KindleLazy - remote control page turns + brightness

[knc1]

Originally Posted by shamanNS

I don't have KT3 so I can't be expected to know things like that for certain.
- - - -

Whoa now.
You asked the questions, I answered the questions.

Nowhere did I make the assumption that you where expected to know those things for certain.

As for your specific concern -
All i.MX6 based Kindle devices have a working OTG port.
Older Kindle devices had the OTG port but also had various hardware and software issues that made it essentially unusable.

Specific device support out of the box is whatever the vendor decides to support (such as their own product(s)).
It is designed and built as an e-book reader, not a general purpose computer system (which is the way we in this forum treat it).

[zxczxc]

if anyone with a jailbroken KT3 who is comfortable, has kterm or ssh, could they do a:

ls /dev/input

for me please? and tell me what they see? (screenshot is ok)
(without any usb device inserted, without KindleLazy running etc)

surely the KT3 would have the same input devices, of touch screen and power button, like the paperwhite...
but since the report of KL not working... we need to start the tests somewhere!

[knc1]

I have one, not jail broken yet but I can probably JB it if needed.

So I will leave this for another reader to answer.
But if it takes more than a day for a response, PM me a reminder.

The helpdesk is temporarily down to one person, anyone is welcome to contribute.

[jscris]

If there's someone with a jailbroken Voyage who can get the keycodes for the light adjustment, that would be great, too. Assuming light adjustment will be possible on the KV.... The-ebook-reader.com blog has the video and article.

[zxczxc]

Originally Posted by jscris

If there's someone with a jailbroken Voyage who can get the keycodes for the light adjustment, that would be great, too. Assuming light adjustment will be possible on the KV.... The-ebook-reader.com blog has the video and article.

The keycodes are exactly the same.
By keycodes/the numbers in the config file, they are just if you want to do something like...
connect a normal keyboard to KindleLazy...
you could edit the config file to make 'K' next page, '7' and 'W' previous page, spacebar for brightness up, etc etc...

Since everyone so far seems to be using the exact same controller I have, you don't need to change the codes because brightness is already linked to the volumeup/down buttons


The problem is KindleLazy/me!
I can make something easily and fast to make brightness work, but ideally I want something future proof that will work in a nice way, for example even if we plug in 4 keyboards and two mice, I still want it to work. At the moment, it wouldn't, they would all be ignored apart from the first 1 or 2.

Honestly if you want it and need it now/soon I'll do it, let me know!
But if you are just asking/telling to be helpful, don't worry, I know about it thanks

[jscris]

Originally Posted by zxczxc

The keycodes are exactly the same.

The problem is KindleLazy/me!
I can make something easily and fast to make brightness work, but ideally I want something future proof that will work in a nice way, for example even if we plug in 4 keyboards and two mice, I still want it to work. At the moment, it wouldn't, they would all be ignored apart from the first 1 or 2.

Honestly if you want it and need it now/soon I'll do it, let me know!
But if you are just asking/telling to be helpful, don't worry, I know about it thanks

Thanks! Now I understand. It's not important enough for you to spend time on it now, but thank you so much for the offer! The one more thing I would love is for page turns to work with koreader, but I understand that may not be going to happen. I am extremely happy with it as it is.

[zxczxc]

Originally Posted by jscris

Thanks! Now I understand. It's not important enough for you to spend time on it now, but thank you so much for the offer! The one more thing I would love is for page turns to work with koreader, but I understand that may not be going to happen. I am extremely happy with it as it is.

The reason why page turns work but brightness doesn't is because the controller we have is weird...
When plugged in, it appears to the computer as I think 4 separate devices... strange!
The volume buttons (that we use in KL for brightness) are on a separate 'device' to the page turning buttons.

Kindle paperwhite's input devices are:
input0 - this is only the power button.
input1 - this is the touch screen.
that's it. when we plug my usb controller and start KindleLazy we see...
input2 - this contains keys like pageup/page down, that we use to turn pages.
input3 - the volume keys happen to be here.

So I made KindleLazy read from input2 and input3, to get all the buttons I wanted. A different, similar presentator may appear just as 1 device, and so all keys might be on input2 - no problem.

On your Kindle Voyage you have:
input0 - power button
input1 - touch screen
input2 - the real PAGE TURN BUTTONS, paperwhite doesn't have this.
then when you insert the usb and start KL...
input3 - controller buttons containing pageup/down
input4 - controller buttons containing volume

BUT KindleLazy only reads from input2 and 3... not input4 where the part of the controller appears as on Voyage.
So the 'input' the volume button commands are coming from is being ignored... thus no brightness control!

