Initial CX II support #165

Vogtinator · 2019-04-30T17:51:47Z

Boots ~~(a bit)~~
~~Needs some refactoring~~
~~Some parts are missing suspend/resume support~~
~~Flash creation dialog mostly useless~~
~~No USB support (neither controller nor protocol)~~
Some parts for complete CX II support are missing, see TODO.md

EnumC · 2019-07-24T07:50:32Z

@Vogtinator Thanks for all the work you've done! Were you able to dump boot1, boot2, manuf, os, and diag so far? If so, how did you go about doing it?

Vogtinator · 2019-07-24T11:39:16Z

Were you able to dump boot1, boot2, manuf, os, and diag so far?

maybe

flashdialog.cpp

Vogtinator · 2019-10-26T17:51:35Z

Yay:

adriweb · 2019-10-26T17:52:09Z

\o/

debrouxl · 2019-10-26T20:10:08Z

Congratulations :)

mosamadeeb · 2019-11-14T13:17:13Z

Is it working fine? Why did the Travis build fail?

mosamadeeb · 2019-11-24T04:58:41Z

@Vogtinator I can't build this pr in qt, is there a certain commit that's causing the problem? same as what's failing travis?

Vogtinator · 2019-11-24T11:10:03Z

@SutandoTsukai181 What is the error message? The error in travis is caused by a too old compiler.

Vogtinator · 2020-03-15T11:43:52Z

@Vogtinator I found a linux kernel source with drivers for a LOT of Faraday IPs (there are more of them outside mach-faraday too, like FOTG210)

https://github.com/SleepyRabbit/arm-linux-3.3/tree/master/linux-3.3-fa/arch/arm/mach-faraday

None of those (except for the DMAC) are used in the CX II and FOTG210 is actually part of mainline: https://github.com/torvalds/linux/blob/master/drivers/usb/gadget/udc/fotg210.h

Vogtinator · 2020-03-15T11:50:20Z

The only missing emulation parts here are nspire specific ones (like the new Aladdin PMU) and figuring out some quirks like why the On/Home button doesn't work.
Some cleanup is also necessary before this can be merged...

zephray · 2020-10-16T02:53:55Z

Hi, sorry about the probably stupid question, but how do I install OS for CX II in the emulator?
I have a copy of boot1, boot2, manuf, as well as complete NAND dump from my CX II-T CAS (not a CX II CAS, this might cause some issue).

I first tried to use the create flash dialog, I am able to get it to create a image that contains valid manuf information, but I would have to manually install boot2 via xmodem. I have tried either my dump or the img from OS update file would work. Once that's installed, it could boot into maintenance menu, but then I wasn't able to send the OS update file over via USB. (Nothing happens when file is being drag-n-dropped there).

Then I thought probably preload OS image no longer works and the USB is not fully working, I could go with full NAND dump. I modified the code to use load_file_part() function in flash.c to pretty much convert the NAND dump (128MB) to the NAND with OOB data (132MB) that the emulator uses. This time it could boot to the os loader stage then hit a prefetch abort:

BOOT: loading complete (75 ticks), launching image.
Boot ROM Time: 0.33 seconds
Boot Loader Time: 0.75 seconds
Warning (13ef0000): Prefetch abort: address=13ef0000 status=0

I am wondering what would be the correct way to install the OS?
Thanks

adriweb · 2020-10-16T04:08:39Z

Are you sure you have an actual full dump of your boot1/2 of the CX II? (Specifically the part with the keys, IIRC (?))

Because TI put some additional security stuff there that makes dumps not work the usual way :P
Vogtinator was able to work around those "security" measures, but for now the tools aren't public just yet.
Once it's ready, the files should work just fine as usual on Firebird.

You can come to #ndless-dev on IRC (EFNet servers) to talk more directly if needed.

zephray · 2020-10-16T04:11:49Z

I don't know, that's why I am asking. If that's the case then I am pretty much sure my dumps are not complete/ correct.

Vogtinator · 2020-10-16T06:20:47Z

Yeah, that's the usual symptom of an incomplete bootrom dump. The OS doesn't get decrypted, so it executes all zeros, eventually running off past the end of RAM.

youssef112358 · 2020-11-01T13:21:58Z

i appreciate all the work you do guys and i wish i could help, but i just begun learning programming so i wont be able to help anytime soon that why i don't really understand evrything you say but i just wanted to ask if i can emulate the cx ii using firebird on android or is not not ready yet ? i now the question is a bit stupid but i'm very new to this "world" and i hope i didn't bother you with my question
and thanks a hole lot for all the amazing work you do

Vogtinator · 2020-11-01T13:31:03Z

i just wanted to ask if i can emulate the cx ii using firebird on android or is not not ready yet ?

The main missing part for that is a new dumping program, which is WIP. On top of that, there are some bugs and unimplemented features, but those aren't that common.

youssef112358 · 2020-11-01T19:20:17Z

i just wanted to ask if i can emulate the cx ii using firebird on android or is not not ready yet ?

