Database Users

Linux on the Wizard

Hi.
Has anyone tried to make linux/familiar boot on the Wizard yet?
Any luck?

mind said:
Hi.
Has anyone tried to make linux/familiar boot on the Wizard yet?
Any luck?
Click to expand...
Click to collapse
Oh this looks interesting. What could you do with a linux boot?

mind said:
Has anyone tried to make linux/familiar boot on the Wizard yet?
Click to expand...
Click to collapse
Have you tried to run haret ?
If it works, can you post the
framebuffer address and 'dump mmu' here ?

It might take a bit longer for anything like this to appear, but it would be interesting when it does.

I haven't tried it. The reason for my question was that I'm going to do some development work for Opie (opie.handhelds.org), and need a machine that boots and runs familiar to test stuff. And I would like to get me a wizard
Familiar has a list of machines tested on familiar ( booting or not ). But only Wallaby (XDA), Himalaya (XDA II), Blueangel (XDA III) and Universal (MDA Pro) are listed.
So if anyone in this forum wich has a wizard and would like to try out linux/opie on it; please let me know if it boots!!!
There are a lot of advantages having linux on you're XDA instead of windows mobile. The most obvious being that if you're stuck in some corner ( lets say mp3 audio over bt isn't working ), you could always do some work on the ( BlueZ ) code you're self to try to fix the problem. Also there's alot of cool software available that you don't have on WM.
And oh, you're free
Best regards,

mind said:
Familiar has a list of machines tested on familiar ( booting or not ). But only Wallaby (XDA), Himalaya (XDA II), Blueangel (XDA III) and Universal (MDA Pro) are listed.
Click to expand...
Click to collapse
The only machine where Linux WORKS right now
is Blueangel (with GPE, never seen reports about opie). Himalaya, Universal and Magician need some work. Alpine should work with some
effort too.
Wallaby is not really working (no SD) because
nobody is interested in writing this driver.
Power management (i.e. suspend) is missing now.
So if anyone in this forum wich has a wizard and would like to try out linux/opie on it; please let me know if it boots!!!
Click to expand...
Click to collapse
To run opie, you need a running kernel, which does not exist. To write the kernel it is necessary to run 'haret'. So, the first step
is to try 'haret' !

To run opie, you need a running kernel, which does not exist. To write the kernel it is necessary to run 'haret'. So, the first step
is to try 'haret' !
Click to expand...
Click to collapse
Just tried it. Haret starts, but prints out the message "Exception reading coprocessor 15 register .." I can connect to Haret on port 9999 and can issue some commands like "dump mmu" but there's not much info there...:
Code:
descriptor flags legend:
C: Cacheable
B: Bufferable
0..3: Access Permissions (for up to 4 slices):
0: Supervisor mode Read
1: Supervisor mode Read/Write
2: User mode Read
3: User mode Read/Write
Error: EXCEPTION reading coprocessor 15 register 2
MMU 1st level descriptor table is at FFFFC000
Virtual | Physical | Descr | Description
address | address | flags |
----------+----------+---------+-----------------------------
00000000 | | | UNMAPPED
ffffffff | | | End of virtual address sp

richardt007 said:
Haret starts, but prints out the message "Exception reading coprocessor 15 register .." I can connect to Haret on port 9999 and can issue some commands like "dump mmu" but there's not much info there...:
Click to expand...
Click to collapse
Then you need to sign haret with the "priviledged certificate". Read more here:
http://forum.xda-developers.com/viewtopic.php?t=30399&start=0

Can you tell me how to do that in a little more detail than what's described in the thread? I managed to compile the latest haret.exe using wine and the msvc package. Do I need to alter the asm-files (and which one?) with the code-snippet in that thread or is it a little more complicated than that? Would like to contribute though...

richardt007 said:
Can you tell me how to do that in a little more detail than what's described in the thread?
Click to expand...
Click to collapse
No, it seems that you don't need to change anything, only to sign the executable.
Don't ask me how, i don't know exactly
I've tried makecert.exe with wine-0.9.1 and
it works. You should look into the archives
on this site or read the microsoft documentation.

cr2 said:
Don't ask me how, i don't know exactly
Click to expand...
Click to collapse
There is an explanation now at the end of this
thread.

