[OK now]Kindle7 ($99) 5.6.0.1.6 serial jailbreak meet koreader issue..

[zfu]

I jailbreaked my kindle 7 ($99) by TTL2USB serail!
Boot up to dialog mode by bootm 0xE41000 and login root with passwd 'mario'.
Then I removed the main mode root passwd in file /etc/shadow of mmcblk0p1.



I already got root access of the main boot mode, installed jb-1.14.N and KUAL. I'm able to launch KUAL.

But when I try install koreader by unzip koreader-kindle-arm-linux-gnueabi-v2014.11-222-gdb420d2.zip from /mnt/us
I see the item 'KOReader' in KUAL and when I click 'Start the filemanager' it crashed and go Home directly.

I tried run koreader from terminal to open a html file with below commands.
#cd /mnt/us/koreader
#./koreader.sh ../documents/list.html -d

The crash.log
./luajit: frontend/device/generic/powerd.lua:30: bad argument #1 to 'open' (string expected, got nil)
stack traceback:
[C]: in function 'open'
frontend/device/generic/powerd.lua:30: in function 'read_int_file'
frontend/device/kindle/powerd.lua:21: in function 'init'
frontend/device/generic/powerd.lua:16: in function 'new'
frontend/device/kindle/device.lua:173: in function 'init'
frontend/device.lua:38: in main chunk
[C]: in function 'require'
frontend/ui/uimanager.lua:1: in main chunk
[C]: in function 'require'
./reader.lua:91: in main chunk
[C]: at 0x0000be95

It seems the power lib is not compatible with 5.6.0.1.6.
13 function KindlePowerD:init()
14 local lipc = require("liblipclua")
15 if lipc then
16 self.lipc_handle = lipc.init("com.github.koreader.kindlepowerd")
17 end
18 if self.lipc_handle then
19 self.flIntensity = self.lipc_handle:get_int_property("com.lab126.powe rd", "flIntensity")
20 else
21 self.flIntensity = self:read_int_file(self.fl_intensity_file) --This should be empty
22 end
23 end

How can I run koreader properly?...

Another issue I found on the script.
For kindle 7.
It can display charactors 37 columns x 33 lines. #epis 36 32 a
EIPS_MAXCHARS=37
EIPS_MAXLINES=33
That's why jb-1.14.N not show the text properly.


Update:
With below command, koreader worker perfect!!
#/mnt/us/koreader/koreader.sh /mnt/us/documents

unpack *.bin files:
I tried kindletool https://github.com/yifanlu/KindleTool, it can decode *.bin to *.tar.gz.

[zfu]

BTW, Most of the resource are with prefix *.bin, how can I decode/unzip this file install manually?

[NiLuJe]

Possibly because no-one has ever run KOreader on a KT2 .

I'll take a potshot at it, expect me to ask you to run stuff to confirm it .

As for the EIPS thing, nice catch, thanks. Additional Q., though: does one character block still occupies the same space?

[knc1]

Originally Posted by zfu

BTW, Most of the resource are with prefix *.bin, how can I decode/unzip this file install manually?

You don't.
You use the MR Package Installer to install packages on 5.6.x firmware.
You use Kindletool to take apart and put back together *.bin format packages (ours or Amazons).
(Dah, we have been writing about and supporting 5.6.x series for many months here. - Spend a little time reading about it. )

[NiLuJe]

I pushed something in a fix-kt2 branch here, but I'm slightly concerned at the fact that the lipc init would fail...

What's in /usr/lib/lua on your device?

What does lipc-probe -av outputs?

Also, since that eips thing is weird, what's the output of eips -i?

[zfu]

Update:
I tried
#/mnt/us/koreader/koreader.sh /mnt/us/documents
koreader works perfectly!!
usbnet is also works good.
There may something wrong with KUAL..
MR Package Installer will crash with KUAL..

[zfu]

lua is existed at /usr/bin/lua.

I attached the output of lipc-probe -av.
lipc-probe.av.log.txt

[NiLuJe]

That vaguely sounds like KUAL is not running as root...

Which version of KUAL are you using, and what's the breadcrumb line prefixed with (top-left)?
I'd reboot once just to make sure things are shaken into place.

I'm also going to need way more details that just 'crash' ^^.

To be perfectly honest, I'm kind of surprised KUAL runs *at all*, I thought Amazon killed Kindlets entirely on the KT2 & KV...

----

As for the EIPS thing, to fix it in a better way than just magic numbers, I'm going to need the output of eips -i and possibly a screenshot where one block of eips character can properly be identified to measure it.

----

Sidenote: you'll want my fork of KindleTool to deal with recent devices (and actually work on other platforms than OSX/BSD).

[zfu]

Yes, it like KUAL not running as root, and the script not exec properly.

eips -i output:

Fixed framebuffer info
id: mxc_epdc_fb smem_start: 0x80C00000
smem_len: 3268608 type: PACKED_PIXELS
type_aux: 0 visual: STATIC_PSEUDOCOLOR
xpanstep: 1 ypanstep: 1
ywrapstep: 0 line_length: 608
mmio_start: 0x00000000 mmio_len: 0
accel: 0

Variable framebuffer info
xres: 600 yres: 800
xres_virtual: 608 yres_virtual: 5376
xoffset: 0 yoffset: 0
bits_per_pixel: 8 grayscale: 1
red.offset: 0 green.offset: 0
red.length: 8 green.length: 8
red.msb_right: 0 green.msb_right: 0
blue.offset: 0 transp.offset: 0
blue.length: 8 transp.length: 0
blue.msb_right: 0 transp.msb_right: 0
nonstd: 0 activate: 128
width: 91 height: 122
accel_flags: 0 pixclock: 27027027
left_margin: 8 right_margin: 100
upper_margin: 4 lower_margin: 8
hsync_len: 4 vsync_len: 1
sync: 0 vmode: 0
rotate: 3

[zfu]

oh, I tried your kual version 2.5, it work perfect!
Thanks

[NiLuJe]

Ha, that makes much more sense, yay .
KUAL running successfully on a KT2 is a nice surprise .