The main missing part for that is a new dumping program, which is WIP. On top of that, there are some bugs and unimplemented features, but those aren't that common.

ok thanks a hole lot for all the work you do and your time,i really appreciate it and i hope you find all the help you need for your work to pay off

Works mostly, but the cursor jumps around sometimes. TODO: - Implement reset, suspend, resume - Refactor synaptics touchpad to use the proper I2C abstraction - Simplify touchpad_captivate_read

Always return that something moved/changed, even when there's no contact. That way it actually resets the reference for relative motion.

Reverts last commit, as returning 1<<3 unconditionally broke the power saving mode. The cursor jumping doesn't happen anymore for some reason.

Avoids an endless loop of SET_CONFIGURATION events. Additionally, fix usb_cx2_bus_reset_off not getting called as the OS resets the host controller on connect, which in turn resulted in usblink_reset.

It's not entirely clear why this is necessary, but it helps both on HW and with emulation. Most importantly, this fixes a delay on reconnection as only the second attempt of requesting an address succeeds.

Implement all FDMAs and merge sending and receiving into one function.

On real hardware, that's what the values during runtime are. Previously, it would clear the divider in 0x020 and thus run at twice the speed.

UI parts missing so far. There's no mechanism for preloading the OS AFAICT, so default to booting the installer.

The OS uses those when sending files to the host.

This avoids weird behaviour after resetting when there was garbage data.

usblink calls the correct functions depending on emulate_cx2 now.

- Makes more sense - Also avoid warnings due to unused arguments

Not needed

The values are probably the PWM pulse width and period.

Otherwise it errors when it doesn't need to, for instance when the CX II OS enters deep sleep.

…ep sleep With this, it enters deep sleep successfully, but the sleep itself isn't really implemented. Waking it up manually by jumping to 0 returns it to the home screen, but the clock values are wrong and the watchdog triggers.

It's actually wrong here, only mmio_* needs it.

Just remember written values. This fixes the CX II bootrom bootlooping when PTT is enabled.

jake01756 · 2021-04-17T22:14:53Z

The flash creation dialog for CX II is incomplete and doesn't have the right labels yet

Not only is it incomplete, the file picker on macOS doesn't let me pick an extension that ends in filename.img.tns

sussycatgirl · 2023-03-03T21:56:21Z

i just wanted to ask if i can emulate the cx ii using firebird on android or is not not ready yet ?

The main missing part for that is a new dumping program, which is WIP. On top of that, there are some bugs and unimplemented features, but those aren't that common.

Heya, sorry to bump this old thread but are there any updates on this?

Vogtinator · 2023-03-03T21:57:53Z

CX II support in firebird is pretty much complete, but a new dumping program is still not available unfortunately.

ddsmith7792 · 2023-11-13T23:11:05Z

Sorry to revive an old and dead thread but has there been any developments?

Vogtinator · 2023-11-13T23:22:03Z

Still the same as #165 (comment)

MinePossu · 2024-11-11T08:31:11Z

Pardon me, but what is the eta for the dumper program?

Vogtinator force-pushed the cxII-clean branch from 66df7c3 to 192e0b4 Compare July 27, 2019 09:26

Vogtinator force-pushed the cxII-clean branch from c2c9057 to fa5bcb8 Compare October 24, 2019 19:25

adriweb reviewed Oct 24, 2019

View reviewed changes

flashdialog.cpp Outdated Show resolved Hide resolved

Vogtinator force-pushed the cxII-clean branch 2 times, most recently from 8e960b0 to ff85cb9 Compare March 14, 2020 14:38

adriweb mentioned this pull request Jun 18, 2020

Support for OS 5.0/5.1 ndless-nspire/Ndless#177

Closed

Vogtinator force-pushed the cxII-clean branch 2 times, most recently from c7aec43 to 9b8567d Compare September 16, 2020 17:42

Vogtinator force-pushed the cxII-clean branch from a92a1ed to 6f469ad Compare October 2, 2020 18:57

Vogtinator mentioned this pull request Oct 5, 2020

How to let emulator loading cx ii cas rom #218

Closed

Vogtinator force-pushed the cxII-clean branch from 6f469ad to 36789a9 Compare January 24, 2021 15:57

Vogtinator force-pushed the cxII-clean branch 2 times, most recently from 857fdc8 to eb06373 Compare March 28, 2021 18:54

Vogtinator added 21 commits April 10, 2021 14:07

WIP: capTIvate touchpad support

2450a3a

Works mostly, but the cursor jumps around sometimes. TODO: - Implement reset, suspend, resume - Refactor synaptics touchpad to use the proper I2C abstraction - Simplify touchpad_captivate_read

Avoid cursor jumping with captivate

711e825

Always return that something moved/changed, even when there's no contact. That way it actually resets the reference for relative motion.

Finish captivate implementation

fb1447d

Reverts last commit, as returning 1<<3 unconditionally broke the power saving mode. The cursor jumping doesn't happen anymore for some reason.

Fix OS freezing after reconnecting USB

1b0ec54

Avoids an endless loop of SET_CONFIGURATION events. Additionally, fix usb_cx2_bus_reset_off not getting called as the OS resets the host controller on connect, which in turn resulted in usblink_reset.

Improve reliability of usblink_cx2 handshake

52f4ec1

It's not entirely clear why this is necessary, but it helps both on HW and with emulation. Most importantly, this fixes a delay on reconnection as only the second attempt of requesting an address succeeds.

usb_cx2: Generalize FDMA handling

3e9e0b1

Implement all FDMAs and merge sending and receiving into one function.

Take default clock values from the CX II OS

1876eda

On real hardware, that's what the values during runtime are. Previously, it would clear the divider in 0x020 and thus run at twice the speed.

Implement flash preloading for the CX II

fc02968

UI parts missing so far. There's no mechanism for preloading the OS AFAICT, so default to booting the installer.

Send INT_POWER on On key presses on the CX II as well

4895d1e

usb_cx2: Allow USB IN transfers bigger than the FIFO size

3cd157b

The OS uses those when sending files to the host.

Clear more in touchpad_cx_reset

7c1db9d

This avoids weird behaviour after resetting when there was garbage data.

Allow coexistance of usb and usb_cx2

fc52d87

usblink calls the correct functions depending on emulate_cx2 now.

Update TODO.md

0f7f275

Rename nope_(read/write)_(byte/half/word) to null_

f57da79

- Makes more sense - Also avoid warnings due to unused arguments

Remove spurious include in misc.h

370dccb

Not needed

Implement CX II backlight handling

7be75a7

The values are probably the PWM pulse width and period.

Error only when a write to the fast timer actually enables it

e2c09c8

Otherwise it errors when it doesn't need to, for instance when the CX II OS enters deep sleep.

Update TODO.md

22ef8fb

Don't use FASTCALL for null_(read/write)_(byte/half/word)

3b6318a

It's actually wrong here, only mmio_* needs it.

Implement LED registers

bf0ee23

Just remember written values. This fixes the CX II bootrom bootlooping when PTT is enabled.

Vogtinator force-pushed the cxII-clean branch from d6b0ad7 to bf0ee23 Compare April 10, 2021 12:07

Vogtinator merged commit 05cb7b0 into nspire-emus:master Apr 10, 2021

Vogtinator deleted the cxII-clean branch April 10, 2021 12:10

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Initial CX II support #165

Initial CX II support #165

Vogtinator commented Apr 30, 2019 •

edited

Loading

EnumC commented Jul 24, 2019

Vogtinator commented Jul 24, 2019

Vogtinator commented Oct 26, 2019

adriweb commented Oct 26, 2019

debrouxl commented Oct 26, 2019

mosamadeeb commented Nov 14, 2019

mosamadeeb commented Nov 24, 2019

Vogtinator commented Nov 24, 2019

Vogtinator commented Mar 15, 2020 •

edited

Loading

Vogtinator commented Mar 15, 2020

zephray commented Oct 16, 2020

adriweb commented Oct 16, 2020 •

edited

Loading

zephray commented Oct 16, 2020

Vogtinator commented Oct 16, 2020

youssef112358 commented Nov 1, 2020

Vogtinator commented Nov 1, 2020

youssef112358 commented Nov 1, 2020 •

edited by adriweb

Loading

jake01756 commented Apr 17, 2021 •

edited

Loading

sussycatgirl commented Mar 3, 2023

Vogtinator commented Mar 3, 2023

ddsmith7792 commented Nov 13, 2023

Vogtinator commented Nov 13, 2023

MinePossu commented Nov 11, 2024

Initial CX II support #165

Initial CX II support #165

Conversation

Vogtinator commented Apr 30, 2019 • edited Loading

EnumC commented Jul 24, 2019

Vogtinator commented Jul 24, 2019

Vogtinator commented Oct 26, 2019

adriweb commented Oct 26, 2019

debrouxl commented Oct 26, 2019

mosamadeeb commented Nov 14, 2019

mosamadeeb commented Nov 24, 2019

Vogtinator commented Nov 24, 2019

Vogtinator commented Mar 15, 2020 • edited Loading

Vogtinator commented Mar 15, 2020

zephray commented Oct 16, 2020

adriweb commented Oct 16, 2020 • edited Loading

zephray commented Oct 16, 2020

Vogtinator commented Oct 16, 2020

youssef112358 commented Nov 1, 2020

Vogtinator commented Nov 1, 2020

youssef112358 commented Nov 1, 2020 • edited by adriweb Loading

jake01756 commented Apr 17, 2021 • edited Loading

sussycatgirl commented Mar 3, 2023

Vogtinator commented Mar 3, 2023

ddsmith7792 commented Nov 13, 2023

Vogtinator commented Nov 13, 2023

MinePossu commented Nov 11, 2024

Vogtinator commented Apr 30, 2019 •

edited

Loading

Vogtinator commented Mar 15, 2020 •

edited

Loading

adriweb commented Oct 16, 2020 •

edited

Loading

youssef112358 commented Nov 1, 2020 •

edited by adriweb

Loading

jake01756 commented Apr 17, 2021 •

edited

Loading