PPCGeeks Forums HTC Arrive HTC HD2 HTC Thunderbolt HTC Touch Pro 2 HTC Evo 4G HTC Evo 3D Samsung Galaxy S II Motorola Droid X Apple iPhone Blackberry
Go Back   PPCGeeks > Windows Mobile > WM HTC Devices > HTC Touch Pro 2 > Android On TP2 > Android On TP2 Development
Register Community Search

Notices


Reply
 
LinkBack Thread Tools Display Modes
  #1 (permalink)  
Old 03-01-2011, 03:41 PM
MassStash's Avatar
Regular 'Geeker
Offline
Pocket PC: Rhodium
Carrier: Sprint
Location: Il
 
Join Date: Oct 2008
Posts: 437
Reputation: 285
MassStash is becoming a PPCGeeks regularMassStash is becoming a PPCGeeks regularMassStash is becoming a PPCGeeks regular
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Re: NAND Boot Testing - 01-07: Panel power on/off fixes

booting with new ril...

update: homescreen.... nah, nothing. APN change to #777, nothing anything else i try before work at 4:30?
__________________
Rockin' The Official TouchPro 2 aka "The dopest phone with keyboard still":
F**k windows mobile, NAND flash to droid....

Last edited by MassStash; 03-01-2011 at 04:13 PM.
Reply With Quote
  #2 (permalink)  
Old 03-01-2011, 04:12 PM
[ACL]'s Avatar
VIP Member
Offline
Pocket PC: tp2
Carrier: Sprint
Location: NY
 
Join Date: Feb 2010
Posts: 1,534
Reputation: 6350
[ACL] is a trusted member of the community[ACL] is a trusted member of the community[ACL] is a trusted member of the community[ACL] is a trusted member of the community[ACL] is a trusted member of the community[ACL] is a trusted member of the community[ACL] is a trusted member of the community[ACL] is a trusted member of the community[ACL] is a trusted member of the community[ACL] is a trusted member of the community[ACL] is a trusted member of the community
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Re: NAND Boot Testing - 01-07: Panel power on/off fixes

Quote:
Originally Posted by MassStash View Post
booting with new ril...

update: homescreen.... nah, nothing. APN change to #777, nothing
fak...

gonna have to test the symlink method..
__________________
Reply With Quote
  #3 (permalink)  
Old 03-01-2011, 04:13 PM
MassStash's Avatar
Regular 'Geeker
Offline
Pocket PC: Rhodium
Carrier: Sprint
Location: Il
 
Join Date: Oct 2008
Posts: 437
Reputation: 285
MassStash is becoming a PPCGeeks regularMassStash is becoming a PPCGeeks regularMassStash is becoming a PPCGeeks regular
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Re: NAND Boot Testing - 01-07: Panel power on/off fixes

yea hows i do that? symlink of pppd eh?

Update: answer to my question is remake the initrd.
So i think acl's putting service on his beast tonight to knock this one out....
Did what i Could haha

Last edited by MassStash; 03-01-2011 at 08:26 PM.
Reply With Quote
  #4 (permalink)  
Old 03-02-2011, 12:34 AM
[ACL]'s Avatar
VIP Member
Offline
Pocket PC: tp2
Carrier: Sprint
Location: NY
 
Join Date: Feb 2010
Posts: 1,534
Reputation: 6350
[ACL] is a trusted member of the community[ACL] is a trusted member of the community[ACL] is a trusted member of the community[ACL] is a trusted member of the community[ACL] is a trusted member of the community[ACL] is a trusted member of the community[ACL] is a trusted member of the community[ACL] is a trusted member of the community[ACL] is a trusted member of the community[ACL] is a trusted member of the community[ACL] is a trusted member of the community
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Re: NAND Boot Testing - 01-07: Panel power on/off fixes

I think i got it..

For now try this and tell me if it works.

adb into the phone and do the following

# /system/bin/pppd /dev/smd1 &
# ps w

Look for the rild service and kill it

kill <pid> <--- pid is the number shown when you do a "ps w"

Now that you have that look at your logcat and you will see pppd firing up. Try to browse the web and let me know how we cookin
Reply With Quote
  #5 (permalink)  
Old 03-02-2011, 03:36 AM
natemcnutty's Avatar
VIP Member
Offline
Pocket PC: VZW Touch Pro 2
Carrier: Verizon Wireless
Threadstarter
 
Join Date: Nov 2009
Posts: 845
Reputation: 3070
natemcnutty is still contributing even after becoming a VIPnatemcnutty is still contributing even after becoming a VIPnatemcnutty is still contributing even after becoming a VIPnatemcnutty is still contributing even after becoming a VIPnatemcnutty is still contributing even after becoming a VIPnatemcnutty is still contributing even after becoming a VIPnatemcnutty is still contributing even after becoming a VIPnatemcnutty is still contributing even after becoming a VIPnatemcnutty is still contributing even after becoming a VIPnatemcnutty is still contributing even after becoming a VIPnatemcnutty is still contributing even after becoming a VIP
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Re: NAND Boot Testing - 01-07: Panel power on/off fixes

OK, regarding the mtdparts, I was thinking that maybe we should emulate the official Google dev phones which should help as we try to more closely stick to AOSP. If we get a bootloader that supports it, we should be able to use fastboot with this layout, and we may even be able to add CWM/ROM Manager support

mtdparts=msm_nand:
0x00480000@0x01000000(recovery) - (4,608 KB)
0x00300000@0x01480000(boot) - (3,072 KB)
0x09100000@0x01780000(system) - (148,480 KB)
0x05F00000@0x0A880000(cache) - (97,280 KB)
0x0F7E0000@0x10780000(userdata) - (253,824 KB)
0x000A0000@0x1FF60000(misc) - (640 KB)

The cache size may seem large, but that data is going to be cached somewhere, and it would ultimately end up in the userdata section anyway. We can conform to AOSP by having its own partition, and we should move the dalvik cache to that if it is being enabled. I think the more "best practices" we can use, the better performance we are going to get from our phone. Plus, roughly 250 MB of space for applications and such should be sufficient since we can just use app2sd.

For system, that should be more than enough space even if we were putting modules into the system. The latest Cyanogenmod systems are roughly 110 MB fully extracted. Not sure of the size of the sense builds, but I really think that should be plenty for the chefs.

I will be trying this partition layout tomorrow to make sure it works (don't see any reason it shouldn't). I'll also try changing /cache to mount on the cache partition and make sure that is working OK.
Reply With Quote
  #6 (permalink)  
Old 03-02-2011, 02:29 PM
[ACL]'s Avatar
VIP Member
Offline
Pocket PC: tp2
Carrier: Sprint
Location: NY
 
Join Date: Feb 2010
Posts: 1,534
Reputation: 6350
[ACL] is a trusted member of the community[ACL] is a trusted member of the community[ACL] is a trusted member of the community[ACL] is a trusted member of the community[ACL] is a trusted member of the community[ACL] is a trusted member of the community[ACL] is a trusted member of the community[ACL] is a trusted member of the community[ACL] is a trusted member of the community[ACL] is a trusted member of the community[ACL] is a trusted member of the community
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Re: NAND Boot Testing - 01-07: Panel power on/off fixes

Quote:
Originally Posted by natemcnutty View Post
OK, regarding the mtdparts, I was thinking that maybe we should emulate the official Google dev phones which should help as we try to more closely stick to AOSP. If we get a bootloader that supports it, we should be able to use fastboot with this layout, and we may even be able to add CWM/ROM Manager support

mtdparts=msm_nand:
0x00480000@0x01000000(recovery) - (4,608 KB)
0x00300000@0x01480000(boot) - (3,072 KB)
0x09100000@0x01780000(system) - (148,480 KB)
0x05F00000@0x0A880000(cache) - (97,280 KB)
0x0F7E0000@0x10780000(userdata) - (253,824 KB)
0x000A0000@0x1FF60000(misc) - (640 KB)

The cache size may seem large, but that data is going to be cached somewhere, and it would ultimately end up in the userdata section anyway. We can conform to AOSP by having its own partition, and we should move the dalvik cache to that if it is being enabled. I think the more "best practices" we can use, the better performance we are going to get from our phone. Plus, roughly 250 MB of space for applications and such should be sufficient since we can just use app2sd.

For system, that should be more than enough space even if we were putting modules into the system. The latest Cyanogenmod systems are roughly 110 MB fully extracted. Not sure of the size of the sense builds, but I really think that should be plenty for the chefs.

