > 
>
> interesting.  i wonder why mine fails?  did either of you start
> with an older (e.g. 656/703) build that included a yum'd gcc?
>
> paul
> =-
>  paul fox, [EMAIL PROTECTED]
> ___
> Devel mailing list
> Devel@lists.laptop.org
> http://lists.laptop.org/listinfo/devel
>
>
>   
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: video bleeds through somewhat between sessions

[NoiseEHC]

It is the video chip's feature that it can display a video overlay over 
the RGB bitmap. The pixels where the overlay can be seen is defined by a 
colorkey (what was 0xFF00FF in the example), or the alpha component of 
the display RGB bitmap (not used on the XO since the change 16 bit 
bitmaps). What you are seeing that the X server does not disable the 
video overlay while switching programs. It can be an error or just some 
braindamaged X stuff. Either way, it has nothing to do with bitmap 
operations.

Mikus Grinbergs wrote:
>>> G1G1, Joyride 2241.  In one Terminal session started mplayer -- it
>>> was playing a movie.  Went to another Terminal session, and entered
>>> some commands.  Noticed that not all of the text on that screen was
>>> equally distinct - some of it was paler than others.  Noticed that
>>> *which* text was paler changed from second to second.  Realized that
>>> the paler text in the second Terminal screen corresponded to the
>>> *brightest* areas of the movie frame then being shown in the first
>>> Terminal screen (the one I had switched way from).
>>>   
>> Video is muxed to the visible screen through the use of a color key -
>> given a rectangle of some size, the hardware compares all of the pixels
>> in that rectangle against a set color - if they match, then a pixel of
>> the video frame is shown, otherwise not.
>>
>> The color is specified by the video application - most applications use
>> very saturated colors similar to those used in "green" or "blue" screens.
>> My favorite is hot pink (0xFF00FF).  IIRC, mplayer uses an off-shade color
>> of grey, so it is easier to run into the possibility that other applications
>> will match the color key, especially with automatic shading such as
>> anti-aliasing.
>> 
>
> I did NOT understand at all what you are trying to say.  [I used the
> words in the subject line to try to describe that what *would* have
> been shown by one program (whose output "window" I was NOT looking
> at) "bled through" to affect what *was* being shown from ANOTHER
> program (whose output "window" I WAS looking at).]
>
>
> In my mind, a 'session' can request that certain pixels be displayed
> on the screen.  [If a text-output program is running, it will
> request pixels which make up an image of a text character.  If a
> video-output program is running, it might request pixels which make
> up an image of a white cloud.]  My point is that each program (that
> is, 'session') supplies __its own__ set of rendering requests.  I
> would expect that if I am looking at an area of the display which I
> thought was dedicated to output from program A, then I will  ONLY
> see pixels as requested by program A (no muxing!).
>
> What was happening to me was:  while looking at the screen showing
> output from program A (Terminal 'session'), the pixels themselves
> (i.e., for black text characters) appeared to have been requested by
> program A.  But the intensity of those pixels (how black they were)
> appeared to be MODIFIED by whatever intensity program B (mplayer
> 'session') wanted for that spot on the screen.  Since it is the
> driver interface (or something) software that ACCEPTS the
> pixel-rendering requests issued by the running programs, I would
> expect that when I switch away from session B, the rendering
> requests still bing issued by session B would be IGNORED in their
> entirety (until B again has the "focus").
>
>
> Why should the video have to be 'muxed' from the simultaneous output
> of multiple sessions ?  Why, when session A has the "focus", should
> anything done by session B affect  HOW  A's output gets shown ?
>
> mikus
>
>
> ___
> Devel mailing list
> Devel@lists.laptop.org
> http://lists.laptop.org/listinfo/devel
>
>
>   
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Test results on Joyride 2230?

[NoiseEHC]

I have downloaded it even before it was selected as a weekly stuff.
Record-54 crashes the build (the machine freezes). I did not create a 
TRAC entry since my last TRAC entry did not receive any attention so I 
just did not care (actually I do not know what to expect). Just started 
the thing and somehow it showed my video and later the machine freezes 
(mouse stops, keyboard cannot switch to console). A week earlier it 
could even show my video. Otherwise it seems to work (I tested it very 
sparingly).

Greg Smith wrote:
> Hi All,
>
> Can we get a "show of hands" on who downloaded joyride 2230?
>
> If you didn't try it yet, instructions for downloading it are available 
> at: http://wiki.laptop.org/go/Friends_in_testing
>
> If you did install it, can you send a quick reply. Any comments on what 
> you tried and your impression of its stability is also welcome.
>
> Thanks,
>
> Greg S
>
> ___
> Devel mailing list
> Devel@lists.laptop.org
> http://lists.laptop.org/listinfo/devel
>
>
>   
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Test results on Joyride 2230?

[NoiseEHC]

Forgot to mention: the software updater draws the progress bar, and
after several megabytes of data, it jumps back to the beginning without
any error message. It cannot update anything.
Also, the TRAC which was created by me was about that Speak/Physics
crashes the machine. Updating software is not an option since in those
cases an user level program can crash Linux which is unacceptable.
http://dev.laptop.org/ticket/7608

Morgan Collett wrote:
> On Tue, Aug 5, 2008 at 14:03, NoiseEHC <[EMAIL PROTECTED]> wrote:
>   
>> I have downloaded it even before it was selected as a weekly stuff.
>> Record-54 crashes the build (the machine freezes). I did not create a
>> TRAC entry since my last TRAC entry did not receive any attention so I
>> just did not care (actually I do not know what to expect). Just started
>> the thing and somehow it showed my video and later the machine freezes
>> (mouse stops, keyboard cannot switch to console). A week earlier it
>> could even show my video. Otherwise it seems to work (I tested it very
>> sparingly).
>> 
>
> Record-56 is the latest. The software updater in the control panel
> should now pick that up.
>
> Regards
> Morgan
>
>
>   


___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Test results on Joyride 2230?

[NoiseEHC]

Okay, it is a question I have been wanting to ask for a very long time. 
Is it better to spam the TRAC with crashes or similar malfunctions about 
alpha quality releases (like latest joyrides), or just be silent since 
it happens occasionally and you do not have the time and manpower to 
handle all those entries? I am asking this because there is not mention 
about this in the wiki and I would like to hear something "official" 
about this issue. For example normally I would not spam TRAC with 
usability issues or crash report about the F9 rebase, but I would file 
reports if you would make it the preferred method.

For example:
http://dev.laptop.org/ticket/7824
I know this since about 2200 but did not want to record since there are 
much much bigger issues about the XO. Would have been better to just 
file a ticket and forget that?

BTW: I will test record on a newer build before I will file this ticket, 
it may have been magically fixed :)

Daniel Drake wrote:
> On Tue, 2008-08-05 at 14:03 +0200, NoiseEHC wrote:
>   
>> I have downloaded it even before it was selected as a weekly stuff.
>> Record-54 crashes the build (the machine freezes). I did not create a 
>> TRAC entry since my last TRAC entry did not receive any attention so I 
>> just did not care (actually I do not know what to expect). 
>> 
>
> File a ticket, please. Pretty please. Even if we don't reply for a
> while, it's doing us a favour, and should only take a couple of minutes
> of your time.
>
> Thanks,
> Daniel
>
>
>
>
>   
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Please help test our new 8.2.0 weekly beta, joyride-2263!

[NoiseEHC]

Probably it would help if the linked pages would contain a link to the 
specified build's directory at
http://xs-dev.laptop.org/~cscott/xo-1/streams/joyride/build2263/devel_jffs2/
Just my 0.02$

Michael Stone wrote:
> We are thrilled to announce a new test build, joyride-2230, valid until
> Wednesday, August 13.
>
> Please help test it according to the detailed instructions at
>
>http://wiki.laptop.org/go/Friends_in_testing
>
> while we still have time to fix issues you might find!
>
> Our specific interest this week continues to be activity compatibility: 
>
>   "Does your favorite activity still run on joyride-2263?"
>
> Currently known issues are recorded at: 
>
>http://wiki.laptop.org/go/Test_Group_Release_Notes#Build_Joyride_2263
>
> New issues should be filed in our bug-tracking system (dev.laptop.org)
> according to 
>
>http://wiki.laptop.org/go/Submitting_bugs
>
> or by notifying us by other means.
>
> Thanks!
>
> Michael
> ___
> Devel mailing list
> Devel@lists.laptop.org
> http://lists.laptop.org/listinfo/devel
>
>
>   
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Please help test our new 8.2.0 weekly beta, joyride-2263!

[NoiseEHC]

2263 works for me.
Something fixed the system crash made by Record I have experienced on 
2230 (so I will not file a ticket for that).
Speak cannot be started (the icon blinks for a long time and then the 
application stops starting).

What started with 2263 is that when the system wakes the display acts 
strangely. What I see is similar when there is a noisy (analog) TV 
broadcast and horizontal distortions run trough the display in a 
vertical direction.
If I am in Browse and switch to the Neighborhood view then sometimes the 
XO starts drawing the view and for a frame I can see the Browse then it 
jumps back to drawing. It can be the same thing as the distortion thing 
or dunno. Maybe a DCON bug or X uses wrong panel settings?

My another $0.02 is that it would be a good idea to zip the .toc and 
.usb files together in every build since
1. Can be compressed to 40% and as such faster to download
2. The .zip file could act as a checksum (it is not too easy to checksum 
downloads on a windows machine...)

Michael Stone wrote:
> We are thrilled to announce a new test build, joyride-2230, valid until
> Wednesday, August 13.
>
> Please help test it according to the detailed instructions at
>
>http://wiki.laptop.org/go/Friends_in_testing
>
> while we still have time to fix issues you might find!
>
> Our specific interest this week continues to be activity compatibility: 
>
>   "Does your favorite activity still run on joyride-2263?"
>
> Currently known issues are recorded at: 
>
>http://wiki.laptop.org/go/Test_Group_Release_Notes#Build_Joyride_2263
>
> New issues should be filed in our bug-tracking system (dev.laptop.org)
> according to 
>
>http://wiki.laptop.org/go/Submitting_bugs
>
> or by notifying us by other means.
>
> Thanks!
>
> Michael
> ___
> Devel mailing list
> Devel@lists.laptop.org
> http://lists.laptop.org/listinfo/devel
>
>
>   
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Datastore profiling

[NoiseEHC]

The best would be writing with LZO and later recompress with ZLIB. Also 
the clear install image (which is created in the build) should use the 
slowest ZLIB compression, but I am not sure if this latest idea worths 
implementing.

Erik, you asked how long would it take to finish my decompression 
routines. I did not answer since I was not sure but now it is 100% that 
it will no go into 8.2.0 (currently I am working 12+ hours and could not 
switch on my XO for a week...)