That's the full explanation


Ideally I would like to be able to simply 'detect' what inputs the usb device is, and read them. But it's easier said than done. Hardcoding it to input2 and input3 was so easy, but as you can see, ugly and doesn't work in all cases.



And yeah, lucida - who showed how to make it work with koreader, I tried but for me it didn't work. When I next work on KL I'll try again, maybe it was a silly mistake of mine. I'm a bit tired of kindle at the moment so it will be a few days at least

[jscris]

Originally Posted by zxczxc

Ideally I would like to be able to simply 'detect' what inputs the usb device is, and read them. But it's easier said than done. Hardcoding it to input2 and input3 was so easy, but as you can see, ugly and doesn't work in all cases.

And yeah, lucida - who showed how to make it work with koreader, I tried but for me it didn't work. When I next work on KL I'll try again, maybe it was a silly mistake of mine. I'm a bit tired of kindle at the moment so it will be a few days at least

Got it! And I will be patient!

[knc1]

Originally Posted by zxczxc

- - - -
Ideally I would like to be able to simply 'detect' what inputs the usb device is, and read them. But it's easier said than done. Hardcoding it to input2 and input3 was so easy, but as you can see, ugly and doesn't work in all cases.

As in examine the /sys tree?

Start with /sys/class/input/* and see if you can find something useful.
I.E:
on my pw-3, version 5.6.1.1 (oldest there is for pw-3) -
input0 is linked to a location with max77696-onkey (the keyboard) in the pathname.
input1 is linked to a location with main_ttsp_core (the touch screen) in the pathname.

I expect the KT(2, 3) to have a different name for the touch screen, since they use the IR scanner of the KT rather than the capacitative touch screen of the others.

[lucida]

KT3 definitely wouldn't work since it uses a different 3.10.53 kernel even it still features an i.MX6. insmod will fail with current set of usbhid kernel modules(since these were compiled with PW3 kernel source) so we don't even have proper /dev/input entry to begin with.

KT2/PW2/PW3/KV/KOA are all iMX6 with a 3.0.35 kernel, among these 5 devices, PW2 KT2 are untested rest 3 300ppi devices are confirmed to work with USB OTG.

Without KT3 I really can't do much here. On the other hand, wifi remote should work regardless hardware/OTG.

Actually I'm using wifi remote myself with a usb mouse on an openwrt router.

[knc1]

What is the problem with re-compiling your module for a different kernel version?

Edit:
I can answer my own question now -

The sources of the kernel modules are not in the release tar-ball nor are they in your public repository src branch.

You do realize that the kernel is GPLv2 and that means the source of the kernel binaries must be available on request?

I am requesting with this post.

I also suggest that you include your module sources (and corresponding kernel source tree) in your repository.

(Which would not only meet the requirements of GPLv2, but make it simpler to support multiple kernel versions.)

core2quad bin $ ls *.ko
hid.ko  mousedev.ko  usbhid.ko
core2quad bin $ modinfo ./hid.ko
filename:       ./hid.ko
license:        GPL
author:         Jiri Kosina
author:         Vojtech Pavlik
author:         Andreas Gal
depends:        
vermagic:       3.0.35-lab126 preempt mod_unload ARMv7 p2v8 
parm:           debug:toggle HID debugging messages (int)
core2quad bin $ modinfo ./mousedev.ko
filename:       ./mousedev.ko
license:        GPL
description:    Mouse (ExplorerPS/2) device interfaces
author:         Vojtech Pavlik <vojtech@ucw.cz>
alias:          input:b*v*p*e*-e*1,*3,*k*110,*r*a*0,*1,*m*l*s*f*w*
alias:          input:b*v*p*e*-e*1,*3,*k*145,*r*a*0,*1,*18,*1C,*m*l*s*f*w*
alias:          input:b*v*p*e*-e*1,*3,*k*14A,*r*a*0,*1,*m*l*s*f*w*
alias:          input:b*v*p*e*-e*1,*2,*k*r*8,*a*m*l*s*f*w*
alias:          input:b*v*p*e*-e*1,*2,*k*110,*r*0,*1,*a*m*l*s*f*w*
depends:        usb_dr
vermagic:       3.0.35-lab126 preempt mod_unload ARMv7 p2v8 
parm:           xres:Horizontal screen resolution (uint)
parm:           yres:Vertical screen resolution (uint)
parm:           tap_time:Tap time for touchpads in absolute mode (msecs) (uint)
core2quad bin $ modinfo ./usbhid.ko
filename:       ./usbhid.ko
license:        GPL
description:    USB HID core driver
author:         Jiri Kosina
author:         Vojtech Pavlik
author:         Andreas Gal
alias:          usb:v*p*d*dc*dsc*dp*ic03isc*ip*
depends:        hid
vermagic:       3.0.35-lab126 preempt mod_unload ARMv7 p2v8 
parm:           mousepoll:Polling interval of mice (uint)
parm:           ignoreled:Autosuspend with active leds (uint)
parm:           quirks:Add/modify USB HID quirks by specifying  quirks=vendorID:productID:quirks where vendorID, productID, and quirks are all in 0x-prefixed hex (array of charp)