I will be trying this partition layout tomorrow to make sure it works (don't see any reason it shouldn't). I'll also try changing /cache to mount on the cache partition and make sure that is working OK.
So i found this online and wanted to share.

/misc - format raw: contains some storage for the SPL + means for some information to be transferred between the SPL/kernel

/recovery - bootable image (Kernel + ram disk see android documentation) the recovery image

/boot - bootable image (Kernel + ram disk see android documentation) contains the system kernel+ram disk

/system - the system data (not erased on factory restore) what is generally considered the "rom" (yaffs2 usually mounted at /system)

/cache - Temporary storage; or cached data.. usually persistent over reboots but not guaranteed (yaffs2 usually mounted at /cache) also used by recovery SPL for hboot/radio flashes; as the SPL must conduct these. Used by system/recovery for system initiated updates such as ota (not usually done from custom firmwares)


In terms of AOSP, we only need 3 partitions (sys, data and cache). Android doesn't deal with the others. Our SPL unlike the android ones may not actually use any of the partitions. And that will never change so we cant truly run like an oem phone. The other partitions are really for OEMs to decide if they want to implement. For example my galaxy S had a weird layout matching none ive seen before.

For example im sure a real android device may use the /misc mount to keep lets say its touch screen calibration points since spl reads it. But our spl writes it in some weird location we havent wiped (luckily) .. So that partition would be a real waste.

I guess we should table this until Alex decides how LK will be used. Until then i'm OK with adding cache for sure. And maybe 1 more for boot to allow kernel updates. But i doubt LK will be able to do things like update our radio.

With that said this is actually a tough decision. Cotulla himself will only make 3 parititons (sys, data, cache) and Stine doesnt really care much for cache since his OTA updates will be done another way. So with that said, we can use cache and possibly recovery. Rest will be a waste of space. How do you guys feel about that?

Edit: also the old sense was massive. I think 130mb just for system. So i can only imagine the newer sense builds.

Last edited by [ACL]; 03-02-2011 at 02:43 PM.
Reply With Quote
This post has been thanked 1 times.
  #7 (permalink)  
Old 03-02-2011, 04:20 PM
natemcnutty's Avatar
VIP Member
Offline
Pocket PC: VZW Touch Pro 2
Carrier: Verizon Wireless
Threadstarter
 
Join Date: Nov 2009
Posts: 845
Reputation: 3070
natemcnutty is still contributing even after becoming a VIPnatemcnutty is still contributing even after becoming a VIPnatemcnutty is still contributing even after becoming a VIPnatemcnutty is still contributing even after becoming a VIPnatemcnutty is still contributing even after becoming a VIPnatemcnutty is still contributing even after becoming a VIPnatemcnutty is still contributing even after becoming a VIPnatemcnutty is still contributing even after becoming a VIPnatemcnutty is still contributing even after becoming a VIPnatemcnutty is still contributing even after becoming a VIPnatemcnutty is still contributing even after becoming a VIP
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Re: NAND Boot Testing - 01-07: Panel power on/off fixes

Thanks for the lowdown ACL. I'm not sure how much space you need for recovery/root, but I think you're right that we don't need a couple of the others. This is what I'll be using today, and maybe we can settle on this:

mtdparts=msm_nand:
0x00400000@0x01000000(recovery) - (4 MB)
0x0A000000@0x01400000(system) - (160 MB)
0x04000000@0x0B400000(cache) - (64 MB)
0x10C00000@0x0F400000(userdata) - (268 MB)

Gives extra headroom for Sense builds or future builds that get larger. Cache should be sufficient, but we would just take away from userdata and give to cache in the future if needed. And 250 MB are left for userdata (after modules) which is almost exactly what we have on Haret.

Let me know what you think and let's vote this thing
Reply With Quote
This post has been thanked 1 times.
Reply

  PPCGeeks > Windows Mobile > WM HTC Devices > HTC Touch Pro 2 > Android On TP2 > Android On TP2 Development


Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are On


All times are GMT -4. The time now is 07:20 PM.


Powered by vBulletin® ©2000 - 2025, Jelsoft Enterprises Ltd.
Content Relevant URLs by vBSEO 3.6.0
©2012 - PPCGeeks.com