So I guess it is exactly the same. Proof:
Code:
patrick@dpc01:~/kp/12019$ arm-linux-gnueabihf-objdump -S libnickel.so.1.0.0 > libnickel-12019.dump
patrick@dpc01:~/kp/12019$ arm-linux-gnueabihf-objdump -S ../182/libnickel.so.1.0.0 > libnickel-182.dump
patrick@dpc01:~/kp/12019$ diff *.dump
2c2
< libnickel.so.1.0.0: file format elf32-littlearm
---
> ../182/libnickel.so.1.0.0: file format elf32-littlearm
patrick@dpc01:~/kp/12019$ strings libnickel.so.1.0.0 | grep 4.11.12019 -C4
/mnt/onboard/.kobo
3.12.1
Mozilla/5.0 (Linux; U; Android 2.0; en-us;) AppleWebKit/%1 (KHTML, like Gecko) Version/4.0 Mobile Safari/%1 (Kobo Touch %2/%3)
Mobile
4.11.12019
^(gif|fw|p2p|utun)\d+$
Bluetooth
[a-z0-9:]+
00000000000000000000000000000000
patrick@dpc01:~/kp/12019$ strings ../182/libnickel.so.1.0.0 | grep 4.11.11982 -C4
/mnt/onboard/.kobo
3.12.1
Mozilla/5.0 (Linux; U; Android 2.0; en-us;) AppleWebKit/%1 (KHTML, like Gecko) Version/4.0 Mobile Safari/%1 (Kobo Touch %2/%3)
Mobile
4.11.11982
^(gif|fw|p2p|utun)\d+$
Bluetooth
[a-z0-9:]+
00000000000000000000000000000000
patrick@dpc01:~/kp/12019$ sha256sum libnickel.so.1.0.0 ../182/libnickel.so.1.0.0
8b53760cf51f9bc1e434e03d7114ba00fb378e752619a063c6e965a5d574c241 libnickel.so.1.0.0
b162d95e7788b21383448366a0d1208cd84bcaeb6aa791c69c3f14895dfb5121 ../182/libnickel.so.1.0.0
patrick@dpc01:~/kp/12019$ ls -lah
total 464M
drwxr-xr-x 3 patrick patrick 4.0K Nov 26 20:55 .
drwxr-xr-x 12 patrick patrick 4.0K Nov 26 20:07 ..
drwxr-xr-x 2 patrick patrick 4.0K Nov 26 20:08 4.11.12019
-rwxrwxr-x 1 patrick patrick 269K Nov 19 19:44 libadobe.so
-rw-r--r-- 1 patrick patrick 171M Nov 26 20:55 libnickel-12019.dump
-rw-r--r-- 1 patrick patrick 95M Nov 26 20:38 libnickel-12019.hop
-rw-r--r-- 1 patrick patrick 171M Nov 26 20:56 libnickel-182.dump
-rwxrwxr-x 1 patrick patrick 16M Nov 19 19:44 libnickel.so.1.0.0
-rwxrwxr-x 1 patrick patrick 7.5M Nov 19 19:44 librmsdk.so.1.0.0
-rwxrwxr-x 1 patrick patrick 5.1M Nov 19 19:44 nickel
-rwxrwxr-x 1 patrick patrick 27K Nov 19 19:44 sickel
Can someone
please prove me wrong and tell me I missed something. Or that kobo did something funny (@davidfor?)? Something about this doesn't make sense...the build increased by 37, yet nothing changed about libnickel except the hash...?
If nobody finds anything, I'll release the (exact same!?!) patches tomorrow.
Edit: and what's more, the nickel CSS is identical too! The sums are different, but everything (the order, the contents) are exactly the same. This is really, really confusing. I hope and don't hope I've been barking up the wrong tree...
Edit 2: and guess what:
Code:
patrick@dpc01:~/kp/12019$ diff <(xxd ../182/libnickel.so.1.0.0) <(xxd libnickel.so.1.0.0)
17,18c17,18
< 00000100: 474e 5500 97ad a007 cfac 09e4 e492 c37c GNU............|
< 00000110: 288c 2f25 1bed 33a0 f32f 0000 26bc 0000 (./%..3../..&...
---
> 00000100: 474e 5500 2841 af98 d989 b8b4 5565 cd7f GNU.(A......Ue..
> 00000110: ce24 e2cc 968c 787c f32f 0000 26bc 0000 .$....x|./..&...
842127c842127
< 00cd98e0: 2e31 3139 3832 0000 4e41 0000 5e28 6769 .11982..NA..^(gi
---
> 00cd98e0: 2e31 3230 3139 0000 4e41 0000 5e28 6769 .12019..NA..^(gi
843092c843092
< 00cdd530: 4e6f 7620 2036 2032 3031 3800 4d4d 4d20 Nov 6 2018.MMM
---
> 00cdd530: 4e6f 7620 3139 2032 3031 3800 4d4d 4d20 Nov 19 2018.MMM
Did they just use a hex editor to change the version?!?
. Can @NiLuJe, @davidfor, @jackie_w, and @GeoffR take a look and see?
Edit 3: So I did a diff of the whole KoboRoot, and only the version numbers changed. In nickel, there are a few more changes, but they are inconsequential (the change is due to the zlib compression, and the decompressed version is identical). This is really crazy.
Edit 4: It seems like one of the kernel images (zImage-E80K00) differs slightly.
Edit 5: DNSB just
confirmed that it was a hardware fix, which explains only the kernel differing.