I managed to get some output from haret, although don't know the validity of the output.
1. I had to modify the asmstuff.asm to include the cp_0, cp_2 and cp_13 calls in order to get rid of the exception ( GetLastError returned 120=ERROR_CALL_NOT_IMPLEMENTED)
2. Called the cp_2 from _cpu_get_cp.
3. Signed the executable using signcode.exe from the Smartphone 2003 SDK (have no clue if this was needed at all)
Here is some output from the output, can somebody help out to interpret it?
Code:
HaRET(3)# dump gpio
GPIO# D S A INTER | GPIO# D S A INTER | GPIO# D S A INTER | GPIO# D S A INTER
------------------+-------------------+-------------------+------------------
0 O 1 0 | 21 I 0 0 | 42 O 0 0 | 63 O 0 2
1 O 0 0 | 22 I 0 0 | 43 I 0 0 | 64 I 0 0
2 I 1 0 | 23 I 0 0 | 44 I 0 0 | 65 I 0 2 FE
3 I 0 0 | 24 I 0 0 | 45 I 0 0 | 66 I 0 3
4 I 0 0 | 25 I 0 0 | 46 I 0 0 | 67 I 0 3
5 I 0 0 | 26 I 0 0 | 47 O 0 0 | 68 I 0 3
6 I 0 0 | 27 I 0 0 | 48 I 0 0 | 69 I 0 1
7 I 0 0 | 28 I 0 0 | 49 I 0 2 | 70 I 0 0
8 I 0 0 | 29 I 0 0 | 50 I 0 3 | 71 I 0 2
9 I 0 0 | 30 I 0 0 | 51 O 0 3 | 72 I 0 0
10 I 0 0 | 31 I 0 0 | 52 O 0 3 | 73 I 0 2
11 I 0 0 | 32 I 0 0 FE | 53 O 0 1 | 74 I 0 3
12 I 0 0 | 33 I 1 0 | 54 O 0 0 | 75 I 0 3
13 I 0 0 | 34 I 0 0 FE | 55 O 0 2 | 76 I 0 3
14 I 0 0 | 35 O 0 0 | 56 O 0 0 | 77 I 0 1
15 I 0 0 | 36 O 0 0 | 57 O 0 2 | 78 I 0 0
16 I 0 3 | 37 O 0 0 | 58 O 0 3 | 79 I 0 2
17 I 0 3 | 38 O 0 0 | 59 I 0 3 | 80 I 0 0
18 I 0 3 | 39 O 0 0 | 60 I 0 3 | 81 I 0 2
19 I 0 1 | 40 O 0 0 | 61 I 0 1 | 82 I 0 3
20 I 0 0 | 41 O 0 0 | 62 I 0 0 | 83 I 0 3
Code:
HaRET(2)# print "%x" VRAM
20001020
Dump mmu gives a lot of lines, have no idea what the most relevant part of it is, so here a snippet:
Code:
HaRET(7)# dump mmu
----- Virtual address map -----
Descriptor flags legend:
C: Cacheable
B: Bufferable
0..3: Access Permissions (for up to 4 slices):
0: Supervisor mode Read
1: Supervisor mode Read/Write
2: User mode Read
3: User mode Read/Write
MMU 1st level descriptor table is at 10360000
Virtual | Physical | Descr | Description
address | address | flags |
----------+----------+---------+-----------------------------
00000000 | | | UNMAPPED
02000000 | | | UNMAPPED
02101000 | 10463000 | CB0000 | Small page (4K)
...
...
29fff000 | 12b4b000 | CB3 | Small page (4K)
2a000000 | | | UNMAPPED
Then it stops, with the progress bar somewhere at 40%. had to give it a soft reset. So what's the next step...?

richardt007 said:
I managed to get some output from haret, although don't know the validity of the output.
Here is some output from the output, can somebody help out to interpret it?
Click to expand...
Click to collapse
I have looked in the docs, wizard has TI OMAP processor, and haret 'gpio' knows only about
PXA. So , this table is probably useless.
Code:
HaRET(2)# print "%x" VRAM
20001020
From omap730.h:
#define OMAP730_SRAM_START 0x20000000
So your VRAM is in the OMAP built-in SRAM.
Code:
HaRET(7)# dump mmu
MMU 1st level descriptor table is at 10360000
Your RAM starts at 0x10000000.
At least 4MB for the kernel & stuff.
Code:
Virtual | Physical | Descr | Description
address | address | flags |
----------+----------+---------+-----------------------------
02101000 | 10463000 | CB0000 | Small page (4K)
The page in RAM mapped for some system purposes.
Then it stops, with the progress bar somewhere at 40%. had to give it a soft reset.
Click to expand...
Click to collapse
It's normal, "dump mmu" in haret is buggy
So what's the next step...?
Click to expand...
Click to collapse
It is interesting to find out where is your ROM mapped.
You can also sign the itsutils and try to dump the memory with these tools.
Read the linux-omap documentation and code
and dump the OMAP registers. You can also
try to find out from the wince dlls what other
chips are used in this phone.