Erik Garrison wrote:
> On Wed, Aug 13, 2008 at 06:06:03PM +0200, Tomeu Vizoso wrote:
>   
>> On Wed, Aug 13, 2008 at 5:51 PM, Erik Garrison <[EMAIL PROTECTED]> wrote:
>> 
>>> On Wed, Aug 13, 2008 at 11:41:38AM +0200, Tomeu Vizoso wrote:
>>>   
>>>> In most cases, the files that activities put into the DS are moved
>>>> instead of copied, because of the long time that takes writing to
>>>> jffs2 due to compression. That will change the profiling results a
>>>> lot.
>>>> 
>>> I'm working on the jffs2 compression issue.  Compression algorithm
>>> support could greatly improve system i/o performance.
>>>   
>> That looks interesting, but from my POV, would be better for Sugar
>> performance if compression was disabled by default, and was mostly
>> used just by pilgrim when creating the jffs2 image file.
>> 
>
> In July, NoiseEHC reported results from his testing of lzo and zlib
> compression.  To best estimate the results we should expect from
> implementing filesystem-wide compression, he compiled the kernel zlib
> and lzo in userspace to test their performance.  Please refer to
> http://lists.laptop.org/pipermail/devel/2008-July/016956.html
>
> I do not understand why we wouldn't want to accept a >500% increase in
> write performance and a >300% read performance boost at the cost of a
> ~25% increase in the size of compressed data.  Please clarify your POV.
>
> Erik
> ___
> Devel mailing list
> Devel@lists.laptop.org
> http://lists.laptop.org/listinfo/devel
>
>
>   
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Please help test our new 8.2.0 weekly beta, joyride-2301!

[NoiseEHC]

It is back again, it is a blocker, please somebody really fix it!
http://dev.laptop.org/ticket/7608

I feel the touchpad very jumpy, it was not jumpy at all with 708.

Since 2230 (if I remember) the software updater simply refused to update 
Scratch (it jumped back to the start of updating).
With 2301 it now restarts Sugar if it is selected in the update, nice 
progress... :)

Please, please please, add 1 line to the build script to zip the .usb 
and .toc together (fast compression is enough).

Could somebody tell me whether 2263 was announced on this list? (I 
somehow missed that email.)

Michael Stone wrote:
> We are thrilled to announce a new test build, joyride-2301, valid until
> Wednesday, August 20.
>
> Please help test it according to the detailed instructions at
>
> http://wiki.laptop.org/go/Friends_in_testing
>
> while we still have time to fix issues you might find!
>
> Our specific interest this week has changed; this week, we'd love to
> know:
>
>"How many ways can you crash or hang your XO?"
>
> Currently known issues are recorded at: 
>
> http://wiki.laptop.org/go/Test_Group_Release_Notes#Build_joyride-2301
>
> New issues should be filed in our bug-tracking system (dev.laptop.org)
> according to 
>
> http://wiki.laptop.org/go/Submitting_bugs
>
> or by notifying us by other means.
>
> Thanks!
>
> Michael
>
> P.S. - I was asked to include to specific notices in this email; first,
> that 
>
>http://wiki.laptop.org/go/Software_update
>
> gives full details on the new Sugar Control Panel Activity Updater and
> second, that we are hoping in future weeks to offer a new "Coordinated
> Testing" volunteer opportunity as part of the run-up to the 8.2.0 raw OS
> release. Poke me, Kim, Joe, Charlie, S Page, Seth, or Francesca for
> details. Sneak preview:
>
>http://wiki.laptop.org/go/Test_cases_8.2.0
> ___
> Devel mailing list
> Devel@lists.laptop.org
> http://lists.laptop.org/listinfo/devel
>
>
>   
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Wellington testers + Activities vs 8.2-759

[NoiseEHC]

>  - Associating to an AP using WPA is extremely unreliable. Some
> laptops do, some after a few tries, some never. Likely to be the
> timing bug that has been discussed before.
>
>
>   
There is another bug somewhere in the network stack. With my Belkin PreN 
router (MIMO chpset), every connection attempt fails which is initiated 
automatically (asks for a password which almost never works). When I 
wait until the XO gives up connecting and then click on the AP, it 
succeeds 100% (gets the password from networks.cfg). It looks like a big 
fat race condition to me, but unfortunately I do not even know where to 
start looking at the source so I will not fix it that is sure.
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: build 8.2-760, Read activity and ebook mode - how to make them work together?

[NoiseEHC]

Probably it would be the right time for 9.1 to finally make Sugar usable 
from the keyboard?
For example I did not see any activities using accelerators (Ctrl+Key or 
Alt+Key) or at least they are not showing it on the UI.
This problem that cursor keys are lost is widespread. In some activities 
I can press the down button several times until the focus moves from the 
toolbar to the scrollable area but on other activities it stays on the 
toolbar, so I have to open the XO from ebook mode, click and close back. 
It is just lame.
Here is a "design document":
How about that when the user keeps the Alt or Ctrl keys down, the XO 
shows him big hints about the available shortcut keys? It could show 
those shortcuts for the active toolbar page's buttons as well. That way 
beginner users (and people who were cursed by very bad memory like me) 
could use it as F1. (In fact it is similar how default XP shows 
accelerators only when Alt is pressed.) It is possible that this design 
is too similar to M$ ribbon controls but I think that being too similar 
is not a shame.

Gary C Martin wrote:
> On 18 Sep 2008, at 06:55, [EMAIL PROTECTED] wrote:
>
>   
>> I have not seen any documentation on how to navigate in Read in  
>> ebook mode.
>> I tried the various game pad buttons but have not had much success  
>> in navigating.
>> I would like to be able to scroll from page to page in ebook mode,  
>> back and forth.
>> Have the the buttons been implemented for ebook mode? If so any  
>> hints out there on how to make this work?
>> 
>
> Testing Read-51 on an XO, the directional arrows pad scroll a pdf  
> about in steps (if you are zoomed in enough to have somewhere to  
> scroll to) – these buttons rotate to match the screen rotation (a nice  
> touch). The four other separate game buttons are, circle to page-up,  
> cross to page-down, square to zoom-in, tick to zoom-out.
>
> However... you must first put focus on the document (clicking on the  
> page, or hitting tab a couple of times will do it). Hmmm, maybe this  
> is worth a trac ticket as it's going to catch folks out.
>
> It's worth noting focus behaviour changes have happened in a bunch of  
> places (Journal being the obvious one), where navigation keys (cursor,  
> page up/down) will go to unintended*** toolbar buttons unless you  
> manually change focus to the scrollable area.
>
> ***for example I just accidentally set a pdf I have to share with my  
> neighboorhood while re-checking the game pad buttons. And as we all  
> know, for some bizarre reason you can never un-share an activity...
>
> actually there is UI a cheat/hack. Copy the offending activity to  
> the clipboard using Journal, then use the clipboard to keep it in the  
> Journal again. The new copy will be private, and you can just delete  
> the other.
>
> --Gary
> ___
> Devel mailing list
> Devel@lists.laptop.org
> http://lists.laptop.org/listinfo/devel
>
>
>   
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: build 8.2-760, Read activity and ebook mode - how to make them work together?

[NoiseEHC]

> Another possible option, for general use, which we've discussed in the
> past is a palette dedicated to keyboard mappings, revealing a user
> configurable set of actions.  This would make it possible to see all
> of the shortcuts for an activity in one consistent place, and even
> allow the user to adjust them on the fly to suit their tastes (which
> might be good, if the mnemonics, for instance, are designed for
> another language.)  This, of course, requires an API so that
> activities can expose lists of available "functions", and then make
> use of the new mappings throughout their activity.
>   
I would vote against it. Simply make shortcuts/accelerators/mnemonics 
translatable and require activity authors to think about the usability 
of their applications. This standardized "function" thing seems a little 
bit over designed, and the keyboard mapping setting palette is a 
complication. I do not know what programs are you planning for the XO, 
but I can hardly think about anything which would require this level of 
customization (and as we all know, customization has a price, less 
simplicity).
See, actually I do not have much weight in this conversation since I 
will not be the man who will code this thing. Another problem is that 
the only application where I have ever customized the keys was Visual 
Studio so I could be a little bit biased.

___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: build 8.2-760, Read activity and ebook mode - how to make them work together?

[NoiseEHC]

> In any case, I brought this idea less due to the customization idea,
> and more due to the expressed desire to have a list of available
> keyboard shortcuts.  It sounded like you were proposing a system by
> which they appeared all at once -- as hovering bubbles, perhaps? -- on
> screen contextual to the control they related to.  That sounds
> interesting, but extremely complicated to do. On the other hand, it
> might not be too much to generate a list of those available somehow.
> Did I misunderstand what you were saying?  Thanks for any
> clarification.
>   
Actually I did not think that far ahead... :)
My main problem was that there was no way (at least there was no way I 
have known about) to change tabs on the toolbar. (I will use Windows 
slang here since I am not a Sugar programmer yet.) Since the exit button 
is on the Activity tab which usually has no useful function for me (I 
only used to read stuff on the XO), every time I have to exit, I have to 
click on the left side for the Activity tab, and then click on the right 
side for the exit button (which is a very bad design IMHO). Reading a 
big pdf file is similar since zooming and jumping to a page is on 
different tabs.
For the first problem I think there is an accelerator key (Ctrl+Q?) but 
every time I use it, I forget next time I start the application.
For the second problem it would be good if there would be some shortcut 
for the "menus" (tabs on the toolbar). It would make the toolbar similar 
to the ribbon control from M$ but who cares?
Now if somehow somebody would implement those thing there is the next 
problem which is that I will forget every accelerator/shortcut promptly 
so there should be some way to show them to my face.
> Perhaps the list is revealed upon press of alt or ctrl (of course
> filtered appropriate to the modifiers pressed), instead of in a
> palette.  Of course, then you have the problem that you might want to
> alt-click on something, or alt-drag something, and such a window could
> get in the way.  I'm not sure the best way to go about it, but let's
> keep ideating around it. It sounds promising.  For it to work well,
> we'd have to have a way to detect all available accelerators,
> probably, so the activities don't have to set it all up on their own.
> I'm not sure how that would work either.
>   
My thinking was that every function is accessible from the toolbar, so 
every accelerator should have a button, and every button should have a 
hint (with the accelerator in the text). So all I was thinking that if 
somebody presses the Alt key it should underline the shortcut letters in 
the "menu" text (the text labels on the toolbar tabs), and every time 
somebody presses the Ctrl key it should show all hints for all the 
buttons for the active tab on the toolbar. All the available shortcuts 
for all the controls should probably go to the activity's documentation 
IMHO.
Now for the Alt+Click and Ctrl+Click problem: it should only show hints 
when the Ctrl key is pressed for 2 seconds (or 1?) and the mouse cursor 
have not been moving during this period. Moving the mouse would cancel 
it. If you implement it then what would be very important is that this 
"gesture" should be generated by Sugar so the application could handle 
it as a key press (and should not need to measure time, etc). Also 
Sugar/GTK should supply a simple method which would show all the hints.
Another idea: there should be some key to cycle the tabs on the toolbar 
(Ctrl+Tab and Ctrl+Shift+Tab ?). It would be time to define the control 
focusing behavior while changing tabs as well.
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Please help test our new best 8.2.0 candidate, 8.2-763!

