KindleLazy - remote control page turns + brightness

[eruditionastic]

Ok, I think I've figured out what is going on. When I plug the device in, it thinks there is both a keyboard and a mouse active. It's putting the mouse on event2, the volume buttons (brightness) on event3, and the keyboard (up/down/tab/enter) on event4. According to the KindleLazy doc it only reads event2 and event3.

I'm not sure why it's putting a keyboard on both event3 and event4. I tried changing the symlinks to redirect the things to event2/3 but that didn't work. Probably wouldn't last anyway between starts/stops.

[root@kindle by-path]# ls -la
drwxr-xr-x    2 root     root           160 Apr  1 00:27 .
drwxr-xr-x    4 root     root           240 Apr  1 00:27 ..
root 9 Apr  1 00:27 platform-fsl-ehci.0-usb-0:1:1.0-event-mouse -> ../event2
root 9 Apr  1 00:27 platform-fsl-ehci.0-usb-0:1:1.0-mouse -> ../mouse0
root 9 Apr  1 00:27 platform-fsl-ehci.0-usb-0:1:1.1-event-kbd -> ../event3
root 9 Apr  1 00:27 platform-fsl-ehci.0-usb-0:1:1.2-event-kbd -> ../event4
root 9 Jan  7  2000 platform-imx-i2c.0-platform-max77696-onkey.0-event -> ../event0
root 9 Apr  1 00:00 platform-imx-i2c.1-event -> ../event1

It looks like changing line 291 in kindlelazy.cpp from

 eventfd2 = open("/dev/input/event2", O_RDONLY);

to

 eventfd2 = open("/dev/input/event4", O_RDONLY);

would fix my issue?

Any ideas on why my device is using event3 and event4 for the keyboard and others are not? I assume others use event2 and event3.

[NiLuJe]

Using a logical path (c.f., /dev/input/by-path) might be better to handle enumeration quirks like that .

(Plus, some Kindle variants *natively* have more stuff enumerated over USB (usually, an accelerometer or some weird extra thingy on 3G models)).

[eruditionastic]

Originally Posted by NiLuJe

Using a logical path (c.f., /dev/input/by-path) might be better to handle enumeration quirks like that .

(Plus, some Kindle variants *natively* have more stuff enumerated over USB (usually, an accelerometer or some weird extra thingy on 3G models)).

I'm not sure the absolute paths don't change all the time as well between devices. How about the following?

Returns "event4"

grep -B2 'EV=120013' /proc/bus/input/devices | grep -o 'event[0-9]'

Returns "event3"

grep -B2 'EV=1f' /proc/bus/input/devices | grep -o 'event[0-9]'

/proc/bus/input/devices contents:

Spoiler:

[root@kindle input]# cat devices
I: Bus=0000 Vendor=0000 Product=0000 Version=0000
N: Name="max77696-onkey"
P: Phys=max77696-onkey/input0
S: Sysfs=/devices/platform/imx-i2c.0/i2c-0/0-003c/max77696-onkey.0/input/input0
U: Uniq=
H: Handlers=kbd event0
B: PROP=0
B: EV=3
B: KEY=100000 0 0 0

I: Bus=0000 Vendor=0000 Product=0000 Version=0000
N: Name="cyttsp4_mt"
P: Phys=cyttsp4_mt.main_ttsp_core
S: Sysfs=/devices/platform/imx-i2c.1/i2c-1/1-0024/main_ttsp_core.cyttsp4_i2c_adapter/cyttsp4_mt.main_ttsp_core/input/input1
U: Uniq=
H: Handlers=event1
B: PROP=0
B: EV=f
B: KEY=6420 0 0 0 0 0 0 0 0 0 0
B: REL=0
B: ABS=2608000 0

I: Bus=0003 Vendor=0458 Product=0189 Version=0101
N: Name="Genius Wireless Mouse"
P: Phys=usb-fsl-ehci.0-1/input0
S: Sysfs=/devices/platform/fsl-ehci.0/usb1/1-1/1-1:1.0/input/input14
U: Uniq=
H: Handlers=event2 mouse0
B: PROP=0
B: EV=17
B: KEY=1f0000 0 0 0 0 0 0 0 0
B: REL=143
B: MSC=10