You can also sign the itsutils and try to dump the memory with these tools.
Click to expand...
Click to collapse
Signing itsutils was not enough unfortunately, as I had to change the security policy (policy ID=4097) from 1 to 2 to allow RAPI calls.
(see http://msdn.microsoft.com/library/default.asp?url=/library/en-us/dnppcgen/html/wmdeploy.asp)
After that, itsutils ran like a charm:
Code:
c:\src\itsutils\build>pmemmap
v82000000-83000000 -> p00000000-01000000
v8a000000-8a400000 -> p13c00000-14000000
v8c000000-8fc00000 -> p10000000-13c00000
v90000000-90100000 -> p20000000-20100000
v98000000-98100000 -> pfff00000-00000000 (huh?)
v98100000-98200000 -> p04000000-04100000
v98400000-98500000 -> p06000000-06100000
v98500000-98600000 -> p40000000-40100000
A simple process list, 'pps -t' showed _alot_ of .dll's being used (mainly by device.exe, naturally)
Code:
c:\src\itsutils\build>pps
handle n base kern user heap exe
0ddf5fba 5 20000000 0.0 8.2 0 rapiclnt -drndis_peer
0dfb5706 2 0e000000 0.0 0.0 0 elatepushservice.exe
0dfdf8ce 11 1a000000 0.0 0.5 0 repllog.exe /sched /remote /h
0e0f03ba 2 10000000 0.0 0.0 0 sddaemon.exe
0fbb4002 2 c2000000 0.0 0.0 0 NK.EXE
2e0a2402 8 14000000 0.0 0.0 0 srvtrust.exe 80
4de65d06 10 18000000 0.0 0.0 0 cprog.exe -n
4e3767c2 10 08000000 0.0 2.1 0 shell32.exe 50
6fba4bfa 12 04000000 0.0 37.8 0 filesys.exe
8dfb56e2 2 1c000000 0.0 0.0 0 PowerTray.exe
8e91ce1a 107 06000000 0.0 1.8 0 device.exe 20
adf78e9e 1 1e000000 0.0 0.0 0 WiFiTray.exe
ae0b6dc2 11 12000000 0.0 0.0 0 connmgr.exe 70
ae54049a 15 0a000000 0.0 0.0 0 gwes.exe 30
ce376c4a 20 0c000000 0.0 0.0 0 services.exe 60
edc4dd32 9 22000000 0.0 0.0 0 tmail.exe -RunInBKG
ee040e02 1 16000000 0.0 0.0 0 poutlook.exe
0 ........ 0.0 50.4 0 total
The list of partitions is completely unclear to me:
Code:
c:\src\itsutils\build>pdocread -l
57.00M TrueFFS
| 3.06M Part00
| 2.88M Part01
| 51.06M Part02
48.23M TrueFFS
| 3.06M Part00
| 2.88M Part01
| 51.06M Part02
10.00M TRUEFFS
| 3.06M Part00
| 2.88M Part01
| 51.06M Part02
483.88M DSK1:
| 483.70M Part00
STRG handles: afb7d5fe
0 partitions, 0 binary partitions
customerid=00000000 uniqueid= 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
(483.70M) ce9ec2da
3 partitions, 2 binary partitions
customerid=00000000 uniqueid= 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
( 10.00M) 0eae8ede
3 partitions, 2 binary partitions
customerid=00000000 uniqueid= 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
( 48.23M) 0fb2bf42
3 partitions, 2 binary partitions
customerid=00000000 uniqueid= 00 00 00 00 fe 04 01 02 1c 32 08 ee 0d 07 05 4d
( 51.06M) efb2bd82
3 partitions, 2 binary partitions
customerid=00000000 uniqueid= 00 00 00 00 fe 04 01 02 1c 32 08 ee 0d 07 05 4d
( 2.88M) cfb2bd5e
3 partitions, 2 binary partitions
customerid=00000000 uniqueid= 00 00 00 00 fe 04 01 02 1c 32 08 ee 0d 07 05 4d
( 3.06M)
This machine has 128Mb of ROM, is there anyway I can read this using itsutils? pmemdump can successfully read only after 0x82000000 and prints out "Attempt to access invalid address"...
I dumped the first 16Megs after 0x82000000 and there's all kind of garbage, same like on the maui device. (marxx printed the first 8192 bytes in the above thread). Unfortunately, arm-wince-pe-objdump says "File format not recognized" on both dumps... What am I missing here?

richardt007 said:
arm-wince-pe-objdump says "File format not recognized" on both dumps... What am I missing here?
Click to expand...
Click to collapse
arm-wince-pe-objdump expects a PE binary, so
it is necessary to tell it that we have a raw binary dump.
arm-wince-pe-objdump -D -b binary -m arm dump.file

I have found this information about Linux
on HTC Typhoon:
http://vivien.chappelier.free.fr/typhoon/index.html

I was going to recomend a kernel built with omap 730 support, since the 850 seems to just be a 730 with edge. TI's site isn't to informational about the other circuits built into the omap850, or 730.

I have not been able to get the bootloader thing to do a single useful thing. I only get the tricolored screen. When pluged into the usb I get it saying usb in the corner. I set my mini sd up for the omap 730 image off the above site, but no dice.
My phone is the wiza 100 I beleive (cingular 8125).

YAY! PROGRESS! i will join you guys, when i find out more about how memory is organised. TI supports linux on the omap, maybe if we asked... :/

I never have any luck with the installing the damn toolchains, they never come with any instructions, or limited instructions. But I think I may have it. I nticed a lot of you are better coders and embedded developers than I I hope to learn a lot from you. Also, if there are any tutorials you can provide links to to help me learn, that would be perfect.

Help needed to Unlock Orange M2000 (Blue Angel)

I need help badly to unlock my Blue Angel. After running the xda_allunlock 1.6_138, I got the following message and the device is waiting for me to type in valid code:
xdadev_all_unlock started
device=PH20B
key found
ril ioctl 0xc2: 00000000
simlock answer:
invalid sim lock answer
error getting current lock status
xdadev_all_unlock failed
ril ioctl 0xc2: 00000000
ril ioctl 0xc6: 00000000
Detail of my Blue Angel:
Orange M2000
Model: PH20B
R: 1.04.00
Orginal SIM card: Orange
Deeply appreciate your advise and help, many thanks.