[NoiseEHC]

I love it!
I do not know who did this, but it seems that my WPA connection issues 
are gone. It connected automatically 2 times in a row, thank you!

Now after updating from 760, on the first boot the software update came 
up, checked Activities' versions, and when I clicked that download and 
install now (do not remember the button's name), Sugar and X restarted. 
After that I manually run the software update from the control panel and 
this time it worked.


Just tested Record and have a problem:
1. Run Record.
2. Take a picture (name it, actually I do not know what action saves it, 
I just created an unsaved picture but it is not important).
3. Switch to Journal, wait until both the Record activity and the just 
taken picture appears.
4. Now when I start the picture, it switches to the running Record and 
nothing happens. Is it the intended behaviour?
5. If I close Record and then start the picture, it starts Browse and 
shows the picture. Why does it show it in Browse?
6. Now when I want to quit Browse it gives me that "Keep error" question 
("Stop anyway" at least works now).

Versions:
Record-57
Browse-98
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: simple datastore replacement, take two

[NoiseEHC]

>> The Datastore can then provide two accessor functions:
>> get_by_value(key) and get_by_reference(key).  get_by_value() returns the
>> contents of the file as a bytestring in memory.  get_by_reference()
>> returns the path to the metadata file, or another path linked (soft or
>> hard) to that file.  This provides all the needed functionality for large
>> and small metadata entries.
>> 
>
> That sounds interesting. I guess that with a POSIX-like API
> implemented with FUSE we would get equivalent fucntionality?
>   
What is the plan on Windows compatibility of Sugar? I mean, is there a 
version planned which can run on Windows? I ask this because if there is 
a plan like this then probably you should forget everything FUSE related 
since it will be impossible to implement that on Windows as I barked 
about here (there is a lot of inlined text in the message):
http://lists.laptop.org/pipermail/devel/2008-April/013330.html
http://lists.laptop.org/pipermail/devel/2008-April/013345.html

___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

5 sec boot

[NoiseEHC]

If you somehow missed it, there is possible to boot Linux in 5 seconds 
on an EeePC.
http://uk.youtube.com/watch?v=s7NxCM8ryF8

Here is the paper:
http://www.fenrus.org/plumbers_fastboot.ppt

Could somebody explain me whether these results are applicable to the 
XO, and how far are we from it, please?

___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: 5 sec boot

[NoiseEHC]

> can you do the hash as you copy it? it should be pretty close to free at 
> that point (since the CPU is waiting for memory/flash access it can do the 
> hash calculationwhen it would otherwise be stalled)
>
>   
According to my measurements the GeodeLX can fetch a new cache line (32 
bytes) every 20-25 clocks.
Unless you can do the hash <2 clocks/byte then you will only earn the 
looping time (assuming that the hashed blocks fit into the L1 cache).
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: 5 sec boot

[NoiseEHC]

> On Sun, 5 Oct 2008, NoiseEHC wrote:
>
>>> can you do the hash as you copy it? it should be pretty close to 
>>> free at that point (since the CPU is waiting for memory/flash access 
>>> it can do the hash calculationwhen it would otherwise be stalled)
>>>
>>>
>> According to my measurements the GeodeLX can fetch a new cache line 
>> (32 bytes) every 20-25 clocks.
>> Unless you can do the hash <2 clocks/byte then you will only earn the 
>> looping time (assuming that the hashed blocks fit into the L1 cache).
>
> is that speed reading from ram or from flash?
>
>
DDR RAM, if neither the L1 nor L2 was hit. I did not measure raw NAND 
read speed, I do not even know how to begin with it.
see:
http://wiki.laptop.org/go/Geode_LX#Memory.2C_Cache_and_TLB

___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: 5 sec boot

[NoiseEHC]

When I will finally have some time (currently I am working even on 
weekends) I will finish my half made zlib decompression code.
Where is that Security hash's code?

Mitch Bradley wrote:
>
> Memory to memory copy: 500 MB/s
> Raw NAND FLASH read:20 MB/s
> Security hash:   4 MB/s
>
> So overlapping hash calculation with NAND FLASH read is of limited 
> value, and trying to overlap anything with memory copy is almost 
> certainly counterproductive.
>
> This discussion seem to be degenerating into a brainstorming session 
> about an sub-problem that is pretty well under control (the firmware 
> component of the boot time).  I've been working diligently on that 
> sub-problem for nearly 2 years now, and I think I have an excellent 
> grasp of where the cycles are going and what can be done to improve 
> it.  The only significant opportunity at this point is to reduce the 
> JFFS2 time, which will require either partitioning or abandoning JFFS2 
> for the boot files, or both.  UBI+UBIFS is one workable approach in 
> the context of a Linux-only machine.  There are some others, such as 
> Redboot partitions with a small boot partition and a large system 
> partition, with various FS possibilities for the two partitions.  The 
> quickest path to a deliverable system would be Redboot + JFFS2 boot 
> partition + UBI system partition.
>
> The rest of the "fruit" on the tree is solidly in the OS domain, 
> encompassing kernel startup, userland startup/initscripts, X startup, 
> and Sugar / application startup.  I would encourage each of you to 
> address the areas in which you have special expertise, and then to 
> take action.
>
>
>
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: 5 sec boot

[NoiseEHC]

> To what end?  AFAIK the zlib decompression (both in OFW and in the OS) 
> is not one of the primary problem areas.
>
Changing fs read from CPU bound to IO bound would change a lot of 
things, for example the boot could utilize a little bit of more 
concurrency. Unfortunately we will only see its effect when the code is 
written (or we will see that it does not help). It is not a problem area 
but first there are projects which would gain on this speed (like etoys 
loading AFAIK), and second every speed optimization (like boot time and 
activity launch time) should take into consideration the faster reading 
speed (which is theoretical at this point).
See, I am NOT suggesting that you should waste time on these 
"nonproblems" but I am not a Linux programming expert so I simply cannot 
help you at this time (finishing 8.2.0) so I spend my (currently not 
existing) time on these "nonproblem" projects.
>> Where is that Security hash's code?
> The computation code is at http://dev.laptop.org/git?p=bios-crypto .  
> It is based on LibTomCrypt.
>
> The specific hashes and signatures that we use are detailed in the 
> "Computation" column of 
> http://wiki.laptop.org/go/Firmware_security#Summary
>
> The security computation code has undergone an audit by outside 
> security experts.  Any changes to the core code would require an 
> additional audit.
>
Exactly which algorithm does the firmware use and how much data does it 
hash? Just out of curiosity...

___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: journal is hard (was Re: notes from the field - Mongolia)

[NoiseEHC]

> We can do a little better than that, actually, by making it all one
> prompt.  It can have a name field, already filled out with the best
> darn attempt at a name we can manage, a tag field (and perhaps even a
> list of popular tags as well, to apply to it with a click or a
> drag/drop), and buttons for [Erase] (Or [Don't Keep]?) and [Keep].
>   
A little better solution would be if the words in the name would be 
treated as tags and if the save "dialog" would offer autocomplete for 
those tags. Tagging via the Journal could just set words to "super" tags 
so they would not be shown in the name but would be handled "harder" 
than "soft" tags in searching or in the proposed tag submenu thing. If 
the user would type in the tag via autocompletion then it should be 
treated as an explicit tag.

I am not sure if you can understand it so here is another try from the 
opposite side:

The problem with tagging is that it is painful to select something on 
the XO from a drop down menu (the list of available tags). (Note that 
developing Sugar on a Linux PC is cheating...) The whole notion of 
explicit tagging is also a nuisance and requiring tagging at save time 
is painful. So this proposal just tries to simplify the process from the 
user's perspective (and makes coding the Journal very very hard, but 
since somebody other than me will code it I do not care...). 
Autocompleting, not only tags but "soft" tags too, would result in that 
if the user is doing some project lately then the system would offer him 
the project's name since probably it would be used lately a lot. Also it 
could be used for filesystem paths as well but probably I should see 
that GNOME UI Hackfest video first.

___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: 9.1 Proposal: shutdown menu

[NoiseEHC]

In the old builds I could initiate a shutdown and close the lid and the 
XO just finished the shutdown.
In the new builds when I close the lid the shutdown halts because of 
suspend (I think, can be mistaken). Could it be a little bit more clever 
about when to suspend? Or is it what you are referring with "(this will 
need to interact with lid-close in a sensible manner.)"?

[EMAIL PROTECTED] wrote:
> this feature has been discussed on the list(s) earlier, but i'm not
> sure of its status.  i'd like to make sure it gets on the table.
> (and it just came up in dan's ethiopian report.)
>
> currently it is much easier to "crash" the laptop (by holding the
> power button down) than it is to shut it down cleanly.  while the
> journalled filesystem (currently) can recover from this on the
> next boot, application writes in progress might not.
>
> pushing the power button on the laptop should present a menu or dialog
> allowing shutdown.  as a strawman UI, the pushing the button should
> result in a simple dialog that looks like:
>
> The laptop will suspend in 5 seconds.
>Shutdown | Suspend Now | Cancel
>
> (this will need to interact with lid-close in a sensible manner.)
>
> paul
> =-
>  paul fox, [EMAIL PROTECTED]
> ___
> Devel mailing list
> Devel@lists.laptop.org
> http://lists.laptop.org/listinfo/devel
>
>
>   
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: 9.1 Proposal: Power.