I: Bus=0003 Vendor=0458 Product=0189 Version=0101
N: Name="Genius Wireless Mouse"
P: Phys=usb-fsl-ehci.0-1/input1
S: Sysfs=/devices/platform/fsl-ehci.0/usb1/1-1/1-1:1.1/input/input15
U: Uniq=
H: Handlers=kbd event3
B: PROP=0
B: EV=1f
B: KEY=4837fff 72ff32d bf544446 0 0 1 20f90 8b17c000 677bfa d941dfed 9ed680 4400 0 10000002
B: REL=40
B: ABS=7fff01 0
B: MSC=10

I: Bus=0003 Vendor=0458 Product=0189 Version=0101
N: Name="Genius Wireless Mouse"
P: Phys=usb-fsl-ehci.0-1/input2
S: Sysfs=/devices/platform/fsl-ehci.0/usb1/1-1/1-1:1.2/input/input16
U: Uniq=
H: Handlers=kbd event4
B: PROP=0
B: EV=120013
B: KEY=e080ffdf 1cfffff ffffffff fffffffe
B: MSC=10
B: LED=1f

Explanation for the bitmasked EV field here:
https://unix.stackexchange.com/quest...t-devices-data

[NiLuJe]

That is indeed another solution, provided you're building the path dynamically .

i.e., we do that in KOReader to pickup the Accelerometer on one or both of the Oasis.

In a hard-coded compiled program, I find a by-path path usually does the job, but, yes, it depends on the exact task (and, indeed, here, the USB controller can be different and/or use a different name on various Kindle models).

EDIT: I haven't checked, but if you want to go that way in C/C++, libevdev might have higher-level tools to do that without having to reinvent the wheel . (evtest and its successor evemu certainly parse this in a human-readable format).

[eruditionastic]

FYI, I got it to work for now. I used a hex editor to patch in event4 vs event2 into the file kindlelazy in /mnt/us/extensions/KindleLazy/bin. Here's the file if it helps anyone. Sheesh, probably 20 hours pondering and trying various ways, and this way took about 5 minutes.

[eruditionastic]

I've successfully gotten KindleLazy to stop and start when a wireless presenter is plugged and unplugged with a Paperwhite 3 2015 Wifi G090G1 5.9.7.

The default behavior of KindeLazy will require a manual press of the start KindleLazy button after plugging your USB device in, and also each time after your Kindle goes to sleep. These scripts should alleviate all instances of having to press the start KindleLazy button.

This will only work if you have the same presenter as the author of KindleLazy specified. If you don't have that one see the end of the post. I've attached the 6 files required to make it work.

The first two files have to go on the OS partition.

98-StopKindleLazy.rules - Place in /etc/udev/rules.d/
99-KindleLazy.rules - Place in /etc/udev/rules.d/

The other 4 files can be placed via plugging the Kindle into your PC:

start.sh
stop.sh
udev.sh
start-reverse.sh

Place all of these in /mnt/us/extensions/KindleLazy/bin/ and that's it. We're done.



If you don't have the same presenter you can modify the two .rules files to make it work. You'll need shell access over wifi or serial, shell via USB will not suffice.

In your shell, run "udevadm monitor" and plug your device in, you'll see output similar to the following:

UDEV  [1551143135.411150] change   /devices/platform/imx-i2c.0/i2c-0/0-003c/max77696-battery.0/power_supply/max77696-battery (power_supply)
KERNEL[1551143139.121037] add      /devices/platform/fsl-ehci.0/usb1/1-1 (usb)
UDEV  [1551143139.124513] add      /devices/platform/fsl-ehci.0/usb1/1-1 (usb)
KERNEL[1551143139.125857] add      /devices/platform/fsl-ehci.0/usb1/1-1/1-1:1.0 (usb)
KERNEL[1551143139.129246] add      /devices/platform/fsl-ehci.0/usb1/1-1/1-1:1.1 (usb)
KERNEL[1551143139.132509] add      /devices/platform/fsl-ehci.0/usb1/1-1/1-1:1.2 (usb)
UDEV  [1551143139.210115] add      /devices/platform/fsl-ehci.0/usb1/1-1/1-1:1.1 (usb)
UDEV  [1551143139.212820] add      /devices/platform/fsl-ehci.0/usb1/1-1/1-1:1.0 (usb)
UDEV  [1551143139.214625] add      /devices/platform/fsl-ehci.0/usb1/1-1/1-1:1.2 (usb)