Orange SPV M500 - HELP A KIWI

Hey
A mate came from th UK to NZ and he gave me his orange SPV m500. He upgraded to the new M600 SPV
I tried the unlock application. I ran it on the phone and it said it was unlocked successfully but when I pop the new SIM in it does not work.
Here is the log file. When it comes to mobes, I am not an expert but I would appreciate any advice from the techies who know a thing or 2 about these mobiles. PLEASE PLEASE PLEASE.
The app I used was xdadev_all_unlock-v1.6.exe
I copied it to my phone and ran it. Was that the right way????
Here is my log file
*********************************
xdadev_all_unlock started
device=PM10B
key found
ril ioctl 0xc2: 00000000
simlock answer: %SIMLOCK= 05
0
current lock status: 05
security answer:
000
security answer2:
at%LISTNETWORKCODE answer: %NetworkCode: 23433F%NetworkCode: 00101F0
at%LISTNETSUBCODE answer: 0
at%LISTSPCODE answer: %SPCode: 000
at%LISTCORPORATECODE answer: 0
at%LISTSIMCODE answer: 0
at%LISTIMSIRANGECODE answer: 0
at%CLEARNETWORKCODE answer:
at%CLEARNETSUBCODE answer:
at%CLEARSPCODE answer:
at%CLEARCORPORATECODE answer:
at%CLEARSIMCODE answer:
at%CLEARIMSIRANGECODE answer:
removing lock 01
htclock answer:
removing lock 04
htclock answer:
simlock answer:
invalid sim lock answer
xdadev_all_unlock started
device=PM10B
key found
ril ioctl 0xc2: 00000000
simlock answer:
invalid sim lock answer
error getting current lock status
xdadev_all_unlock failed
ril ioctl 0xc2: 00000000
ril ioctl 0xc6: 00000000
xdadev_all_unlock done
ril ioctl 0xc2: 00000000
ril ioctl 0xc6: 00000000
xdadev_all_unlock started
device=PM10B
key found
ril ioctl 0xc2: 00000000
simlock answer: %SIMLOCK= 04
0
current lock status: 04
security answer:
000
security answer2:
at%LISTNETWORKCODE answer: 0
at%LISTNETSUBCODE answer: 0
at%LISTSPCODE answer: 0
at%LISTCORPORATECODE answer: 0
at%LISTSIMCODE answer: 0
at%LISTIMSIRANGECODE answer: 0
at%CLEARNETWORKCODE answer:
at%CLEARNETSUBCODE answer:
at%CLEARSPCODE answer:
at%CLEARCORPORATECODE answer:
at%CLEARSIMCODE answer:
at%CLEARIMSIRANGECODE answer:
removing lock 04
htclock answer:
simlock answer:
invalid sim lock answer
xdadev_all_unlock done
ril ioctl 0xc2: 00000000
ril ioctl 0xc6: 00000000

Hi,
I was curious to help, so I just gave it a try to my SPV M500 /aka HTC Magician, by the way)
It went great in less than a minute. I just copied xdadev_all_unlock-v1.6.exe to my SD card, then launched the app in the phone using file explorer. A pop up came after 20 seconds to confirm the phone had been successfully unlocked
You might give it another try, using the tool + info found on the link below. If still bad, you could try to hard reset and do the thing again...
http://forum.xda-developers.com/viewtopic.php?t=24779&highlight=xdadev

working!
thanx man

Hi mate, we have the same phones here at work, Orange SPV M500 but we use them with Vodafone GPRS sims, so we use the same program to unlock them. BUT I find that running it once doesn't always work, so I run it three times in a row, wait till you get the message on screen before re-running it and you will be fine mate.

i did with all instruction i copy to SD card then i got this info when i enter code it shows invalid plzz help
xdadev_all_unlock started
device=PM10B
key found
ril ioctl 0xc2: 00000000
simlock answer: %SIMLOCK= 05
0
current lock status: 05
security answer:
000
security answer2:
at%LISTNETWORKCODE answer: %NetworkCode: 23433F%NetworkCode: 00101F0
at%LISTNETSUBCODE answer: 0
at%LISTSPCODE answer: %SPCode: 000
at%LISTCORPORATECODE answer: 0
at%LISTSIMCODE answer: 0
at%LISTIMSIRANGECODE answer: 0
at%CLEARNETWORKCODE answer:
at%CLEARNETSUBCODE answer:
at%CLEARSPCODE answer:
at%CLEARCORPORATECODE answer:
at%CLEARSIMCODE answer:
at%CLEARIMSIRANGECODE answer:
removing lock 01
htclock answer:
removing lock 04
htclock answer:
simlock answer:
invalid sim lock answer
xdadev_all_unlock done
ril ioctl 0xc2: 00000000
ril ioctl 0xc6: 00000000
Click to expand...
Click to collapse

Why not a bigger swap?

Ive been thinking since we have always been told dont use a bigger than 32mb swap on any(most) hero ROMs
why not use a 256 mb swap to make the g1's internal ram to be "equal" to the htc hero
Maybe it would work or has this been tested before and end up being a total fail???
correct me please cuz im considering trying this out to see if it actually works
but am not sure if i want to, because of the obvious precautions.