[NoiseEHC]

> as a G1G1 user one annoyance of enabling power savings is the screen 
> dimming while I am reading a page. I thought this was supposed to be 
> improved, but I saw the same thing when I upgraded to 767 a week ago.
>
>
>   
+1

I did not try 767 but I have found it very irritating in 763.
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: 9.1 Proposal: Top five performance problems

[NoiseEHC]

The Geode X drive copyes every bit of data to the command ring buffer by 
using the CPU so that is sure that those "almost no CPU cycles" thing is 
at least a bit stretch... :) According to Jordan Crouse it will not be 
better but he was not too concrete so in the end I am not sure what he 
was really talking about, see:
http://lists.laptop.org/pipermail/devel/2008-May/014797.html

Benjamin M. Schwartz wrote:
> -BEGIN PGP SIGNED MESSAGE-
> Hash: SHA1
>
> Erik Garrison wrote:
>   
>> What about changing the kind of visual feedback we give.  Instead of
>> pulsing icons what about icons with a string of dots beneath, a progress
>> bar, flashing, or another kind of overlay feedback which requires fewer
>> visual changes (frames) and/or could be overlaid on top of existing
>> icons without calculating a new animation for every icon?
>> 
>
> We have GPU-accelerated alpha compositing on the XO, so we could do the
> current animation using almost no CPU cycles.  It's just a question of
> figuring out how to access that compositing.  As far as I'm aware, no
> effort in this direction has been made.  I don't know if "composite" here
> requires the use of "Composite" in the window manager or not; my knowledge
> of X is minimal.
>
> - --Ben
> -BEGIN PGP SIGNATURE-
> Version: GnuPG v2.0.9 (GNU/Linux)
>
> iEYEARECAAYFAkkCPQAACgkQUJT6e6HFtqSlSwCfVrZfVFFUqbwgBuLJCckGmHDc
> S40An2vtXMot6/rz9YmceB38geDaQhH4
> =aOse
> -END PGP SIGNATURE-
> ___
> Devel mailing list
> Devel@lists.laptop.org
> http://lists.laptop.org/listinfo/devel
>
>
>   
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: 9.1 Proposal: Top five performance problems

[NoiseEHC]

Could you be a bit more specific, please? What did you mean when you 
talked about that moving a little bit more of the driver to kernel level 
would not help? (This was the mentioned thread I had with Bernie.)

Also could somebody enlighten me why we does not use DirectFB? Is it 
because of there are not enough developers or because it is not 
technically sound?

Jordan Crouse wrote:
> On 25/10/08 00:00 +0200, NoiseEHC wrote:
>   
>> The Geode X drive copyes every bit of data to the command ring buffer by 
>> using the CPU so that is sure that those "almost no CPU cycles" thing is 
>> at least a bit stretch... :) According to Jordan Crouse it will not be 
>> better but he was not too concrete so in the end I am not sure what he 
>> was really talking about, see:
>> http://lists.laptop.org/pipermail/devel/2008-May/014797.html
>> 
>
> Indeed - many CPU cycles are used during compositing.  There is a lot of
> math that happens to generate the masks and other collateral to render
> the alpha icon on the screen.  The performance savings in the composite
> code comes from not having to read video memory to get the src pixel
> for the alpha operation(s).  That performance savings is already available
> in the X driver today.
>
> Jordan
>
>   
>> Benjamin M. Schwartz wrote:
>> 
>>> -BEGIN PGP SIGNED MESSAGE-
>>> Hash: SHA1
>>>
>>> Erik Garrison wrote:
>>>   
>>>   
>>>> What about changing the kind of visual feedback we give.  Instead of
>>>> pulsing icons what about icons with a string of dots beneath, a progress
>>>> bar, flashing, or another kind of overlay feedback which requires fewer
>>>> visual changes (frames) and/or could be overlaid on top of existing
>>>> icons without calculating a new animation for every icon?
>>>> 
>>>> 
>>> We have GPU-accelerated alpha compositing on the XO, so we could do the
>>> current animation using almost no CPU cycles.  It's just a question of
>>> figuring out how to access that compositing.  As far as I'm aware, no
>>> effort in this direction has been made.  I don't know if "composite" here
>>> requires the use of "Composite" in the window manager or not; my knowledge
>>> of X is minimal.
>>>
>>> - --Ben
>>> -BEGIN PGP SIGNATURE-
>>> Version: GnuPG v2.0.9 (GNU/Linux)
>>>
>>> iEYEARECAAYFAkkCPQAACgkQUJT6e6HFtqSlSwCfVrZfVFFUqbwgBuLJCckGmHDc
>>> S40An2vtXMot6/rz9YmceB38geDaQhH4
>>> =aOse
>>> -END PGP SIGNATURE-
>>> ___
>>> Devel mailing list
>>> Devel@lists.laptop.org
>>> http://lists.laptop.org/listinfo/devel
>>>
>>>
>>>   
>>>   
>> ___
>> Devel mailing list
>> Devel@lists.laptop.org
>> http://lists.laptop.org/listinfo/devel
>>
>> 
>
>   
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

testing 8.2.1 - 800

[NoiseEHC]

I had some time lately and tested the latest signed release.

I have reimaged (with 4 button pressed start) my XO and after that 
installed activities via olpc-update and some yum  
mc/gcc/make/X11-devel/Xv-devel. I did not copied my developer key to 
/security this time. The extreme power management is unchecked (the 
default setting).

1. While doing some "yum list something" on the virtual console my XO 
freezed and only could restarted by holding the power button for 5 secs. 
Nothing was written to the console (but it was the 2. console). The file 
what yum generated stopped abruptly halfway.
2. The wireless works as it did with 763 or something (I have never 
installed 767). It cannot connect to a Belkin PreN with WPA (100% 
displays the password dialog and no matter what I type into it the 
machine just redisplays the dialog). After I cancel the dialog then I 
can just click on the AP and it connects 100%. I do not really remember 
how did it work last year but it seems that I no longer have to wait 
until the XO stops scanning mesh networks (and it does not seem to stop 
it anyway).
3. After using the XO for 2 hours while developing some XVideo using 
applcation and closing the lid several times and stealing some 
unprotected wifi to browse the web (I was waiting in a hospital for 2 
hours), I finally shutdowned the machine and it freezed with "libertas: 
tx watch dog timeout". After I pressed the power twice to sleep and 
wakeup, the shutdown continued, displayed the shutdown screen and the 
the XO switched off.
The error message was uploaded here: 
http://wiki.laptop.org/go/Image:Xo_freeze_on_shutdown.png

So it is a regression since the old image was not used to crash or 
freeze during shutdown and the wireless seems to be just as broken as it 
was (probably the suspected network manager race condition what I was 
speculated lately).
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: OS/X11 support for XO-1 hardware?

[NoiseEHC]

If you need it for some game then here is how to do it: attached.

A little question to Jordan Crouse or anybody else who can answer.
Here Jodran told me that the Geode can do XV flipping:
http://lists.laptop.org/pipermail/devel/2007-May/005208.html
It can be that he either did not reflect to that one part of my message 
or I am just too stupid to make it work. So what I do not see in the XV 
documentation is how to allocate two xvideo frames and flip between 
them? What this code currently does is allocating one frame and doing 
XvShmPutImage. Because its speed depends on the size of the xvideo frame 
I think it does blittting (copying) and not flipping (switching).

So how to do flipping?

ps:
Actually I would need 3 frames for triple buffering but I would be happy 
even with 2 frames.



Chris Marshall wrote:

With the spin-off of Sugar development to sugarlabs,
it is nice to see the development continued.

However, it seems that the OLPC layoffs and refocus
has scuttled the work to complete some OS and system
software support for the XO-1 hardware features.

For example, I have been waiting for the video scaler
support to allow for adjustable display resolutions on
the XO.  Among other things, it would allow programs
that don't understand a 1200x900 but only 6x4" display
to work at a more usable resolution where the graphic
elements and text/fonts are consistent and visible to
the naked eye...  It would allow for much improved
video performance since you could play back a 320x240
video on the full screen at considerable CPU savings.

I had thought this capability would be coming with
the Fedora 10 move in 9.1.0.  With that release now
scuttled, I'm wondering more generally, are these
pieces being picked up anywhere?

Would it make sense to have an 8.2.2 release involving
the move to Fedora 10 but pretty much the same as
8.2.1 otherwise?

--Chris
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel


  




xvscale2.tar
Description: Binary data
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

rotate button sucks on the XO

[NoiseEHC]

Hello!

Just today I have noticed some things about the rotate button (which is 
below the directional buttons on the display part):
1. When the screen is rotated the mouse does not so if I turn the XO to 
be able to read letters, I cannot navigate with the mouse.
2. An Xvideo RGB overlay displays the big nothing (black) while the 
screen is rotated.

If somebody will fix it to be usable then it would be a good idea to 
program the rotate button so that holding pressed for 2 seconds would 
turn on-off the backlight (color to mono and back).
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: rotate button sucks on the XO

[NoiseEHC]

p...@laptop.org wrote:
> while i can understand the frustration when something that seems
> simple and obvious doesn't work, starting wout with " sucks"
> probably isn't the best way to get people to listen to your issues.
>   
 From my experience, it is the most straightforward way to get some 
reply to my mostly arcane questions... :) I am sorry if I offended people.
> how do other people feel about this problem?  are there any good
> reasons to _not_ make the touchpad rotate with the screen?  (i
> actually think this might be an almost, but not quite, trivial
> addition to the grab key daemon i mentioned to the list last
> week.  matching the touchpad "orientation" to the orientation
> of the screen initially would be the tricky part -- if they were
> out of sync, it'd be a real drag.)
>   
Unfortunately I do not have enough linux programming experience to do 
this otherwise I would have been already done that.
> noiseehc wrote:
>  > Hello!
>  > 
>  > Just today I have noticed some things about the rotate button (which is 
>  > below the directional buttons on the display part):
>  > 1. When the screen is rotated the mouse does not so if I turn the XO to 
>  > be able to read letters, I cannot navigate with the mouse.
>  > 2. An Xvideo RGB overlay displays the big nothing (black) while the 
>  > screen is rotated.
>  > 
>  > If somebody will fix it to be usable then it would be a good idea to 
>  > program the rotate button so that holding pressed for 2 seconds would 
>  > turn on-off the backlight (color to mono and back).
>
> is this simply to make the backlight controllable from ebook mode?
> because shift-increase and shift-decrease (or is it ctrl?) accomplish
> this pretty simply now.
>   
Yes, it would be just for the ebook mode. I think it is a more sane 
proposition than for example overloading the power button with extra 
functionality.
> paul
> =-
>  paul fox, p...@laptop.org
> ___
> Devel mailing list
> Devel@lists.laptop.org
> http://lists.laptop.org/listinfo/devel
>
>
>   

___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: rotate button sucks on the XO

[NoiseEHC]

Aaron Konstam wrote:

On Sun, 2009-03-01 at 17:06 +0100, NoiseEHC wrote:
  
Hello! 

Just today I have noticed some things about the rotate button (which is 
below the directional buttons on the display part):
1. When the screen is rotated the mouse does not so if I turn the XO to 
be able to read letters, I cannot navigate with the mouse.


You can navigate but in a sense sideways. Movng the arrow up makes it go
up. But relative to the text it is wrong.
  


Yeah, just if I turn the laptop to read the text then it is a little bit 
impossible to hit anything with the cursor...
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: rotate button sucks on the XO

[NoiseEHC]

Eben Eliason wrote:
> This whole argument, I feel, is fruitless.  That's just my opinion, of course.
>
> The touchpad isn't readily accessible in handheld mode, and was never
> made to be.  I'll continue to suggest that the cursor simply be
> automatically hidden in handheld mode, and that a simple means for
> taking full advantage of the handheld buttons which are present be
> made available to activities in a standardized way.
>   
This argument rests on the wrong assumption that the user can only 
rotate the screen in handheld mode. Of course the user can open the 
laptop as a book and read it rotated while using the touchpad with one 
of his thumbs.
> A suggestion for how this standardized system might work is laid out
> rather clearly at http://wiki.laptop.org/go/Browse#Handheld_Mode.  I'd
> very much like to see an API for the press and press-and-hold states
> of these buttons so that activities could take advantage of it easily
I have read this page but it does not talk about screen rotation at all. 
Unfortunately (last time I checked) most of the activities are handling 
keyboard focus badly and they usually need some help with the touchpad 
to focus to their scrollable area. In handheld mode it means opening the 
screen a little bit as david Lang has just said.

