08-05-2013, 12:41 PM | #1 |
Going Viral
Posts: 17,212
Karma: 18210809
Join Date: Feb 2012
Location: Central Texas
Device: No K1, PW2, KV, KOA
|
[KUALette] Hardware Report
This is a corrected and extended function button for reporting the K5 hardware environment.
Tested on Kpw, should also work on Kindle Touch but not tested. Un-archiving the zip file will create a directory 'HardWare' with contents and a sub-tree (same as other KUAL buttons). Copy the 'HardWare' folder, and its sub-tree into your */extensions directory for KUAL (same as other KUAL buttons). If you should find a file named: '1' in */extensions/HardWare/bin, that file can be safely deleted. It is the result of the typos in version-1.0. Titled as a "KUALette" because there is no 'upstream source' - its all scripting, what you see in the files is all there is (plus the lab126 functions it calls). Code:
Hardware Version Report Board: celeste Revision: 22 Platform: yoshime3 Battery Status Report Battery ID: 10 Battery capacity: 99% Charging voltage: 4190 mv. Charging current: 112 ma. Current charge: 1376 maH. Last measured discharge: 1386 ma. Last edited by knc1; 08-05-2013 at 02:25 PM. |
08-05-2013, 02:04 PM | #2 |
( ͡° ͜ʖ ͡°){ʇlnɐɟ ƃǝs}Týr
Posts: 6,586
Karma: 6299991
Join Date: Jun 2012
Location: uti gratia usura (Yao ying da ying; Mo ying da yieng)
Device: PW-WIFI|K5-3G+WIFI| K4|K3-3G|DXG|K2| Rooted Nook Touch
|
Awesome.
KUALette is "Kual-as" |
08-06-2013, 11:13 AM | #3 |
Going Viral
Posts: 17,212
Karma: 18210809
Join Date: Feb 2012
Location: Central Texas
Device: No K1, PW2, KV, KOA
|
Kernel memory corruption
This seems to be occurring more often than it should.
Well, in point of fact, it should **never** happen but in the Kindle kernels it is a known problem (according to lab126's own comments in the log writer functions). Today's challenge - Script a corruption detector for this hardware reporter. This problem was fixed back in the days of the version 2.5.x kernel, but it seems to have gotten "un-fixed" in the Amazon kernel build. |
08-06-2013, 12:46 PM | #4 |
BLAM!
Posts: 13,478
Karma: 26012494
Join Date: Jun 2010
Location: Paris, France
Device: Kindle 2i, 3g, 4, 5w, PW, PW2, PW5; Kobo H2O, Forma, Elipsa, Sage, C2E
|
@knc1: Huh. Incidentally, the 2.5 -> 3.x 'update' was when they switched from a vanilla/freescale kernel to vanilla+rt/freescale kernel (and all hell broke lose for the full lifecycle of the 3.0.x updates, until they finally stopped trying to use an rt preempt kernel with a host of probably not that pretty 3rd party modules).
I haven't seen a PREEMPT kernel build since, although they still use a kernel at least based on the -RT patchset. |
08-06-2013, 12:53 PM | #5 | |
Going Viral
Posts: 17,212
Karma: 18210809
Join Date: Feb 2012
Location: Central Texas
Device: No K1, PW2, KV, KOA
|
Quote:
**Kernel** version 2.5.x not Amazon firmware version 2.5.y, sorry, again. That was a loooong time ago as time passes on kernel.org and a ps: In the -RT kernels from Amazon that still have the .config info compiled in - lab126 has the RT stuff disabled in the config. Duh... No telling what that did, patching the kernel and then "disabling" the new features. I bet it left "cruff" all over the source code. - - - - Reaches for the blood pressure medication each time lab126 comes into the discussion . . . . |
|
08-06-2013, 12:56 PM | #6 |
BLAM!
Posts: 13,478
Karma: 26012494
Join Date: Jun 2010
Location: Paris, France
Device: Kindle 2i, 3g, 4, 5w, PW, PW2, PW5; Kobo H2O, Forma, Elipsa, Sage, C2E
|
@knc1: Yeah, they did disable most of the RT preempt stuff in the end (3.1), but the -RT kernel touches a lot of things, and I don't think everything's 'gone'. I see the softirq kernel threads, for example, which IIRC were (still are? I though it was merged at one point :?) an -RT thing.