Well, the phone won't kill off running applications until it's out of memory, so more memory = you end up with a hell of a lot more applications running, which is sortof pointless.
And it'll put a lot more preasure on the memory card.
But... loading an application from swap is probably faster than going through onCreate in most applications.

jf4888 said:
Ive been thinking since we have always been told dont use a bigger than 32mb swap on any(most) hero ROMs
why not use a 256 mb swap to make the g1's internal ram to be "equal" to the htc hero
Maybe it would work or has this been tested before and end up being a total fail???
correct me please cuz im considering trying this out to see if it actually works
but am not sure if i want to, because of the obvious precautions.
Click to expand...
Click to collapse
see this thread
http://forum.xda-developers.com/showthread.php?p=4124061
this post
#1902

toeshot said:
see this thread
http://forum.xda-developers.com/showthread.php?p=4124061
this post
#1902
Click to expand...
Click to collapse
The information in there is somewhat dated with regards to how swap works. Swap, back in the 60's, was program-by-program since multitasking wasn't what it is today. Now its in BLOCKS called MEMORY PAGES. The memory you save to swap is for RUNNING PROCESSES and is based on last-access time and access frequency.
Works like this;
You have 2 running processes, each uses 50 pages, total used is 100 pages. Lets say you have only 40 pages of physical ram and that each process has 10 active pages, that leaves 80 swappable pages, and 20 active pages. Obviously, not all can fit into ram, so 60 are dumped to swap. When one process tries to access memory that has been swapped to disk, there is a page fault and it loads that page from disk into memory and based on the algorithm used, it selects one of the in-memory pages and swaps it to disk.
Now the fun part: Lets say you try running 5 programs, each using 10 highly active pages and you only have 40 available. Now it starts THRASHING -- this is what kills performance, and on a multitasking OS, this is a very definite threat when the number of active memory pages exceeds the number available. When you increase your swap TOO MUCH, android will leave TOO MANY processes ACTIVE and you WILL GET THRASHING. Note that thrashing WILL KILL your sdcard FAST (though not necessarily any faster than you will get fed up with the slowness and reboot it).
Android has a very nice memory management feature where it kills those inactive processes. Having a huge amount of swap space will prevent this memory management feature from working correctly and will result in too many apps running, which will result in thrashing and poor performance. It is of particular importance to recognize this feature. Typical guidelines for setting swap space are not applicable to Android due to its memory management features.
Read this regarding paging (swapping):
http://en.wikipedia.org/wiki/Swap_space
http://en.wikipedia.org/wiki/Page_table