A footnote is that this latter touchpad usage conflicts with the one I 
have talked about halfway on this page, just imagine it :)

ps:
I would like to hear a similarly interesting conversation about the 
xvideo surface and X11 driver, please!


___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: rotate button sucks on the XO

[NoiseEHC]

> in any case, so far i've heard no good argument against rotating
> the touchscreen to match the screen.  it may not be the most
> convenient way to use or hold the laptop, but it would be better than
> the current situation where screen rotation makes the touchpad
> almost completely useless.
>   

The question remains whether we make it rotate to match the closed ebook 
mode or match the rotated opened-like-a-book mode.
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: rotate button sucks on the XO

[NoiseEHC]

Jordan, thanks for the info.
I only started to use the XV overlay because I had this little hope that 
somehow I could get a pointer to a hardware buffer to avoid blitting the 
data but as I see it now the LX driver simply copies the overlay data to 
the hardware buffer so I could just use simple X surfaces as well 
speedwise. Is it true?
Could you please answer those question from my earlier message? (copied 
here):

"A little question to Jordan Crouse or anybody else who can answer.
Here Jodran told me that the Geode can do XV flipping:
http://lists.laptop.org/pipermail/devel/2007-May/005208.html
It can be that he either did not reflect to that one part of my message 
or I am just too stupid to make it work. So what I do not see in the XV 
documentation is how to allocate two xvideo frames and flip between 
them? What this code currently does is allocating one frame and doing 
XvShmPutImage. Because its speed depends on the size of the xvideo frame 
I think it does blittting (copying) and not flipping (switching).
So how to do flipping?
ps:
Actually I would need 3 frames for triple buffering but I would be happy 
even with 2 frames. "




Jordan Crouse wrote:
> Benjamin M. Schwartz wrote:
>> Jordan Crouse wrote:
>>> NoiseEHC wrote:
>>>
>>>> 2. An Xvideo RGB overlay displays the big nothing (black) while the 
>>>> screen is rotated.
>>> Indeed - XV is purposely turned off when the screen is rotated (or 
>>> at least, not displayed):
>>
>> The LX hardware supports rotated blits, right?  So in principle, rotated
>> XV could be added to the driver if someone cared sufficiently...?
>
> Absolutely - and as a special bonus, the LX groks how to rotate YUV 
> data natively, so both YUV and RGB video can be rotated.
>
> Jordan
>
>

___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

What to expect from developers, are there any left? (was Re: rotate button sucks on the XO)

[NoiseEHC]

p...@laptop.org wrote:
>   but like david, i think
> that currently neither olpc nor sugarlabs is going to foster or
> champion their use:  olpc has no resources for s/w development,
> and as far as i can tell, sugarlabs is targeting other h/w
> platforms just as strongly as the XO -- and other platforms don't
> have these screen issues.
>   
Witch the recent disbanding of the development team I simply cannot see 
what will happen to the XO development. I mean that 8.2.1 will be 
released and 9.1.0 is dropped but what I do not understand is what will 
happen with all the development for 9.1.0? What I heard is that those 
will be pushed upstream (whatever that means) but it is not clear if 
reporting bugs or talking about button layouts on the game pad will 
result in a new software release or is just a waste of time. What I mean 
is that should I also subscribe to some Fedora devel list (note that I 
do not know sh*t about linux development, packaging or anything like 
that) to keep informed or what?

Currently I am writing a nice activity which teaches kids what to do 
when alien spaceships attacks Earth and it will take some time to 
finish. What should I do next?

Can some insider comment on these issues please?
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

What to expect from developers, are there any left? (was Re: rotate button sucks on the XO)

[NoiseEHC]

Sorry, I wanted to post it toplevel.

p...@laptop.org wrote:


>   but like david, i think
> that currently neither olpc nor sugarlabs is going to foster or
> champion their use:  olpc has no resources for s/w development,
> and as far as i can tell, sugarlabs is targeting other h/w
> platforms just as strongly as the XO -- and other platforms don't
> have these screen issues.
>   
  
Witch the recent disbanding of the development team I simply cannot see 
what will happen to the XO development. I mean that 8.2.1 will be 
released and 9.1.0 is dropped but what I do not understand is what will 
happen with all the development for 9.1.0? What I heard is that those 
will be pushed upstream (whatever that means) but it is not clear if 
reporting bugs or talking about button layouts on the game pad will 
result in a new software release or is just a waste of time. What I mean 
is that should I also subscribe to some Fedora devel list (note that I 
do not know sh*t about linux development, packaging or anything like 
that) to keep informed or what?


Currently I am writing a nice activity which teaches kids what to do 
when alien spaceships attacks Earth and it will take some time to 
finish. What should I do next?


Can some insider comment on these issues please?

___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: rotate button sucks on the XO

[NoiseEHC]

p...@laptop.org wrote:
> noiseehc wrote:
>  > The question remains whether we make it rotate to match the closed ebook 
>  > mode or match the rotated opened-like-a-book mode.
>
> there's no good answer to this, because there's no way to make it
> "do the right thing" automatically.  the lid switch can't be
> used, because by definition the laptop isn't really in ebook mode
> when you're trying to use the touchpad.  there's no way for the
> laptop to tell that the screen is flipped around, which is what's
> needed.
>
> user configuration might be possible, but frankly, i'd just
> default it to match the "opened" mode.  i've gotten used to (in
> almost-ebook mode) moving my finger opposite to the direction i
> want the pointer to move, and i never rotate the screen in that
> mode precisely because of the rotation issue.  so it would be a
> win just to have the finger-to-pointer relationship be
> predictable, even if it's not "right".
>
>   
+1
The question remains who will code it though :)

___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: What to expect from developers, are there any left? (was Re: rotate button sucks on the XO)

[NoiseEHC]

Daniel Drake wrote:
> It is unlikely that you (as a user, rather than a deployment)
> reporting bugs to OLPC will result in another software release *direct
> from OLPC* (such as 8.2.2), because development of 8.2.x is mostly
> discontinued and will really only be driven by deployments.
> Have you read?
> http://wiki.laptop.org/go/Future_releases
> It may not answer all your questions but it is the most concrete
> documentation that I have seen so far.
>   
I have already read that page and was aware of those issues. 
Unfortunately it does not answer my questions.
> In terms of reporting bugs, the process of "upstreaming" everything
> basically means that OLPC is no longer the distributor and that bugs
> should be reported directly to the people who are "more responsible"
> for the them.
>   
My main problem is that knowing who is more responsible requires knowing 
linux more that I am comfortable with (I am a Windows developer). Here 
are just 3 examples to show my point:
1. Today I noticed that my simple program can crash the whole sugar 
desktop and the X server. Shall I report it to somewhere? I do not even 
know where to look for log files to attach to a bug report. Also if 
nobody will fix it (I cannot fix it that is sure...) then why should I 
care? Does it mean that if no deployment will bark that the desktop&X 
can be crashed then it will not be fixed ever?
2. I can reliably (100%) trigger the "cannot connect to WPA and the 
dialog asks for a password endlessly" bug but unfortunately I do not 
know how to debug that thing. To tell you the truth I do not even know 
where to look for the code of NetworkManager (somebody told that this 
can be the problem) and even if I knew it usually I cannot compile 
downloaded linux code for some arcane reason beyond my understanding. So 
for example in this situation what should I do? Is this NetworkManager 
part of some linux distro, or is it an XO thing? If it is part of 
fedora, who should I report bugs to?
3. Okay, I have forgot the third one... :)
Note that I am totally aware that these things are not your 
responsibility, I would just like to have some answers from somebody. If 
the solution is installing some distro then I will do it, the big 
question is that which one will be the official one?
> What would you do if you ran Ubuntu on your main computer but some of
> the buttons on your keyboard were not working correctly? You would
> file a bug with Ubuntu, who would hopefully either fix the problem on
> their own back, or help you to report the issue to the developers of
> the related package (which would likely be one of the X.org input
> components, in the case of keyboard troubles).
>   
Frankly, if some of my buttons would not work in Ubuntu I would simply 
format the machine and install Windows. :)
> Work with the relevant upstream component. In this case, you are
> working on a sugar activity, so develop it as a platform-neutral
> activity at sugarlabs.org, and work with sugarlabs' standard processes
> of getting activities included in distributions.
>   
This is not an activity in the strictest sense, it is more like a 
library which shows what the XO hardware can do in animation. After that 
probably I will use the lessons learned to optimize GCompris and PyGame 
because currently they look like Powerpoint presentations... So the 
whole point is to work fast on a physical XO hardware. Of course if 
somebody will tell me that the XO is a dead thing and OLPC will cease 
then I will reconsider wasting my time.


___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: What to expect from developers, are there any left? (was Re: rotate button sucks on the XO)

[NoiseEHC]

> I don't know what your simple program does, but it sounds like it
> could be a Sugar bug. You should file a ticket at dev.sugarlabs.org.
> If it is not related to Sugar, we'll try to pass the report along to
> the proper place.
>   
http://dev.sugarlabs.org/ticket/465
> It does sound like NM. Look at ~/.sugar/default/nm
>
> There is still an engineer at OLPC looking into WPA on the latest builds.
>
>   
Reported here, no answer:
http://lists.laptop.org/pipermail/devel/2009-February/023504.html
I just would like to know if testing is worthwhile or just wastes my time.

Thanks anyway!
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: What to expect from developers, are there any left?

[NoiseEHC]

p...@laptop.org wrote:

Cannot comment on the first part, I have no idea how this linux distro 
development thing goes...
> this is a much simpler question:  there's a lot of work going on
> in sugarland to help activity writers.  since activities are released
> independently, the "distribution" aspects that affect XO base s/w
> aren't really an issue.
> http://sugarlabs.org/go/ActivityTeam
>
>   
I have already answered this to Daniel Drake.
>  > 
>  > Can some insider comment on these issues please?
>
> you may be overestimating a) the number of insiders, and b) their
> stash of undisclosed information
 From my point of view every linux/sugar developers are insiders in this 
project especially fedora developers. Do they all have an XO by now or 
OLPC just missed it?
> .
>
> paul
> =-
>  paul fox, p...@laptop.org
> ___
> Devel mailing list
> Devel@lists.laptop.org
> http://lists.laptop.org/listinfo/devel
>
>
>   

___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: XO Gen 1.5

[NoiseEHC]

Hi!
I would like to ask these questions from OLPC staff:

Does this also mean that people who already own XOs will find that new
software is going to require a computer more powerful than they
currently have?  I thought that that was something that was going to be
specifically avoided.