We're interested in this part - "/devices/platform/fsl-ehci.0/usb1/1-1/1-1:1.2"

Copy whatever shows up after the last "add" and run the following command:

udevadm info -a -p /devices/platform/fsl-ehci.0/usb1/1-1/1-1:1.2

You'll get a looong list of stuff, but we're only interested in the first two paragraphs. For the presenter I have, it looks like this:

 looking at device '/devices/platform/fsl-ehci.0/usb1/1-1/1-1:1.2':
    KERNEL=="1-1:1.2"
    SUBSYSTEM=="usb"
    DRIVER=="usbhid"
    ATTR{bInterfaceNumber}=="02"
    ATTR{bAlternateSetting}==" 0"
    ATTR{bNumEndpoints}=="01"
    ATTR{bInterfaceClass}=="03"
    ATTR{bInterfaceSubClass}=="01"
    ATTR{bInterfaceProtocol}=="01"
    ATTR{modalias}=="usb:v0458p0189d1001dc00dsc00dp00ic03isc01ip01"
    ATTR{supports_autosuspend}=="1"

  looking at parent device '/devices/platform/fsl-ehci.0/usb1/1-1':
    KERNELS=="1-1"
    SUBSYSTEMS=="usb"
    DRIVERS=="usb"
    ATTRS{configuration}==""
    ATTRS{bNumInterfaces}==" 3"
    ATTRS{bConfigurationValue}=="1"
    ATTRS{bmAttributes}=="a0"
    ATTRS{bMaxPower}=="100mA"
    ATTRS{urbnum}=="26"
    ATTRS{idVendor}=="0458"
    ATTRS{idProduct}=="0189"
    ATTRS{bcdDevice}=="1001"
    ATTRS{bDeviceClass}=="00"
    ATTRS{bDeviceSubClass}=="00"
    ATTRS{bDeviceProtocol}=="00"
    ATTRS{bNumConfigurations}=="1"
    ATTRS{bMaxPacketSize0}=="64"
    ATTRS{speed}=="12"
    ATTRS{busnum}=="1"
    ATTRS{devnum}=="7"
    ATTRS{devpath}=="1"
    ATTRS{version}==" 1.10"
    ATTRS{maxchild}=="0"
    ATTRS{quirks}=="0x0"
    ATTRS{avoid_reset_quirk}=="0"
    ATTRS{authorized}=="1"
    ATTRS{manufacturer}=="Genius"
    ATTRS{product}=="Wireless Mouse"

Note the values in the first paragraph for KERNEL=="1-1:1.2" and SUBSYSTEM=="usb", and ATTRS{idProduct}=="0189" in the second paragraph. Take the values in the quotes and replace the ones in the two .rules files with yours. You'll probably only need to replace the idProduct value, at least on a Paperwhite 3 2015 Wifi only.

Now run

udevadm test /devices/platform/fsl-ehci.0/usb1/1-1/1-1:1.2

And you will see if your rules are recognizing your device or not. You should see lots of text and two lines similar to the following if it is successful

udev_rules_apply_to_event: RUN '/mnt/us/extensions/KindleLazy/bin/udev.sh' /etc/udev/rules.d/99-KindleLazy.rules:5
udevadm_test: run: '/mnt/us/extensions/KindleLazy/bin/udev.sh'

Thanks for the various helps and to the author. My girlfriend finally has her Valentine's day present.

[Xuhu]

KindleLazy works on PW3 5.9.6.1, with the patched binary from eruditionastic. I'm using the same presenter the KL developer recommended (Model DSIT001 written on the box) and was getting the event4 device name as well.