I am a little confused. I have read the links you provided, but I do not see anywhere were it says to decrease swap size. In fact after reading the articles I feel it supports my case even more. (See http://en.wikipedia.org/wiki/Thrash_(computer_science) Cause of Thrashing or even the overview)as for android killing sleeping process. that's not true either. here is an Process List of an idle G1 the S (Right Before the Process Name) stands for sleeping
# ps
USER PID PPID VSIZE RSS WCHAN PC NAME
root 1 0 292 208 c00ca8d0 0000caec S /init
root 2 0 0 0 c006c4bc 00000000 S kthreadd
root 3 2 0 0 c005bb94 00000000 S ksoftirqd/0
root 4 2 0 0 c00898bc 00000000 S watchdog/0
root 5 2 0 0 c008d1a0 00000000 S rcu_sched_grace
root 6 2 0 0 c0068620 00000000 S events/0
root 7 2 0 0 c0068620 00000000 S khelper
root 8 2 0 0 c0068620 00000000 S suspend
root 9 2 0 0 c0068620 00000000 S kblockd/0
root 10 2 0 0 c0068620 00000000 S kmmcd
root 11 2 0 0 c0068620 00000000 S btaddconn
root 12 2 0 0 c0068620 00000000 S btdelconn
root 13 2 0 0 c0068620 00000000 S kondemand/0
root 14 2 0 0 c0068620 00000000 S qmi
root 15 2 0 0 c0037830 00000000 D rpcrouter
root 16 2 0 0 c0068620 00000000 S detection/0
root 17 2 0 0 c00399f4 00000000 S krpcserversd
# ps
USER PID PPID VSIZE RSS WCHAN PC NAME
root 1 0 292 208 c00ca8d0 0000caec S /init
root 2 0 0 0 c006c4bc 00000000 S kthreadd
root 3 2 0 0 c005bb94 00000000 S ksoftirqd/0
root 4 2 0 0 c00898bc 00000000 S watchdog/0
root 5 2 0 0 c008d1a0 00000000 S rcu_sched_grace
root 6 2 0 0 c0068620 00000000 S events/0
root 7 2 0 0 c0068620 00000000 S khelper
root 8 2 0 0 c0068620 00000000 S suspend
root 9 2 0 0 c0068620 00000000 S kblockd/0
root 10 2 0 0 c0068620 00000000 S kmmcd
root 11 2 0 0 c0068620 00000000 S btaddconn
root 12 2 0 0 c0068620 00000000 S btdelconn
root 13 2 0 0 c0068620 00000000 S kondemand/0
root 14 2 0 0 c0068620 00000000 S qmi
root 15 2 0 0 c0037830 00000000 D rpcrouter
root 16 2 0 0 c0068620 00000000 S detection/0
root 17 2 0 0 c00399f4 00000000 S krpcserversd
root 18 2 0 0 c00979e0 00000000 S pdflush
root 19 2 0 0 c00979e0 00000000 S pdflush
root 20 2 0 0 c009cda8 00000000 S kswapd0
root 21 2 0 0 c0068620 00000000 S aio/0
root 24 2 0 0 c00388d0 00000000 S kadspd
root 36 2 0 0 c0068620 00000000 S panel_on/0
root 43 2 0 0 c01fb17c 00000000 S mtdblockd
root 44 2 0 0 c0201db0 00000000 S USB mass_storag
root 46 2 0 0 c0068620 00000000 S synaptics_wq
root 47 2 0 0 c0068620 00000000 S kstriped
root 48 2 0 0 c0068620 00000000 S hid_compat
root 49 2 0 0 c02ed850 00000000 S krfcommd
root 52 2 0 0 c0230fe8 00000000 S mmcqd
root 53 2 0 0 c01e5df4 00000000 S loop0
root 54 2 0 0 c01e5df4 00000000 S loop1
root 56 1 3332 156 ffffffff 0000e8f4 S /sbin/adbd
root 66 2 0 0 c014bc74 00000000 S kjournald
system 109 1 808 168 c0242cb0 afe0c45c S /system/bin/servicemanager
root 110 1 848 256 c00ca8d0 afe0c584 S /system/bin/vold
root 111 1 668 152 c02504d8 afe0cdec S /system/bin/debuggerd
radio 112 1 7896 696 ffffffff afe0cacc S /system/bin/rild
root 113 1 73092 13704 c00ca8d0 afe0c584 S zygote
media 114 1 24272 1248 ffffffff afe0c45c S /system/bin/mediaserver
bluetooth 117 1 1172 224 c00ca8d0 afe0d25c S /system/bin/dbus-daemon
root 118 1 800 192 c02cf988 afe0c1bc S /system/bin/installd
compass 120 1 3372 472 ffffffff afe0c45c S /system/bin/akmd
system 133 113 154716 23712 ffffffff afe0c45c S system_server
radio 168 113 107580 17276 ffffffff afe0d3e4 S com.android.phone
app_7 174 113 119796 20292 ffffffff afe0d3e4 S android.process.acore
app_62 208 113 99700 15032 ffffffff afe0d3e4 S com.p1.chompsms
app_1 221 113 110712 16964 ffffffff afe0d3e4 S com.google.process.gapps
app_20 234 113 98256 14464 ffffffff afe0d3e4 S android.process.media
root 267 2 0 0 c00386f0 00000000 D audmgr_rpc
app_36 310 113 95004 13496 ffffffff afe0d3e4 S com.android.alarmclock
app_21 344 113 106684 16440 ffffffff afe0d3e4 S com.htc.android.mail
root 368 1 4064 408 c00c3180 afe0c1bc S su
app_65 369 113 95608 13488 ffffffff afe0d3e4 S com.phonefusion.voicemailplus.android
app_54 418 113 95020 13692 ffffffff afe0d3e4 S org.freecoder.widgets.overclock
app_27 500 113 106180 15536 ffffffff afe0d3e4 S com.google.android.gm
app_7 561 113 98652 13564 ffffffff afe0d3e4 S com.android.inputmethod.latin
app_43 568 113 106064 15484 ffffffff afe0d3e4 S com.koushikdutta.klaxon
app_32 761 113 98412 15280 ffffffff afe0d3e4 S com.android.calendar
app_71 775 113 94428 13048 ffffffff afe0d3e4 S com.roundedlabs.widgets.gps
app_42 1058 113 94800 17008 ffffffff afe0d3e4 S com.twidroid:remote
root 1083 56 740 332 c0059630 afe0d08c S /system/bin/sh
root 1086 1083 884 340 00000000 afe0c1bc R ps
Now I don't have a swap partition active nor am I using swapper and I have 2k of free memory. If select on a 32mb for a swap partition that partition would fill quickly with the sleeping process, yes they would go to the swap partition However the swap partiton would not be completely filled to 32 meg because of the partition overhead. the memory that I freed up be partially available, meaning I would not see a full 32mb. however if I select 100mb swap partition it would not fill up, file swapping would better as there is more available space to put the sleeping programs until needed, freeing up more ram for the phone when I start to use it.
One thing to remember is that even though the android Cupcake will multi thread and multi task processes in the background it will not truly multi task applications. For example you can not open up a sms message and then open up google maps without the sms going into the background and going to sleepand if you don't have GPS on once the map is loaded it will go to sleep until you interact with it.

[Q] Android Wallpaper Service on Nook Simple Touch (the removal of)

Hi all,
On a Nook Simple Touch (E-ink reader), there is a service I am unable to locate and delete.
The details:
Code:
Label: Android System
Name: system
Flags: PERSISTENT, SYSTEM
Clients: I
Package: android
Pack. label: Wallpaper
Pack. details: android/com.android.internal.service.wallpaper.ImageWallpaper
While I accept that the system requires resources, this service degrades the performance of the e-reader every now and then to irritatingly slow levels. When it is persistently using CPU time, the screen is very noticeably unresponsive, perhaps once a day or so.
grep reports that this service is a "Default built-in wallpaper that simply shows a static image. "
Added to the rather odd fact that I do not recall having a wallpaper beyond a sheet of white, which I would gladly sacrifice in return for a reliably responsive system.
I additionally don't know how a wallpaper service takes almost more memory than my web browser.
I have been unable to locate documentation on this service, what it does, where it resides and how to disable it.
If anyone could provide any insight on this service, it would be greatly appreciated.
Thanks
(tl;dr: Android's imagewallpaper service uses more RAM/CPU than a web browser, and I would like to kill it or understand why)

The service you mention is the default and very simple implementation of a WallpaperService.
Basically, it gets an intent to use a drawable as a background and draws it when needed.
It really should not be using any CPU or memory.