This is going to be a hard trap not to fall into, although several of
the primary activities (Browse and Write) are based on desktop
products that are not necessarily aimed at low-power or embedded
systems, so I don't know if things will actually be any different.
  
What about the integrated 3D support? Are you planning to support 
OpenGL? In that case there will be a very wide performance gap (like 3D 
acceleration vs no 3D acceleration)...


Also it seems that this machine will be on par with current Netbooks. 
Are you planning to sell it through normal distribution channels? (As I 
am interested in theat still no Netbook has the display or the 
indestructible design my XO has.)
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: #9307 NORM Not Tri: 100% reliable way to test a DCON creates the wrong colors bug

[NoiseEHC]

I am not sure if anybody reads the OLPC trac anymore but this demo can 
show us a DCON bug which probably should be fixed in the gen 1.5 
version. Since the program reproduces the bug in 100% of the time, I 
wish you good debugging. (If it turns out to be just an X bug then I do 
not know whether it will be fixed ever so never mind.)


Zarro Boogs per Child wrote:
> #9307: 100% reliable way to test a DCON creates the wrong colors bug
> -+--
>  Reporter:  NoiseEHC | Owner:  jg 
>  
>  Type:  defect   |Status:  new
>  
>  Priority:  normal   | Milestone:  Not Triaged
>  
> Component:  x window system  |   Version:  Mass Production 
> Hardware
>  Keywords:   |   Next_action:  never set  
>  
>  Verified:  0|   Deployment_affected: 
>  
> Blockedby:   |  Blocking: 
>  
> -+--
>  Run the attached program from the Terminal activity on an XO which has 801
>  and has power saving activated and wait until first the screen dims and
>  then until the animation stops. After you wake the machine the colors will
>  be wrong. Repeat 3 times to get back the good colors.
>
>  Now if it a DCON bug then you should probably fix it before gen 1.5 comes
>  out.
>
>   

___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: XO Gen 1.5

[NoiseEHC]

> But this should improve with VIA now having employed Harald Welte of
> gnuviolations.org fame to help them move forward in the open source
> world. They have released their drivers and some manuals for their
> GPUs now. So no 3D just yet, but then that's not exactly a regression
> compared to the geeode video either. Details of Harald's VIA related
> OSS releases can be seen at the link below including links to various
> HW programming manuals.
>   
I do not know. I tried to download the specification to their processor 
and gave up after seeing the massive registration and request forms 
required. It is clearly ridiculous. If somebody has the spec please put 
it onto the wiki, please.
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: [Sugar-devel] [IAEP] 3D engine uses in a no-nonsense GUI (was: XO Gen 1.5)

[NoiseEHC]

I think most of those effects can be just as easily be done by the 2D 
engine (like what the Geode has). Of course it would need a LOT of 
coding, like killing the stupid X driver model with the X server 
process, using a compositing windows manager, rewriting GTK+ to use some 
form of retained rendering mode (like a super optimized Cairo library 
with some scene manager functionality) and finally fixing all of those 
GTK applications which became buggy because of this rendering change. Oh 
yeah, almost forgotten that moving to OpenGL would need just the same 
amount of work.
So not only would it chew trough the XO 1.5's battery like crazy but it 
would not run on the XO 1.0 so does it really worth that effort?


Martin Langhoff wrote:

On Tue, Apr 21, 2009 at 1:42 PM, Martin Dengler
 wrote:
  

Imagine if it actually looked like the demo:
http://www.sugarlabs.org/index.php?template=gallery&page=media_01



Exactly my thoughts. There are a couple of things we have to be
mindful of as we step into the wild 3D world...

 - memory footprint -- those smooth transitions count on having
various full-screen buffers, one for each screen you might want to
slide smoothly into.

 - batery burn -- the OpenGL API was originally designed for high end
workstations, and has some battery-depleting features such as a hi-res
timer event that triggers all the time and prevents sleep

The iPhone uses these smooth UI tricks (to a fantastic effect), and
its battery life is a fraction of every other phone -- I'd say that
the 3D niceness is part of the reason why.

cheers,



m
  


___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: [Sugar-devel] [IAEP] 3D engine uses in a no-nonsense GUI (was: XO Gen 1.5)

[NoiseEHC]

> Actually, GNOME 3.0 is moving into that direction (requiring OpenGL):
> http://lwn.net/Articles/327845/
>
>   
Hehe, seems like that I have just invented Clutter... :)
More seriously, it seems that Sugar just runs ahead of Gnome and 
reinvents almost everything which will be created by Gnome people (or 
will not be created since that article was just a plan). Do you feel 
comfortable that your efforts will not go into Gnome 3.0, or is there 
something I do not know about?
Journal = GNOME Zeitgeist
Karma = Clutter minus the OpenGL acceleration
Sugar = "social desktop"
etc

___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: plz test Build 802 + Firmware Q2E41 = Candidate Release 8.2.1

[NoiseEHC]

As I understand 802 differs from 801 only in the firmware. Is it true? 
Shall I test 802 if I have been using 801 for a long time? Will anybody 
fix kernel/X errors if I report them?

Holt wrote:
> The build is now signed so you don't even need a developer key:
> http://wiki.laptop.org/go/Friends_in_testing
>
> Helping us test WPA (WPA2 especially) would be most useful, since these 
> wifi connections sometimes fail as much as 20% of the time, when Release
> 8.2.0 seemed to fail only ~10% of the time.  And of course try:
> http://wiki.laptop.org/go/1_hour_smoke_test
>
> Please help us clean up Release Notes on the way!
> http://wiki.laptop.org/go/Release_notes/8.2.1
>
> Recap -- all you should need are these 2 files burned onto a USB stick:
> http://download.laptop.org/xo-1/os/candidate/802/jffs2/os802.img (233MB)
> http://download.laptop.org/xo-1/os/candidate/802/jffs2/fs.zip (155K)
>
> Follow the usual procedure (http://wiki.laptop.org/go/USB_update), grab 
> Activities from your XO's Control Panel later, and Buzz (IRC Live Chat)
> if you get stuck: http://forum.laptop.org/chat
>
> Thanks!
> --Holt
>
> ___
> Devel mailing list
> Devel@lists.laptop.org
> http://lists.laptop.org/listinfo/devel
>
>
>   

___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: plz test Build 802 + Firmware Q2E41 = Candidate Release 8.2.1

[NoiseEHC]

> A small amount of testing would be very good, yes.  We don't expect
> any changes to be visible outside of the firmware and battery charging
> (behavior should be better in the presence of batteries with extremely
> low charge), but we should double-check that everything looks normal.
>   
Seems a little strange, I do not remember seeing this before (but it can 
be that I just did not watch other upgrades).
1. Inserted the USB stick, did a 4 button update with the power plugged 
in (and then left the stick inserted into the XO).
2. The machine rebooted without pretty boot (the little man with the 
dots). I could see the linux boot messages with the 1L->X logo on top 
and that is the strange thing. Then it asked for my name (as in first boot).
3. Then I shut down, and started to see if there was pretty boot. It 
restarted immediately after talking about the firmware and then updated 
the firmware. (I have no idea how could it work if the kernel would 
depend on a feature in the new firware.)
4. Then it started with pretty boot and works not (except the wireless 
needs the same canceling and then manual connection as I reported 
hundreds of times).

Now wireless does not work with my PRE-N Belkin router (WPA) as usual. 
When it connects automatically then it throws up the password dialog and 
no matter what I do it will not connect and shows me the password dialog 
repeatedly. What I have to do is to cancel the dialog and click the AP 
icon when it does not blink. Then it accepts (and remembers) the given 
password. What I did is to look at the suspend/resume process by 
pressing the power button and what I noticed that after a suspend the XO 
looses the connection to the AP and it does the exact same thing as 
after a power up. So after every resume I have to cancel the dialog and 
connect manually.

Another thing is that after extended use the shutdown process halts 
halfway and I have to press Alt+2 to continue. It then shows the 
shutdown graphics and then the XO switches off. (It is also a long 
standing bug.)

___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Ambient light sensing via LED response

[NoiseEHC]

> But why do you say you would need 1 mV accuracy ?   Bright sunlight  
> is far stronger than
> the light sources he used.
>   
I am not an engineer so forgive me if I am saying something stupid, but 
is not the goal to switch off the backlight if and only if there is no 
difference between the switched on and switched off state? I mean that 
this implies that you have to measure the mV at a light level when you 
cannot see the difference (what can be a much fainter light than the sun).
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: ffreep support on Geode LX (XO-1)

[NoiseEHC]

I have just tested it on my XO and the Geode DOES NOT support the ffreep 
instruction. It could explain the halting shutdown when it stalls with a 
signal 15 (which happens to be SIGILL) and only continuing it when I 
switch to the other console (as I reported in [1]). So fixing it and 
creating a 803 is absolutely necessary IMHO.


[1] http://lists.laptop.org/pipermail/devel/2009-May/024356.html

Sascha Silbe wrote:

Hi!

While trying to use sugar-jhbuild on DebXO (Debian on XO-1), I 
encountered several programs that crashed with SIGILL, apparently 
during execution of ffreep. While the "AMD Athlon Processor x86 Code 
Optimization Guide" [1] claims that "although insufficiently 
documented in the past, [ffreep] is supported by all 32-bit x86 
processors", the AMD Geode LX datasheet [2] doesn't list ffreep.


As ffreep was added to gcc 3.4 [3] and Build 801 seems to use 4.3.0, 
I'm wondering whether it has been patched/configured in some way to 
avoid this issue or whether the processor actually supports it and 
something else on my machine is broken.



[1] 
http://www.amd.com/us-en/assets/content_type/white_papers_and_tech_docs/22007.pdf 

[2] 
http://www.amd.com/files/connectivitysolutions/geode/geode_lx/33234d_lx_ds.pdf 


[3] http://gcc.gnu.org/ml/gcc-patches/2002-11/msg01386.html

CU Sascha



___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel



__ Information from ESET NOD32 Antivirus, version of virus signature 
database 4052 (20090504) __

The message was checked by ESET NOD32 Antivirus.

http://www.eset.com

  


___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: ffreep support on Geode LX (XO-1)

[NoiseEHC]

> please file a ticket at dev.laptop.org, with details on how to
> reproduce the ffreep issue using build 802.  (if it's only
> reproducible with debxo (unclear from what's been written so
> far), then the priority (and the fix) will likely be very
> different.)
>
>   
I cannot reproduce it reliably so I do not know what should I file as a 
bug. What I have seen with 800 (months ago) can be seen here:
http://wiki.laptop.org/go/Image:Xo_freeze_on_shutdown.png
I think that it is sure that at least some parts of the 80x versions 
contain some code compiled with illegal instructions.

