NAND Testing - 05-25 Update: New LK, Recovery.img, Kernel Updates through Recovery

[Lmiller1708]

Originally Posted by arrrghhh

dzo responded! Gotta love the guy. Here's what he told me... I think the HD2 guys came to the same conclusion, AFAIK: (seems like you have as well )


I hope it does too. Let me know how it goes!

Thanks arrrghhh!

More for reference...

Here is a Normal Vogue ROM NBInfo 2.1rc2 VOGUE '..\OS.nb.payload' has val - Vogue
Here is a Normal TP2 ROM NBInfo 2.1rc2 TP2 '..\OS.nb.payload' has valid - TP2

Not too sure how to change the partition table just yet, even if we figure this out what are the values we need to change them too, ideas?

[[ACL]]

Originally Posted by arrrghhh

dzo responded! Gotta love the guy. Here's what he told me... I think the HD2 guys came to the same conclusion, AFAIK: (seems like you have as well )


I hope it does too. Let me know how it goes!

guy is a legend.. to think, dzo knows we exist.

Unfortunately i know shit about partition tables.. good thing nate has a better grasp.

also who is an SPL master ? someone who knows all those mtty commands? would be nice to just do this via spl and call it a day.

Worse case cant we sniff the usb traffic to mtty and see how winmo does its nonsense ?

[natemcnutty]

Originally Posted by [ACL]

guy is a legend.. to think, dzo knows we exist.

Unfortunately i know shit about partition tables.. good thing nate has a better grasp.

also who is an SPL master ? someone who knows all those mtty commands? would be nice to just do this via spl and call it a day.

Worse case cant we sniff the usb traffic to mtty and see how winmo does its nonsense ?

I am totally stoked by that information. For partition tables, I'm going to take a look into that. What we need at this point is to find out where the NAND starts, position tinboot with the small partition table so it only uses one block, figure out the offset for the rest of it, and put that into the mtdparts command. It's all HEX based for sizes, and is pretty easy to compute. I think following along the way the Vogue does things for now would be good until we can learn why they did them that way :P

Also, I'll try to get info on mtty, and you can always use snoopy pro: http://sourceforge.net/projects/usbsnoop/

[[ACL]]

Originally Posted by natemcnutty

I am totally stoked by that information. For partition tables, I'm going to take a look into that. What we need at this point is to find out where the NAND starts, position tinboot with the small partition table so it only uses one block, figure out the offset for the rest of it, and put that into the mtdparts command. It's all HEX based for sizes, and is pretty easy to compute. I think following along the way the Vogue does things for now would be good until we can learn why they did them that way :P

Also, I'll try to get info on mtty, and you can always use snoopy pro: Usb Sniffer for Windows | Download Usb Sniffer for Windows software for free at SourceForge.net

there is really no need right ? We are sending the imgfs in the payload. Vogue adds this separately through tinboot. So we we need to mod our payload and strip the imgfs portion so we can add it to tinboot as well. We can actually use their emtpy imgfs which will just wipe one block. I dont think the partition table on the vogue even matches the mtd parts.

I can try to snoop the mtty tonight and see what the whole flash process looks like.

who isnt working right now? come on someone lets try this !!!

[Lmiller1708]

Originally Posted by [ACL]

there is really no need right ? We are sending the imgfs in the payload. Vogue adds this separately through tinboot. So we we need to mod our payload and strip the imgfs portion so we can add it to tinboot as well. We can actually use their emtpy imgfs which will just wipe one block. I dont think the partition table on the vogue even matches the mtd parts.

I can try to snoop the mtty tonight and see what the whole flash process looks like.

who isnt working right now? come on someone lets try this !!!

Tried running it on Windows 7 but it didn't work for me.

[natemcnutty]

Originally Posted by Lmiller1708

Tried running it on Windows 7 but it didn't work for me.

Yeah, to get mtty or snoopy pro to work on Win7, you have to restart the computer with driver enforcement disabled, then install the driver, then go at it, and you will have to restart with driver enforcement disabled every single time you want to do it.

I actually keep an old XP laptop laying around just for the purpose of running old stuff like this

[natemcnutty]

Originally Posted by [ACL]

there is really no need right ? We are sending the imgfs in the payload. Vogue adds this separately through tinboot. So we we need to mod our payload and strip the imgfs portion so we can add it to tinboot as well. We can actually use their emtpy imgfs which will just wipe one block. I dont think the partition table on the vogue even matches the mtd parts.

I can try to snoop the mtty tonight and see what the whole flash process looks like.

who isnt working right now? come on someone lets try this !!!

Yeah, what I am looking at is finding out exactly where the NAND starts so we can put tinboot at the very beginning and maximize the amount of space we have to use on NAND. I guess it doesn't really matter at this point since we know the locations we are writing to are valid.

I guess we could try using the vogue's current tinboot build, make the MTYPE change, mem start change and other changes and see how it works. I was in the process of doing that last night just to see if it worked by chance, but I got pulled away by the wife. Unfortunately, we have a dinner party at some friends tonight, so I won't have a computer all night

[[ACL]]

Originally Posted by natemcnutty

Yeah, what I am looking at is finding out exactly where the NAND starts so we can put tinboot at the very beginning and maximize the amount of space we have to use on NAND. I guess it doesn't really matter at this point since we know the locations we are writing to are valid.

I guess we could try using the vogue's current tinboot build, make the MTYPE change, mem start change and other changes and see how it works. I was in the process of doing that last night just to see if it worked by chance, but I got pulled away by the wife. Unfortunately, we have a dinner party at some friends tonight, so I won't have a computer all night

i already diff'ed the tinboot sources and they aren't that different. Lmiller actually already added the emptyimgs portion at the end but one thing missing is removing it from the payload. I'm not sure how to do this.. is there a partition table editor ? lol or do we have to hex edit it out?

Or take the partitions table and hex edit it to match our layout. This way we dont remove data, just modify it.

[MassStash]

lmiller, you tryin the driver again or you need me to try?

[[ACL]]

Originally Posted by MassStash

lmiller, you tryin the driver again or you need me to try?

damn.. you look hungry.. like you wanna test. sounds like you are ready to modify some partition tables..