But it is, and using 10-20% of CPU and about a quarter of my RAM.
How would one go about diagnosing the root cause, and subsequently disabling it?
I tried 'pm disable servicename' but it only shifted the exact same CPU and RAM usage to the Google checkin service, and after I disabled that, to the Android system itself.
What might be causing this?

Where are you getting these statistics from?
Are you getting these statistics over ADB without an app running?
What are you using for a launcher?
Do you have any widgets on the desktop?
What does logcat say?
What does ps say?

Renate NST said:
Where are you getting these statistics from?
Click to expand...
Click to collapse
From the task manager built into Relaunch (nookdevs . com/ReLaunch), a barebones file manager/launcher without animations nor wallpaper.
Renate NST said:
Are you getting these statistics over ADB without an app running?
Click to expand...
Click to collapse
This is persistent over a week-long period of regular usage, as in, the resource consumption is constant regardless of what I am doing with the device (ie. fresh boot, web browser, reading).
I am unaware of the command to pull these statistics over ADB, though I suspect they will be similar.
Renate NST said:
What are you using for a launcher?
Click to expand...
Click to collapse
Relaunch (nookdevs . com/ReLaunch)
Written for the Nook, spartan feature set includes only file manager, basic app selection, wifi and battery information.
UI is simplistic, about what you'd expect from a file manager.
No animations.
Renate NST said:
Do you have any widgets on the desktop?
Click to expand...
Click to collapse
I have a listing of files in the current directory, beyond that nothing.
Renate NST said:
What does logcat say?
Click to expand...
Click to collapse
pastebin . com/j4EWAH0R
(Does not seem to have anything relevant, but the lagging problems only manifest themselves once a day or so.)
Renate NST said:
What does ps say?
Click to expand...
Click to collapse
Code:
USER PID PPID VSIZE RSS WCHAN PC NAME
root 1 0 552 248 c00d4988 0000d6cc S /init
root 2 0 0 0 c0078ee4 00000000 S kthreadd
root 3 2 0 0 c006925c 00000000 S ksoftirqd/0
root 4 2 0 0 c0090d48 00000000 S watchdog/0
root 5 2 0 0 c0075624 00000000 S events/0
root 6 2 0 0 c0075624 00000000 S khelper
root 12 2 0 0 c0075624 00000000 S suspend
root 127 2 0 0 c0075624 00000000 S kblockd/0
root 133 2 0 0 c0075624 00000000 S cqueue
root 143 2 0 0 c0075624 00000000 S twl4030-irqchip
root 144 2 0 0 c01a4b54 00000000 S twl4030-irq
root 170 2 0 0 c0075624 00000000 S omap2_mcspi
root 177 2 0 0 c0075624 00000000 S ksuspend_usbd
root 182 2 0 0 c01df4e4 00000000 S khubd
root 187 2 0 0 c0075624 00000000 S kmmcd
root 218 2 0 0 c00a573c 00000000 S pdflush
root 219 2 0 0 c00a573c 00000000 S pdflush
root 220 2 0 0 c00aa36c 00000000 S kswapd0
root 222 2 0 0 c0075624 00000000 S aio/0
root 230 2 0 0 c0075624 00000000 S epd-su/0
root 231 2 0 0 c0075624 00000000 S epd-dspck/0
root 371 2 0 0 c01f7818 00000000 S usb_mass_storag
root 385 2 0 0 c0075624 00000000 S zforce_wq
root 404 2 0 0 c0075624 00000000 S kondemand/0
root 416 2 0 0 c0075624 00000000 S hid_compat
root 485 2 0 0 c0215cb8 00000000 S mmcqd
root 716 2 0 0 c0215cb8 00000000 S mmcqd
root 720 2 0 0 c012dcdc 00000000 S kjournald
root 724 2 0 0 c012dcdc 00000000 S kjournald
root 725 1 700 276 c017e784 afe0c74c S /system/bin/sh
root 728 1 836 360 c00d4988 afe0cb14 S /system/bin/OMXPolicyManager
root 729 1 1900 360 ffffffff afe0c7ec S /system/bin/OMXResourceManager
root 734 1 216 64 c007dd0c 0001413c S /system/bin/bridged
system 735 1 764 264 c022646c afe0c9ec S /system/bin/servicemanager
root 737 1 816 400 c00d4988 afe0cb14 S /system/bin/vold
root 738 1 628 244 c0248ce4 afe0d37c S /system/bin/debuggerd
root 740 1 171004 8552 c00d4988 afe0cb14 S zygote
media 741 1 28516 1820 ffffffff afe0c9ec S /system/bin/mediaserver
bluetooth 742 1 1068 348 c00d4988 bef345b4 S /system/bin/dbusdaemon
root 743 1 756 328 c02ab6e8 afe0c74c S /system/bin/installd
keystore 746 1 1600 272 c0248ce4 afe0d37c S /system/bin/keystore
root 754 2 0 0 c0075624 00000000 S pvr_timer/0
root 782 2 0 0 c0075624 00000000 S pvr_workqueue
root 790 1 16832 12172 c016e7e8 afe0c74c S /system/bin/omap-edpd.elf
system 810 740 353844 54808 ffffffff afe0c9ec S system_server
root 820 2 0 0 c0075624 00000000 S pvr_display_syn
system 1003 740 258112 16916 ffffffff afe0d974 S com.android.settings
root 1008 2 0 0 c0075624 00000000 S sdio_wq
root 1015 2 0 0 c0075624 00000000 S tiwlan_wq
wifi 1032 1 1988 668 c00d4988 afe0cb14 S /system/bin/wpa_supplicant
app_2 1043 740 245724 12396 ffffffff afe0d974 S com.android.inputmethod.latin
app_3 1108 740 249840 11984 ffffffff afe0d974 S com.bn.cloud.svc:svc
system 1134 740 252272 18352 ffffffff afe0d974 S com.bn.devicemanager
system 1181 740 261060 10192 ffffffff afe0d974 S com.bn.waveformdownloader.svc
app_22 1201 740 249236 11240 ffffffff afe0d974 S com.android.vending
app_23 1558 740 308052 62676 ffffffff afe0d974 S com.UCMobile
app_29 1742 740 253832 15392 ffffffff afe0d974 S com.eink.norefresh
app_29 1756 1742 636 344 c0227544 afe0c74c S /system/bin/logcat
app_34 3105 740 259656 21560 ffffffff afe0d974 S com.harasoft.relaunch
dhcp 3173 1 304 164 c00d4988 0000ee6c S /system/bin/dhcpcd
app_30 3247 740 250176 15752 ffffffff afe0d974 S siir.es.adbWireless
root 3289 1 3372 184 ffffffff 0000f444 S /sbin/adbd
root 3369 3289 700 328 c017e784 afe0c74c S /system/bin/sh
root 3370 3289 636 344 c0227544 afe0c74c S logcat
app_17 3395 740 251488 11776 ffffffff afe0d974 S com.google.process.gapps
app_20 3413 740 244860 10292 ffffffff afe0d974 S com.android.setupwizard
root 3442 3289 700 328 c0066b58 afe0d61c S /system/bin/sh
root 3446 3442 844 340 00000000 afe0c74c R ps
Thanks
Edit: The PID/UID of the problematic process is 810/1000, which corresponds to "system 810 740 353844 54808 ffffffff afe0c9ec S system_server" from the ps result. I'm searching for insight on system_server.