> the failure to switch to the UL-warning screen during shutdown
> is a secondary effect of whatever it is you're seeing, and if
> reproducible should have a second ticket filed.
>
>   
On Windows if there is a debugger installed and a program crashes then 
Windows asks if I wanna attach a debugger. Can something like this be 
done on the XO? Or shall I always run it from a debugger? If so then how 
can I do it? Or can I create a crash dump file somehow? It happens quite 
regularly but I cannot give your instructions. Do not you experience it?
> paul
>
>  > Sascha Silbe wrote:
>  > > Hi!
>  > >
>  > > While trying to use sugar-jhbuild on DebXO (Debian on XO-1), I 
>  > > encountered several programs that crashed with SIGILL, apparently 
>  > > during execution of ffreep. While the "AMD Athlon Processor x86 Code 
>  > > Optimization Guide" [1] claims that "although insufficiently 
>  > > documented in the past, [ffreep] is supported by all 32-bit x86 
>  > > processors", the AMD Geode LX datasheet [2] doesn't list ffreep.
>  > >
>  > > As ffreep was added to gcc 3.4 [3] and Build 801 seems to use 4.3.0, 
>  > > I'm wondering whether it has been patched/configured in some way to 
>  > > avoid this issue or whether the processor actually supports it and 
>  > > something else on my machine is broken.
>  > >
>  > >
>  > > [1] 
>  > > 
>  > 
> http://www.amd.com/us-en/assets/content_type/white_papers_and_tech_docs/22007.pd
>  > f 
>  > >
>  > > [2] 
>  > > 
>  > 
> http://www.amd.com/files/connectivitysolutions/geode/geode_lx/33234d_lx_ds.pdf
>  
>  > >
>  > > [3] http://gcc.gnu.org/ml/gcc-patches/2002-11/msg01386.html
>  > >
>  > > CU Sascha
>  > >
>
> =-
>  paul fox, p...@laptop.org
> ___
> Devel mailing list
> Devel@lists.laptop.org
> http://lists.laptop.org/listinfo/devel
>
>
> __ Information from ESET NOD32 Antivirus, version of virus signature 
> database 4059 (20090507) __
>
> The message was checked by ESET NOD32 Antivirus.
>
> http://www.eset.com
>
>
>
>
>   

___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Fwd: The XO-1.5 software plan.

[NoiseEHC]

> The 1GHz C7 is still a slow cpu, as it seems from reviews of similar 
> netbooks:
>
> http://www.notebookreview.com/default.asp?newsID=4352
>
> For most tasks it is slower than an 600MHz Celeron M and that's not 
> exactly fast. Does anyone more familiar with the hardware have any 
> idea of how fast it is when compared to the Geode?

 From my measurements of the Geode and the very limited "documentation" 
of the C7 I can speculate that the integer unit can have similar speed 
to the Geode clock-by-clock (but can have a better branch predictor and 
faster movsb/movsd implementation) and probably the floating point unit 
is better integrated so floating point code does not block the integer 
unit like on the Geode. So if we do not consider the 3d unit or the 
probably better flash hardware (scatter-gather support) it will have 
exactly the same speed on similar clock speeds but of course it can go 
more than 2x faster.

___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Fwd: The XO-1.5 software plan.

[NoiseEHC]

Please, always use reply-all. Answers inlined where I have an answer.

Tiago Marques wrote:
On Sat, May 16, 2009 at 6:42 PM, NoiseEHC <mailto:noise...@freemail.hu>> wrote:



The 1GHz C7 is still a slow cpu, as it seems from reviews of
similar netbooks:

http://www.notebookreview.com/default.asp?newsID=4352

For most tasks it is slower than an 600MHz Celeron M and
that's not exactly fast. Does anyone more familiar with the
hardware have any idea of how fast it is when compared to the
Geode?


>From my measurements of the Geode and the very limited
"documentation" of the C7 I can speculate that the integer unit
can have similar speed to the Geode clock-by-clock (but can have a
better branch predictor and faster movsb/movsd implementation) and
probably the floating point unit is better integrated so floating
point code does not block the integer unit like on the Geode. So
if we do not consider the 3d unit or the probably better flash
hardware (scatter-gather support) it will have exactly the same
speed on similar clock speeds but of course it can go more than 2x
faster.




And the memory should also help, it should be enough. But still, IMHO 
Xfce would still be a better fit, especially since these laptops go to 
places where things being snappy is almost a requirement.


I do not think that the memory speed is a bottleneck on the XO-1. The 
problem is that the Geode is an in-order processor and an uncached 
memory read block the processor for >25 clocks. It will be the same on 
the C7 unless it has a Core2 Duo category speculative memory prefetcher 
but of course I doubt it... BTW the faster memory will not hurt either.


What about random write performance of the flash memory this time? 
That will be a show stopper if it's below at least 0.5MB/s. But that 
would be hard without cache for the flash.


The 0.5 MB/s came mostly from the zlib compression code. With LZO the 
Geode could compress 10MB/sec so it would have been a big help in write 
performance but the conclusion from most of the developers was that the 
biggest win would be having per inode compression setting (like not 
compressing zip and jpeg files) but of course nothing was implemented.
BTW I have a half-made asm zlib decompressor what I have left rotting 
since it became impossible to debug (hallowed are gcc developers and the 
holy UNIX command piping, gcc generates 1 line of debug info for a whole 
asm block). I have another half-made asm decompressor for LZO but it 
seems that the creator of LZO f***ed up the code and it has unused 
opcodes so I tried to actually document the LZO compressor but my 
efforts stalled since kernel developers were fired from OLPC (I will not 
integrate such code to the kernel that is sure).
The conclusion is that if the XO 1.5 will use a normal filesystem then 
compression will not be supported so flash write speed will not be a 
bottleneck.


As for Gnome/Xfce/KDE, whatever, how are you considering that older 
students guess where F1-12 are? Are any changes planned for the 
keyboard stamping to accommodate this change in direction or are you 
taking that as part of the learning process?



I do not know so reposted to devel.

Best regards,

Tiago Marques


___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Fwd: The XO-1.5 software plan.

[NoiseEHC]

Sorry, should have explained myself better, as I was also talking 
about memory speed and not size, this time.

Ahh, if you wrote about memory size then never mind my comments. :)


Thing is, most flash controller implementations are crap, and it will 
probably be the case with  the one in Gen 1.5. I'm quoting 0.5MB/s in 
*random writes* to the file system, nothing to do with compression. 
Most decent SSDs can write at last 1MB/s with some topping 2MB/s, in 
random patterns, sequential is about 150MB/s+. Sequential is not the 
problem when using SD cards or most USB drives, random writes is, when 
you're trying to have an OS on it.

The best drives around, from Intel, can do 20+MB/s in random writes.

Most SSDs on the market are based on J-Micron controllers that can do, 
at most, 0.04MB/s in random writes. This causes the system to 
frequently stall when some app is performing heavy writes to arbitrary 
locations. Random reads are mostly very fast with every type of flash 
you can get.


http://www.anandtech.com/storage/showdoc.aspx?i=3531&p=25 



0.5MB/s in RR should be enough to avoid most stalls.

I hope that Mich Bradley will educate us but it seems to me that the 
hidden eraseblock handling can be the problem with those devices (and if 
it is true then compression will not help it either). It seems to be 
that some tests are required with physical hardware, a paper processor 
will not be enough... :)


Are there any plans using UBIFS?

___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Bootloader question

[NoiseEHC]

The kernel init improvements will certainly bring 15 other seconds.
Maybe some parallelisation of the sysvinit will save some time, say 5
seconds (low end estimation)
  



Parallelization will not help at all if you are using JFFS2.  The low 
level NAND driver that JFFS2 uses busy waits for I/O, and then JFFS2 is 
CPU-bound on the decompression step, preventing any useful concurrency.


The busy-wait could be changed to an interrupt - if only someone had 
time to do the work and test it extensively.  The decompression is going 
to be CPU bound no matter what you do, so the only option is to arrange 
for the important files not to be compressed (thus increasing the NAND 
footprint).
  


Hi Guylhem!

What I have been told: The busy waiting happens because there is no 
scatter-gather support in the NAND driver so the interrupt rate is high 
and it is faster to busy wait than to context switch. Probably it would 
help to interrupt for large IO and busy wait for small IO but it needs 
testing.
I promise you that if you happen to make the required efforts to speed 
up booting then I will finish my fixed LZO decompressor code. It would 
make reading compressed files actually faster, just I am not a Linux 
kernel developer so integrating that with Linux would be your job.


BTW why the doctors cannot just close the lid and open when needed?

___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

dev.laptop.org expored

[NoiseEHC]

dev.laptop.org uses an invalid security certificate.
The certificate expired on 2009.06.06. 16:03.
(Error code: sec_error_expired_certificate)

___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: A1 Motherboard specs diagram

[NoiseEHC]

> There are no free 3D drivers.  I have heard nothing to indicate that there
> are likely to be soon.  I would be surprised if OLPC were to ship the
> proprietary drivers, though I cannot speak for them.
>   

Then please test the xvideo extension with sleep/resume.
I assumed (wrongly) that there was some way to stretch-blit bitmaps to 
the video memory on X (as the Geode video processor is clearly up to 
this rescaling task). Unfortunately it turned out (or I am mistaken) 
that the only way to scale animations is to use the xvideo extension. 
However it is unusable on the XO-1 because of this bug. If the XO-1.5 
will not have working OpenGL out of the box then probably it would be 
wise to have the only usable X feature for games in a working condition.
http://dev.laptop.org/ticket/9307

ps:
If somebody could prove me wrong in that X is not a clusterfuck and can 
do strectch blits then I would be happy...

___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: A1 Motherboard specs diagram

[NoiseEHC]

> You are mistaken.  The Geode LX has a two scaler units, and neither can
> feed back to the main CPU.  One of them is in the Geode Display Controller
> (not the DCON), and simply scales the entire screen to the output.  The
> other is in the Video Controller, and can be used only for overlay
> scaling.  Either way, the scaled output is never written to video memory.
>  The Graphics Controller, which can manipulate video memory, does not
> contain a scaler.
>   
Yes, you are right, now I remember. Seems that somewhat the names of the 
multiple video units got mixed up in my head.
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

kernel debugging

[NoiseEHC]

Hi!

Recently I have started porting Android-x86 to the XO-1 and 
unfortunately hit a wall. I know that it can be done since somebody 
already booted it but the hard drive crashed in his laptop and he had no 
backups... The porting progresses a little bit slowly because I know 
almost nothing about Linux but at least it is some motivation to learn it.


My problem is that I cannot debug the kernel nor see its debug output 
because the XO-1 does not have a standard serial port.