Last edited by NiLuJe; 08-06-2013 at 12:59 PM. |
08-06-2013, 01:38 PM | #7 |
Going Viral
Posts: 17,212
Karma: 18210809
Join Date: Feb 2012
Location: Central Texas
Device: No K1, PW2, KV, KOA
|
@NiLuJe: I am agreeing with you, that is probably not the most solid kernel build Amazon ever shipped.
|
08-06-2013, 01:40 PM | #8 | |
Going Viral
Posts: 17,212
Karma: 18210809
Join Date: Feb 2012
Location: Central Texas
Device: No K1, PW2, KV, KOA
|
Quote:
The date/time stamps on kernel events have been turned off in 5.3.3 and 5.3.4 (at least, maybe others). |
|
08-06-2013, 01:48 PM | #9 |
Going Viral
Posts: 17,212
Karma: 18210809
Join Date: Feb 2012
Location: Central Texas
Device: No K1, PW2, KV, KOA
|
Process listing
@NiLuJe: How many times does USBnet-0.9.N run dropbear?
Code:
root 2853 0.0 0.1 1672 404 ? SNs 16:23 0:00 /mnt/us/usbnet/bin/busybox telnetd -F root 2860 0.0 0.1 1964 508 ? SNs 16:23 0:00 /usr/bin/dropbear -P /mnt/us/usbnet/run/sshd.pid -n -b /dev/null root 4083 0.2 0.3 2032 872 ? SNs 17:35 0:00 /usr/bin/dropbear -P /mnt/us/usbnet/run/sshd.pid -n -b /dev/null Code:
[root@kindle kernel]# who -a LOGIN ttymxc0 01:52 Aug 6 15:57:38 root pts/0 00:00 Aug 6 16:11:14 192.168.15.201:37316 root pts/0 00:00 Aug 6 17:35:16 192.168.15.201 Last edited by knc1; 08-06-2013 at 01:50 PM. |
08-06-2013, 02:15 PM | #10 | |
Going Viral
Posts: 17,212
Karma: 18210809
Join Date: Feb 2012
Location: Central Texas
Device: No K1, PW2, KV, KOA
|
Quote:
FLAG=$(cat /sys/module/printk/parameters/time) if FLAG == 'N' then dmesg -c echo 'Y' > /sys/module/printk/parameters/time fi Or, something like that to turn on the kernel message time stamps. Code:
[root@kindle parameters]# echo 'Y' > time [root@kindle parameters]# dmesg [ 101.048759] KERNEL: I pmic: PWRON1 event: [ 101.246992] cyttsp: I cyttsp_proc_write::command=unlock: |
|
08-06-2013, 02:25 PM | #11 |
BLAM!
Posts: 13,478
Karma: 26012494
Join Date: Jun 2010
Location: Paris, France
Device: Kindle 2i, 3g, 4, 5w, PW, PW2, PW5; Kobo H2O, Forma, Elipsa, Sage, C2E
|
@knc1: Yep, one master process, and a fork for each connection. OpenSSH does the same, except the child processes have a more verbose name .
Code:
root 3387 0.0 0.2 3180 756 ? Ss 20:00 0:00 /usr/sbin/sshd root 5602 2.6 0.6 5892 1748 ? Ss 20:25 0:00 \_ sshd: root@pts/0 root 5606 0.0 0.2 2432 624 pts/0 Ss 20:25 0:00 \_ -sh root 5609 0.0 0.3 2604 956 pts/0 R+ 20:26 0:00 \_ ps faux Code:
source /etc/rc.d/functions && echo "$(productid) @ rev $(hwrevision) ($(hwbuildid) build)" Last edited by NiLuJe; 08-06-2013 at 02:34 PM. |
08-07-2013, 10:58 AM | #12 |
Going Viral
Posts: 17,212
Karma: 18210809
Join Date: Feb 2012
Location: Central Texas
Device: No K1, PW2, KV, KOA
|
Kernel memory corruption
Maybe, maybe not . . . .
There is a definite pattern here: Code:
[root@kindle root]# dmesg [ 101.048759] KERNEL: I pmic: PWRON1 event: [ 101.246992] cyttsp: I cyttsp_proc_write::command=unlock: [ 164.429048] cyttsp: I cyttsp_proc_write::command=lock: [ 69.146976] KERNEL: I aplite_charger: Restart charging: [ 345.405187] KERNEL: I aplite_charger: Battery full Stop charging: [ 184.179650] KERNEL: I aplite_charger: Restart charging: [ 460.437894] KERNEL: I aplite_charger: Battery full Stop charging: [ 269.212394] KERNEL: I aplite_charger: Restart charging: [ 8.599683] KERNEL: I aplite_charger: Battery full Stop charging: [ 354.245148] KERNEL: I aplite_charger: Restart charging: [ 63.632433] KERNEL: I aplite_charger: Battery full Stop charging: [ 439.277848] KERNEL: I aplite_charger: Restart charging: [ 148.665061] KERNEL: I aplite_charger: Battery full Stop charging: [ 524.350571] KERNEL: I aplite_charger: Restart charging: [ 233.737835] KERNEL: I aplite_charger: Battery full Stop charging: [ 42.512368] KERNEL: I aplite_charger: Restart charging: [ 318.770530] KERNEL: I aplite_charger: Battery full Stop charging: [ 127.545076] KERNEL: I aplite_charger: Restart charging: [ 245.636135] aplite_charger: I pmic:faulti:sense_0=0x200:battery dies/charger times out. [ 339.794269] PM: Syncing filesystems ... done. [ 339.862853] Freezing user space processes ... (elapsed 0.00 seconds) done. [ 339.866307] Freezing remaining freezable tasks ... (elapsed 0.00 seconds) done. [ 339.867194] Suspending console(s) (use no_console_suspend to debug) [ 339.868346] cyttsp_suspend: Enter [ 339.888562] WiFi chip powered down [ 339.889415] PM: suspend of devices complete after 21.077 msecs [ 339.889478] kernel: I PM:suspend:duration=21.077: [ 339.890472] PM: late suspend of devices complete after 0.925 msecs [ 339.890530] kernel: I PM:latesuspend:duration=0.925: [ 339.935129] PM: early resume of devices complete after 0.270 msecs [ 339.935206] kernel: I PM:earlyresume:duration=0.270: [ 339.988619] WiFi chip powered on [ 340.123421] mmc0: queuing CIS tuple 0x01 length 3 [ 340.130823] mmc0: queuing CIS tuple 0x1a length 5 [ 340.134048] mmc0: queuing CIS tuple 0x1b length 8 [ 340.134704] mmc0: queuing CIS tuple 0x14 length 0 [ 340.296101] PM: resume of devices complete after 360.528 msecs [ 340.296117] kernel: I PM:resume:duration=360.528: [ 340.296276] Restarting tasks ... done. [ 340.701503] g_ether gadget: high speed config #1: CDC Ethernet (ECM) [ 342.290615] Battery: Resume - TimeSuspended(secs)=41841, NAC=0x1d3c, RSOC=96, capacity=96% [ 342.369272] Battery: Resume - PostUpdate NAC=0x1d08, RSOC=96 The messages sent into the kernel's buffer by the lab126 scripting certainly looks like it has problems. Almost as if they where using random numbers for the time stamps. Groan - I may have to set this little "corruption detection" project aside until I have enough time to read through whatever it is lab126 is doing with its message logging. |
08-07-2013, 11:56 AM | #13 |
BLAM!
Posts: 13,478
Karma: 26012494
Join Date: Jun 2010
Location: Paris, France
Device: Kindle 2i, 3g, 4, 5w, PW, PW2, PW5; Kobo H2O, Forma, Elipsa, Sage, C2E
|
@knc1: No idea about the native/java stuff, but what's in shell uses logger from busybox, which, AFAIK is a simple wrapper around the syslog() stuff.
(Well, at least reimplementing it in Python that way achieved the same kind of results). |
Tags |
launcher add-ons |
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
Classic New Hardware | JeffElkins | Barnes & Noble NOOK | 5 | 12-06-2010 07:55 AM |
Free Report (Kindle) - Economic Report of the President | koland | Deals and Resources (No Self-Promotion or Affiliate Links) | 5 | 02-13-2010 12:07 PM |
Any New Hardware for the UK? | river | Which one should I buy? | 8 | 06-24-2009 04:16 PM |
Ergonomics for Hardware? | emellaich | News | 14 | 06-03-2009 05:30 PM |
Hardware vs. Software | weatherman | Amazon Kindle | 7 | 04-18-2009 05:22 AM |