As for eips, try to run this

#!/bin/sh
##
#
#  eips check
#
#  $Id$
#
##

# We need to get the proper constants for our model...
kmodel="$(cut -c3-4 /proc/usid)"
case "${kmodel}" in
	"13" | "54" | "2A" | "4F" | "52" | "53" )
		# Voyage...
		SCREEN_X_RES=1088	# NOTE: Yes, 1088, not 1072 or 1080...
		SCREEN_Y_RES=1448
		EIPS_X_RES=16
		EIPS_Y_RES=24		# Manually mesured, should be accurate.
	;;
	"24" | "1B" | "1D" | "1F" | "1C" | "20" | "D4" | "5A" | "D5" | "D6" | "D7" | "D8" | "F2" | "17" | "60" | "F4" | "F9" | "62" | "61" | "5F" )
		# PaperWhite...
		SCREEN_X_RES=768	# NOTE: Yes, 768, not 758...
		SCREEN_Y_RES=1024
		EIPS_X_RES=16
		EIPS_Y_RES=24		# Manually mesured, should be accurate.
	;;
	"C6" | "DD" )
		# KT2...
		SCREEN_X_RES=600	# NOTE: Might be 608?
		SCREEN_Y_RES=800
		EIPS_X_RES=16
		EIPS_Y_RES=24
	;;
	* )
		# Handle legacy devices...
		if [ -f "/etc/rc.d/functions" ] && grep "EIPS" "/etc/rc.d/functions" > /dev/null 2>&1 ; then
			. /etc/rc.d/functions
		else
			# Touch / Basic
			SCREEN_X_RES=600	# _v_width @ upstart/functions
			SCREEN_Y_RES=800	# _v_height @ upstart/functions
			EIPS_X_RES=12		# from f_puts @ upstart/functions
			EIPS_Y_RES=20		# from f_puts @ upstart/functions
		fi
	;;
esac
# And now we can do the maths ;)
EIPS_MAXCHARS="$((${SCREEN_X_RES} / ${EIPS_X_RES}))"
EIPS_MAXLINES="$((${SCREEN_Y_RES} / ${EIPS_Y_RES}))"

# Adapted from libkh[5]
eips_print_bottom_centered()
{
	# We need at least two args
	if [ $# -lt 2 ] ; then
		echo "not enough arguments passed to eips_print_bottom_centered ($# while we need at least 2)"
		return
	fi

	kh_eips_string="${1}"
	kh_eips_y_shift_up="${2}"

	# Get the real string length now
	kh_eips_strlen="${#kh_eips_string}"

	# Add the right amount of left & right padding, since we're centered, and eips doesn't trigger a full refresh,
	# so we'll have to padd our string with blank spaces to make sure two consecutive messages don't run into each other
	kh_padlen="$(((${EIPS_MAXCHARS} - ${kh_eips_strlen}) / 2))"

	# Left padding...
	while [ ${#kh_eips_string} -lt $((${kh_eips_strlen} + ${kh_padlen})) ] ; do
		kh_eips_string=" ${kh_eips_string}"
	done

	# Right padding (crop to the edge of the screen)
	while [ ${#kh_eips_string} -lt ${EIPS_MAXCHARS} ] ; do
		kh_eips_string="${kh_eips_string} "
	done

	# Sleep a tiny bit to workaround the logic in the 'new' (K4+) eInk controllers that tries to bundle updates,
	# otherwise it may drop part of our messages because of other screen updates from KUAL...
	usleep 150000	# 150ms

	# And finally, show our formatted message centered on the bottom of the screen (NOTE: Redirect to /dev/null to kill unavailable character & pixel not in range warning messages)
	eips 0 $((${EIPS_MAXLINES} - 2 - ${kh_eips_y_shift_up})) "${kh_eips_string}" >/dev/null
}

##
eips_print_bottom_centered "Is the white edge to edge" 1

And see if the white background extends to the right edge of the screen. If it does, great, if it doesn't, see if tweaking SCREEN_X_RES to 608 for the KT2 case fixes it.

Thanks .

[Dr. Drib]

Moderator Notice

Please take a few moments to read our Posting Guidelines, in particular the part about the maximum image size allowed (600x600 pixels). Images that violate those guidelines will be deleted.

Thanks.

[zfu]

It shows right at the center of bottom.
'Is the white edge to edge'
EIPS_X_RES=16
EIPS_Y_RES=24

[NiLuJe]

@zfu: Great, thanks .

And, just to confirm: is the white background hitting the right edge, or is there a sliver left? (probably need to do this over a full-screen cover or something to catch that).

I'll fix my stuff, that'll end up in the next snapshots whenever they're released .

[knc1]

Originally Posted by NiLuJe

@zfu: Great, thanks .

And, just to confirm: is the white background hitting the right edge, or is there a sliver left? (probably need to do this over a full-screen cover or something to catch that).

I'll fix my stuff, that'll end up in the next snapshots whenever they're released .

What is this?
With unpaid, volunteers, responding to bug reports the same day?


- - - -

Meanwhile, MiUi (current Doukan owner - I think) - who's slogan is "Redefining Android" seems to be having problems with its own (Chinese) user community:
https://translate.googleusercontent....ZBE8W1fWc0e3tQ

(What? No source available? - Phooey, that sounds true to form.
I guess the Chinese Kindle owners will have to switch to our KOReader.)