1.
I have read this page:
http://wiki.laptop.org/go/Serial_adapters
What is not clear to me is whether I should buy the converter which is 
on the picture (I did not get one in my XO's box) or shall I fabricate 
one or shall I request one from OLPC? Will it be good for XO-1.5?

What is another option would be buying one of these:
http://wiki.laptop.org/go/Serial_adapters#Third_Party_Adapters
What is not clear that how will I fabricate a cable which connects those 
adapters to the XO?


2.
I looked into using the USB port with gadget support. I am attaching the 
kernel config if you are interested in. I have compiled amd5536udc.ko, 
g_serial.ko and g_android.ko as modules (so I can look at error messages 
while doing an insmod). (Kernel 2.6.29, g_android.ko would enable 
debugging via USB with the adb - Android debugging bridge.)
If I do not insmod amd5536udc.ko, then both g_android.ko and g_serial.ko 
bark me about missing usb_gadget_register_driver symbol.
If I do insmod amd5536udc.ko then both modules bark about "(No such 
device)" error.
What is not clear whether this amd5536udc.ko is usable at all because I 
am booting the XO-1 from an USB stick so can the AMD southbridge be used 
for both host and device at the same time?

What device does it need?

3.
It happened that one of the hardware in my room had a FTDI usb to serial 
converted. I pluggeg it into the XO-1 and insmod usbserial.ko then 
insmod ftdi_sio.ko and both worked. However I do not have a 
"/dev/ttyUSB..." entry. It can be that because the system does not run 
udev or similar it does not automatically create /dev entries? If so 
then how can I do this? (I see that the init srcript uses "mdev -s") 
Another issue is that this serial stuff will be loaded much later than 
kernel start so if I define "console=/dev/ttyUSB0" in the kernel 
parameters and the device will be created later then will Linux pick up 
the device at that time?


4.
Of course the 3. point will not work that easily since I have only one 
serial cable and that has a plug and a receiver so it cannot link the 
two receivers on the ends. Will I have to use a null-modem cable or 
what? It also turned out that my new computer does not even have a 
serial port outside so I will have to buy a serial adapter anyway.
What would be much simpler if I could just somehow make the USB 
connection work. What I think that there is no standard cable which has 
2 Type A plugs at both ends but it seems that if I tear down the 
surrounding metal ring from the Type A receiver then it can be plugged 
into the Type A receiver on the XO-1 and it has the correct wires in the 
correct position (since the connector has to be flipped to fit into the 
receiver). Is it correct?



Can somebody help me with any solution? (One would be enough...)
I have attached the files if somebody wants to look at them.
I use "make iso_img TARGET_PRODUCT=xo1 TARGET_BUILD_TYPE=debug" to build 
files then copy from out/debug/target/product/xo1 the files initrd.img, 
ramdisk.img, kernel and system.img into /boot in the usb drive (also 
olpc.fth). It uses the source from:

http://code.google.com/p/android-x86/wiki/GetSourceCode



android-xo1.tar.gz
Description: application/gzip
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: New F11-for-XO1 build is available and needs testing

[NoiseEHC]

I copied both files to a usb drive then booted with holding the esc key. 
Then 'copy-nand u:\os3.img' then it copied and I got back the firmware 
prompt. It did not bark about wrong crc or something like that. Then I 
switched the XO-1 off then switched on and all I can get is a Boot 
failed error message. Reimaging with 802 works so it is not a hardware 
fault. How could others run this stuff?



Steven M. Parrish wrote:
> I have taken over creating builds for the XO1 from Daniel, and have created a 
> new build of F11 for the XO1.  This build "OS3"  includes the pretty boot 
> animations, and also replaces olpcsound with csound.
>
> I would apprectiate testing and feedback.  I will have another build out 
> early 
> next week as well.
>
> Builds can be downloaded from http://dev.laptop.org/~smparrish/xo-1/builds/
>
> The usual cautions apply: this is development code, there will be
> bugs. Installation instructions: use copy-nand to install os3.img with
> os3.crc.
>
> Steven
>
> =
> Steven M. Parrish
> -
> gpg fingerprint: 4B6C 8357 059E B7ED 8095 0FD6 1F4B EDA0 A9A6 13C0
> http://tuxbrewr.fedorapeople.org/
> irc.freenode.net: SMParrish @ #fedora-kde, #fedora-devel, #fedora-olpc, #sugar
> ___
> Devel mailing list
> Devel@lists.laptop.org
> http://lists.laptop.org/listinfo/devel
>
>
>   

___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: kernel debugging

[NoiseEHC]

Somehow every time I asked about debugging in these forums I have got no 
answer at all. Now I have a weak feeling that Linux developers are "real 
programmers" who do not use debuggers. I hope it is not the case... :)
> 2.
> I looked into using the USB port with gadget support. I am attaching 
> the kernel config if you are interested in. I have compiled 
> amd5536udc.ko, g_serial.ko and g_android.ko as modules (so I can look 
> at error messages while doing an insmod). (Kernel 2.6.29, g_android.ko 
> would enable debugging via USB with the adb - Android debugging bridge.)
> If I do not insmod amd5536udc.ko, then both g_android.ko and 
> g_serial.ko bark me about missing usb_gadget_register_driver symbol.
> If I do insmod amd5536udc.ko then both modules bark about "(No such 
> device)" error.
> What is not clear whether this amd5536udc.ko is usable at all because 
> I am booting the XO-1 from an USB stick so can the AMD southbridge be 
> used for both host and device at the same time?
> What device does it need?
>
I succeeded compiling kdb into the kernel, and I have an USB keyboard 
(and successfully realized what must be compiled into the kernel) so I 
can press the Pause button to break into the debugger. However when I 
exit kdb crashes the machine so this way I cannot debug the g_android.ko 
driver.
I would appreciate some answer to any one of these questions:
1. What is the Pause button on the XO keyboard? (And which is the SysRq?)
2. If there are no such buttons then how can I change the required 
button? Is it wired into the kernel or to kdb's keyboard handler?
> 3.
> It happened that one of the hardware in my room had a FTDI usb to 
> serial converted. I pluggeg it into the XO-1 and insmod usbserial.ko 
> then insmod ftdi_sio.ko and both worked. However I do not have a 
> "/dev/ttyUSB..." entry. It can be that because the system does not run 
> udev or similar it does not automatically create /dev entries? If so 
> then how can I do this? (I see that the init srcript uses "mdev -s") 
> Another issue is that this serial stuff will be loaded much later than 
> kernel start so if I define "console=/dev/ttyUSB0" in the kernel 
> parameters and the device will be created later then will Linux pick 
> up the device at that time?
>
It seems that if I use mdev -s then /dev/ttyUSB0 appears. What I found 
is that in the hardware I pulled the FTDI converter had a cable similar 
what I just needed so I plugged the cable into my PC's motherboard, its 
plug into the serial cable, the serial cable's plug into the FTDI bridge 
and the bridge into the XO-1. Then I started getty ttyS0 or something 
similar and nothing happened. Exactly how could I test the setup?

___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Disk layout for XO-1.5

[NoiseEHC]

>
> Once BTRFS is mature, yum learns to snapshot-upgrade-or-revert and we
> switch to that combo, we will be able to retire olpc-update, the
> symlink trees and the fancy overlays.
>   
Do you have a prediction (I mean an educated guess) about when will 
BTRFS be mature? What I heard last time that they had to change the on 
disk layout so I do not know why do you seek for example firmware 
support for BTRFS. (I am really just interested in some roadmap.)
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Candidate "paper cut" bugs for a new 8.2.x release?

[NoiseEHC]

Hi!

I do not know what was the conclusion about this _completely 
hypothetical_ case but does fixing the Geode VGA driver match the "paper 
cut" criteria?


Since I am porting (very slowly since 2 months ago I did not know 
anything at all about the Linux boot process for example) Android to the 
XO-1 there were some display errors. Android uses FBIOPUT_VSCREENINFO to 
flip (and not blit!!!) between two frames in video memory and without a 
patched Geode FB driver it causes a very awful flicker since the Geode 
FB driver turns off the video output then sets params then turns on the 
video output. Similar flicker can be seen on the XO when Sugar wakes up 
from a frozen DCON so possibly my fix would be applicable here as well.


If fixing the driver is desirable then probably I could fix the other 
Geode driver bug, the video overlay fuckup during wakeup. My little 
question is this: does anybody know how and where the sleep happens? Is 
the video output turned off in the FB or the X driver? Is the FB driver 
used at all when the X driver takes control?


ps:
The attached file is not a patch but a VERY ugly "fix" for trying... :) 
Of course I will create a normal patch which can be pushed upstream but 
only if it is needed...


Martin Langhoff wrote:

In the _completely hypothetical_ case that I had some time and chance
to spin a 8.2.x release aimed at fixing the "paper cuts"[1] and
low-risk bugs that hinder XO-1 deployability _today_ in the field -
have *you* got any candidates? Tell me about them :-)

I am specially hoping to round up bugs that have patches that have
been tested in the field. So low low risk stuff that makes life easier
and better for those that really matter: kids and teachers in the
field.

No promises for now, but I sure want to find a way to do this.
Daniel's piled up a few good patches I am aware of, but I am sure
there are more (from him and others).

cheers,



martin
1 - https://wiki.ubuntu.com/PaperCut
  




geode_fix.tar.gz
Description: application/gzip
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Booting squashfs from olpc.fth

[NoiseEHC]

You probably need to do a dcon-unfreeze as well.

\ OLPC boot script
unfreeze
dcon-unfreeze
" u:\android\initrd.img" to ramdisk
" u:\android\kernel" to boot-device
" root=/dev/ram0 console=tty0 androidboot.hardware=xo1" to boot-file boot


Sebastian Silva wrote:

Hello I'm trying to boot into Trisquel like you
would boot SOAS.

My problem is with making the olpc.fth file
currently i'm trying with the following

\ Boot script for SD Boot
\ created from http://wiki.laptop.org/go/Custom_bootloader
" ro boot=casper rootdelay=1 splash console=ttyS0,115200 console=tty0 
fbcon=font:SUN12x22" to boot-file

" sd:\boot\vmlinuz" to boot-device
" sd:\boot\initrd.img" to ramdisk
unfreeze
boot

But the kernel does not seem to load as it freezes...
Let me know what other data I can give
to help me debug this.

Thanks
--
Sebastian Silva
Colectivo FuenteLibre
http://blog.fuentelibre.org/


___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel
  


___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: os33 - video regression ?

[NoiseEHC]

> To the best of my knowledge Flash has never supported the XVideo
> extension.  The reason for this is that Xv scales YUV data and Flash
> uses RGB data.  Now could this be converted and scaled absolutely, but
> Adobe has decided they are not going down that road.
>   
Xv can blit both YUV and RGB data to the overlay. I do not know why do 
not they support Xv but this cannot be the reason...
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel