Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


Messages - InspiredByCas

Pages: 1 2 [3]
31
TI-Nspire / Re: Calling all Linux Kernel developers!
« on: December 02, 2012, 03:36:47 pm »
Quote
And what is in the second RAM?
There's the SRAM (flip-flops, fast) and the SDRAM (slow, has to be refreshed dynamically)

I know that already, but what is the content of these RAMs? SRAM is the normal operating RAM like in a PC, I think, but what's in the other one?

32
TI-Nspire / Re: Calling all Linux Kernel developers!
« on: December 02, 2012, 02:52:06 pm »
Quote
Or why can nClock run in background with linucx?
It cannot ;)
Like the rest, it is completely replaced by Linux and its userspace.
But how knows nClock the right time after any softreboot (linucx, chrash on calc etc.)?

Quote
Are there two different RAMs?
Yes, there are indeed two different RAMs, but the main parts of both TI's OS and Linux use the same RAM, in the same area.

And these parts which are not used by OS and linucx are these the areas where all these 'special commands' write their to?
And it this the area which is mapped to /dev/ram0 ?
And what is in the second RAM?

33
TI-Nspire / Re: Calling all Linux Kernel developers!
« on: December 02, 2012, 10:41:16 am »
There is one thing which I don't understand:
How can the OS lock the LED when it is overwritten by linucx in the RAM?
Or how can the OS restart the calc when reset button is pressed? (In earlier OS all user data was deleted when pressing reset button on the back.)
Or why can nClock run in background with linucx? I think it does because after the 'reboot' command in linucx nClock has the right time on startup, rebooting with reset button messes up nClock and it shows the wrong time.

Are there two different RAMs?

And if you write to these addresses in the RAM to do something special (watchdog, LED, power management), wouldn't the OS or linucx (or whatever runs) be overwritten partially?

34
TI-Nspire / Re: Calling all Linux Kernel developers!
« on: December 01, 2012, 02:00:57 pm »
Try starting linucx in Press-to-test Mode. If you put the linuxloader (or the script) in the ndless startup folder, it should be executed, even in Press-to-test Mode. I am not sure if it works with linuxloader, but other things in ndless startup folder are executed in Press-to-test Mode.
The reason why I write this, is that Norse (the cool app for morsing around with the LED, not compiled for 3.1) worked sometimes when started i Press-to-test Mode, as I read somewhere.

You can exit Press-to-test Mode anytime by sending a complete EMPTY file with the file name 'Exit Test Mode.tns' (just create a empty text file and rename it, but watch out, it's case-sensitive!) to the - regulary only and empty - folder in the CAS.
To start this Mode, just hold Esc+On when calc is off.
Then try your LED code again and try reading if the values in the memory are different from normal mode (they should be, even before writing new values), because the LED blinks in Press-to-test mode.

Also cool news with the power management :)

35
TI-Nspire / Re: Calling all Linux Kernel developers!
« on: November 29, 2012, 11:14:39 am »
Is there any chance that it will support USB 2.0?

36
TI-Nspire / Re: Calling all Linux Kernel developers!
« on: November 27, 2012, 11:54:22 am »
Quote
And I'm too lazy to change the version number every 10 minutes..
Add an automated script that adds a timestamp to the version number so you don't have to?

The main problem was that he don't know which kernel with which features he is testing.....

37
TI-Nspire / Re: Calling all Linux Kernel developers!
« on: November 27, 2012, 06:40:44 am »
Quote
(even if the framebuffer doesn't work)
I don't think the initrd and framebuffer problem are related..
I could really have tested the wrong kernel the whole time *embarrassed*

Try making a different login text for each kernel...
That will help maybe.

38
TI-Nspire / Re: Calling all Linux Kernel developers!
« on: November 24, 2012, 09:48:35 am »
Yeah! ;D
My 'USB power upgrader' works!
It automatically switches on when input voltage from the CAS comes and switches off automatically when input voltage is off (when the calc is off through ctrl+on). A little green LED displays the state.
I can connect the output to a USB fan, and it's working...

BUT: when i connect to USB stick, the calc reboots instantly, even if it is running linux. (edit: oh, only with the very first version of linucx, with the newer the stick lights up) I think I connected the data wires wrong, so that data+ is connected to data- and vise versa. Not even one device is displayed by lsusb (watch -n1 lsusb).
So it restarts to prevent corruption of any devices...
Connecting to a hub doesn't do anything... not even hub LED is working... Only power consumption goes up to 550 mA.
I will try swapping the data wires...

If it is finished, I will post pictures.


If it is totally off, it doesn't need any power from the batteries (0 mA), when input comes in, it needs about 5 mA, and with my USB fan it needs 200 mA.
The power consumption at the batteries should be double as at the USB output port, because the L7805 has a efficiency of ~50% with 9 volts input.


1. EDIT: Just the power works, data transfer is not working, because of the two separated input/output power circuits, but the USB host needs to connect the data wires to ground to enumerate USB devices.

Now inventing a complete different circuit...

2. EDIT: My new circuit works! Even hubs are working with less power consumption than before :D Just the off-switching part does not work correctly, there is always a output... Working on it.
The trick was - as I thought - the GND, which must be the same for input and output.

39
TI-Nspire / Re: Calling all Linux Kernel developers!
« on: November 24, 2012, 03:47:10 am »
Quote
Here, the "build yourself" means for me to open emacs, and start coding. Is that what you wanted to say?
You could do that if you have the patience and time, but http://hackspire.unsads.com/wiki/index.php/C_and_assembly_development_introduction_on_Linux should be faster if you're not trying to compile it on a 4004 ;D
Your prefix will be arm-none-eabi-, you just have to copy the binarys created into /usr/bin, /opt/cross or whatever directory is in your $PATH.
Ah ! For sure I already have that.
But now imagine I have succesfuly compiled the kernel, how to compile stuff that is not included in that kernel ? (for example, how to you manage to add the extra-programs that are available by default when you run your kernel ?).
Maybe I've missed one post, but I cannot figure out how you're dealing with the *double* cross compiler (linux <> arm nspire <> arm linux).

Other stuff which is not a driver, is in the initial ramdisk (initrd) which contains the root filesystem, or, since we have USB support, this root filesystem can be copied to a USB stick and it is mounted at boot time like a real hard disk in a real desktop PC running linux.
So once all drivers and stuff which should be in the kernel, is there, boot it and start install apps on the nspire. For that, a USB stick is needed, because it can be mounted writeable.

So I understand this so far, or am I wrong?


btw: how can I give +1 to a post?

40
TI-Nspire / Re: Calling all Linux Kernel developers!
« on: November 23, 2012, 07:27:59 am »
So, I really like linux on ti-nspire, but I want to improve the kernel a little bit, with new commands, inbuild apps and something. But I don't know how  ??? . O can not get the zimage file extracted  :banghead: . So i thought its identic with the files uploaded to Github , but these files have a size of 125mb.

It would be nice if someone could gelp me or if someone could upload the extracted zimage file.

Greetings and waiting for response Stennwellara.

Maybe this helps: http://www.linuxquestions.org/questions/programming-9/how-to-obtain-image-from-zimage-824499/
First link googleing 'decompress zimage'.

But I think, it would be better to compile the kernel new from source. Use buildroot for that, somewhere in the output folder there is a cross-compiler or so... I dodn't understand that, too.

41
TI-Nspire / Re: Calling all Linux Kernel developers!
« on: November 22, 2012, 07:16:46 am »
I changed my plan a bit and decided to make two USB A connectors: At one the input from my self-made OTG cable comes in (so I don't have to cut my cable) and on the other the data and 5 volt comes out (going to a USB hub for example). This is because I didn't want to buy parts which I need (I hate waiting for deliveries), so I want to use parts I found in some old devices. And I just found a double USB A connector on a old mainboard... You can desolder them with a hot air pistol (min. 550°C). :devil: Watch out for dangerous gases when some parts (eg. plastics) melt and the mainboard makes bubbles...
These gases make a good headache. XD

And I think a transistor will do the on/off switching thing, just a LED will display state...

42
TI-Nspire / Re: Calling all Linux Kernel developers!
« on: November 21, 2012, 02:39:46 pm »
Haha, if you short any wires on on the connected CAS with OTG cable it displays a weird message that I should reconnect the device... :w00t:

Yes, i know if I do this too often, it will brick my calc...

43
TI-Nspire / Re: Calling all Linux Kernel developers!
« on: November 21, 2012, 06:36:16 am »
external power support would be really helpful, so i'm betting people will be interested if you can get it working =)
Yeah. I opened a mini USB b plug and shortened the ground (5. pin) with with sense (4. pin) by putting some solder on the back of the contacts. then I used a file to remove a bit of the solder which was too much and would have blocked the metal case of the plug.
If I connect this cable to my CAS and start linux, it gives 5 volts output at the other end of the cable, so that proves the calculator recognizes the short between ground and sense and acts like a USB host.
I found a picture which describes that:
top part is normal usb cable, lower part is OTG cable with pin 4 and 5 shortened.


Now I will cut the cable and add the external power supply. It will a 9 volt battery (or accu) which is connected to a L7805 (a thing with 3 pins and a heatsink connection on the back, can be found in some devices, I found one in a automtic call responser (<= I think there is a better word for that ;)), that will make exactely 5.00 volts independent of input voltage. This power will go into the output of my power supply, but it should not go out to the input, that would destroy my CAS I think.
With a switch, a 'low batt' warnig LED or something like that, it is ideal to use it everywhere and cheat everywhere :D
The output can be connected to any regular USB Hub and I think with a full battery it can supply power to very 'hungry' devices :) The only problem with much devices would be that the L7805 can become very very hot (much more than 100°C)...

I will inform you about my progress...

44
TI-Nspire / Re: Calling all Linux Kernel developers!
« on: November 20, 2012, 04:23:18 pm »
I registered in this forum because of this great topic and I want to support the people here as best as I can. :)

I can code, but I never made (big) things in C/C++...

I'm also going to build a USB Hub for my CAS, with power supply from a battery ;)...
If someone is interested, I will publish the parts and the wirings...

Pages: 1 2 [3]