After a couple KL start/stops, the event4 (kbd) device stopped showing up, but it seems a kindle reboot makes it show up again when KL starts.

Thanks to everyone who worked on KL and the jailbreak, and everything else involved!

[zxczxc]

Thanks @eruditionastic - that's great work.
Even better that it was a work out of love!

I'll see how I can include them in the main release.

About future kindles, I had a PW4 to setup and was ready to jailbreak it, but embarrassingly accidentally updated over wifi... so the chance of getting to make KindleLazy work on PW4 was delayed until I can try again

Once I remember/re-learn how to set up the necessary development tools in order to compile kindle stuff, I'll make a few little updates.

Thanks

[NiLuJe]

@zxczxc: see https://github.com/koreader/koxtoolchain for a refresher course .

[Xuhu]

After a sleep/wake cycle, KL only picks up events from the event1 ('mouse') device, and none from the event4 ('kbd') device. A restart of the kindlelazy binary in kterm fixes that. Is there a way to detect 'wake' events and reopen the input devices ? Or should the udev rules posted earlier fix this as well ?

[eruditionastic]

Originally Posted by Xuhu

After a sleep/wake cycle, KL only picks up events from the event1 ('mouse') device, and none from the event4 ('kbd') device. A restart of the kindlelazy binary in kterm fixes that. Is there a way to detect 'wake' events and reopen the input devices ? Or should the udev rules posted earlier fix this as well ?

Yeah that is the default behavior. After a real "sleep" you have to manually go press the start kindlelazy button.

The udev rules and scripts I posted should make it so you never have to press start or stop button for any reason, at least that's how mine has been working so far. I'd be gratified if someone else would test it out though before I claim it really works all the time

[zxczxc]

Thanks NiLuJe, and eruditionastic.
I should be getting a dev setup soon

Then I can compile the PW4 kernel modules for things like mouse etc, and get someone to test if it works!

[NiLuJe]

Slight caveat for kernel-related work: unless you intend to rebuild and flash a custom kernel, if you want to build modules for a production kernel, you'll have to dig up the relevant *identical* TC that Amazon used, not mine .

[wbkindle]

Thank you zxczxc and eruditionastic!

I just picked up a used PW3 today to try and run KindleLazy. I also got the same remote that zxczxc is using. I have FW 5.9.6.1 and after installing KL - I could adjust the screen brightness with the volume buttons on the remote, but the other buttons didn't do anything. I then saw eruditionastic's update with the new kindlelazy download and it started working - woohoo! Page turns worked perfectly.

I started trying out different settings and after running the 'Screen Upside Down' setting - the page turns no longer worked. Instead, both buttons now brought up the menu (same result as tapping the top of the kindle). I went back to the normal screen settings and the remote buttons still brought up the menu. I turned on and off the kindle, deleted the plugin, re-installed the plugin - still nothing. Eventually, I re-installed KUAL, along with a fresh install of KL and eruditionastic's update and got the buttons working again. Sweet!

I'm totally happy with this but if anybody has an idea why changing the screen settings would cause the remote to stop working, I'd be curious to see if there's a simple fix.

Also - would be awesome to have KL automatically start when the wireless presenter is plugged in but I'm a little lost with the first part of the instructions. Can you clarify what the OS partition is that you're talking about or let me know what I can search for to understand what it's involved? Thanks!

Originally Posted by eruditionastic

The first two files have to go on the OS partition.

98-StopKindleLazy.rules - Place in /etc/udev/rules.d/
99-KindleLazy.rules - Place in /etc/udev/rules.d/

The other 4 files can be placed via plugging the Kindle into your PC:

start.sh
stop.sh
udev.sh
start-reverse.sh

Place all of these in /mnt/us/extensions/KindleLazy/bin/ and that's it. We're done.

[knc1]

What you don't need to know:
Partition 1 (first) is the OS partition for the "Main" system (your Kindle is dual-boot).


Other questions:
If you have access to the command line (ssh, telnet, kterm), then the complete filename and the full pathname where each file belongs is shown in the quote you made.


Just 'cd' (change directory) to the full pathname, place listed file(s) there.