Thanks, oops, sorry. I meant to say "top".
If you are running something that displays stats constantly it could be causing the wallpaper to constantly redraw.
Try a top from ADB, I'm sure that it will show zero usage.
You do have some wackiness going on in your logcat.
The UMSServer is going crazy. Is there a problem with your SD card?

Cross-referencing the PID with the processes reported in the task manager appears to point the issue at "system_server"
I'm not sure of the relevance of this, but it appears other Android devices have encountered issues with this particular process, with symptoms similar to mine.
However, their solutions (disabling GPS, replacing 'system_server' with its 2.3/4.0 counterpart) do not seem to apply, or cause more issues in this case.
The UMS messages only show up during physical USB connection, not during logcat when there is only a wireless connection, so it should be unrelated.
Could you clarify 'top'?
Your explanation makes sense, but it doesn't fully explain why the system slows to a crawl for about ten to twenty minutes out of the blue every now and then, seemingly caused by this process.
However, this may just be an unrelated factor.
I have, in my search for what 'system_server' does, discovered what happens if you replace said file with its 2.3/4.0 counterpart, should anyone wish to know.
Regular functions are unaffected, resource usage by the process is slightly diminished, however the tablet is unable to wake from sleep.
Additionally, removing most of the Google and B&N system services has somewhat circumvented the issue.
Thanks for your help, by the way, wouldn't have gotten past the whole figuring out what ImageWallpaper was without it.

I looked at the code for UMSServerActivity quickly and jumped to the conclusion that it was for mounting of the SD card.
It's actually the USB Mass Storage mounter.
I looked at logcat with a USB drive plugged in and I don't get any of those messages that you have.
From a shell in ADB:
Code:
# top
User 0%, System 0%, IOW 0%, IRQ 0%
User 3 + Nice 0 + Sys 1 + Idle 300 + IOW 0 + IRQ 0 + SIRQ 0 = 304
PID CPU% S #THR VSS RSS PCY UID Name
24453 0% R 1 876K 388K fg root top
2 0% S 1 0K 0K fg root kthreadd
3 0% S 1 0K 0K fg root ksoftirqd/0
4 0% S 1 0K 0K unk root watchdog/0
and so on for another 50 lines...

There is a somewhat more pressing issue, is there a way to have Android save logcat data automatically regardless of system state and ADB connection?
Right now it only has ADB over wifi, which is terminated soon after the device sleeps.
The reason I ask is that right now if I leave it for too long it refuses to wake up, I assumed this was an issue with the replacement of "system_server" in "/system/bin/", as the reverting those changes initially did away with the problem, but it just manifested itself twice over the span of five minutes and I'd like to track
down the source.
Thanks

aliastral said:
I assumed this was an issue with the replacement of "system_server" in "/system/bin/"
Click to expand...
Click to collapse
Um, if you start indiscriminately swapping out system parts you are bound to cause wackiness.
For example: I've been having problems with "mediaserver".
It crashes all the time, but what is in /system/bin/mediaserver is just a stub that gets a bunch of processes running.
I wouldn't think of trying to swap out /system/bin/mediaserver.
The problem probably lies in a bug in one of the audio libraries in /system/lib