[lucida]

Originally Posted by knc1

What is the problem with re-compiling your module for a different kernel version?

Edit:
I can answer my own question now -

The sources of the kernel modules are not in the release tar-ball nor are they in your public repository src branch.

You do realize that the kernel is GPLv2 and that means the source of the kernel binaries must be available on request?

I am requesting with this post.

I also suggest that you include your module sources (and corresponding kernel source tree) in your repository.

(Which would not only meet the requirements of GPLv2, but make it simpler to support multiple kernel versions.)

core2quad bin $ ls *.ko
hid.ko  mousedev.ko  usbhid.ko
core2quad bin $ modinfo ./hid.ko
filename:       ./hid.ko
license:        GPL
author:         Jiri Kosina
author:         Vojtech Pavlik
author:         Andreas Gal
depends:        
vermagic:       3.0.35-lab126 preempt mod_unload ARMv7 p2v8 
parm:           debug:toggle HID debugging messages (int)
core2quad bin $ modinfo ./mousedev.ko
filename:       ./mousedev.ko
license:        GPL
description:    Mouse (ExplorerPS/2) device interfaces
author:         Vojtech Pavlik <vojtech@ucw.cz>
alias:          input:b*v*p*e*-e*1,*3,*k*110,*r*a*0,*1,*m*l*s*f*w*
alias:          input:b*v*p*e*-e*1,*3,*k*145,*r*a*0,*1,*18,*1C,*m*l*s*f*w*
alias:          input:b*v*p*e*-e*1,*3,*k*14A,*r*a*0,*1,*m*l*s*f*w*
alias:          input:b*v*p*e*-e*1,*2,*k*r*8,*a*m*l*s*f*w*
alias:          input:b*v*p*e*-e*1,*2,*k*110,*r*0,*1,*a*m*l*s*f*w*
depends:        usb_dr
vermagic:       3.0.35-lab126 preempt mod_unload ARMv7 p2v8 
parm:           xres:Horizontal screen resolution (uint)
parm:           yres:Vertical screen resolution (uint)
parm:           tap_time:Tap time for touchpads in absolute mode (msecs) (uint)
core2quad bin $ modinfo ./usbhid.ko
filename:       ./usbhid.ko
license:        GPL
description:    USB HID core driver
author:         Jiri Kosina
author:         Vojtech Pavlik
author:         Andreas Gal
alias:          usb:v*p*d*dc*dsc*dp*ic03isc*ip*
depends:        hid
vermagic:       3.0.35-lab126 preempt mod_unload ARMv7 p2v8 
parm:           mousepoll:Polling interval of mice (uint)
parm:           ignoreled:Autosuspend with active leds (uint)
parm:           quirks:Add/modify USB HID quirks by specifying  quirks=vendorID:productID:quirks where vendorID, productID, and quirks are all in 0x-prefixed hex (array of charp)

amazon released wrong source code for 5.8.2
grab 5.8.1 source code you'll get all the codes then.

[knc1]

Originally Posted by lucida

amazon released wrong source code for 5.8.2
grab 5.8.1 source code you'll get all the codes then.

The license states that information must be included with every binary.

Please fix the repository and the release archives.

[lucida]

Originally Posted by knc1

The license states that information must be included with every binary.

Please fix the repository and the release archives.

thanks for your friendly request.

have fun.

https://kindle.s3.amazonaws.com/Kind...4910023.tar.gz

[lucida]

OK, all BS aside, for KT3 to work: if there is no OTG support in stock hardware/kernel then we're out of luck. Otherwise, someone needs to build KT3 specified modules. Dropping PW3 prebuilt kernel module in KT3 wouldn't work since they have different kernel version.