Re: [TESTING] Updated 1/30/11 New Kernel (100% FIXED Panel Wake)
 

I'm seeing pretty outstanding results with 1/30(2) on my Sprint Rhod400 as well. 100% wake so far, and wake is fast, less than a second.

Re: [TESTING] Updated 1/30/11 New Kernel (100% FIXED Panel Wake)
 

just tried this out on Sprint (rhod400 obviously) and it is waking up correctly about 9 out of 10 tries. one thing i am noticing though is that the led goes to green light almost immediately when i turn the screen off (awesome) but it keeps cycling from green to orange every 3 seconds or so...

the setup i'm using for this test is tiad8's latest (FROYO X 2011.01.28 SYSTEM.EXT2 UPDATED). the only mods i did after initial boot were adding APN for sprint, adding my main account which syncs my contacts, setting the brightness, and finally setting my preferred ringtone. after that, i rebooted to WinMo, added this test module/zimage and booted back to android. i have the default startup.txt for sprint other than adding rel_path=andboot.

hope this helps! you guys are kicking ass and making my TP2 rock!

Re: [TESTING] Updated 1/30/11 New Kernel (100% FIXED Panel Wake)
 

Does this happen on FRX04? I haven't had this happen at all, but I also don't get the greed LED immediately either. It's either 5s or 30s it seems to go to green LED after hitting the power button, and I have the same device - RHOD400.

I'd prefer we all test on FRX04, but I can't force ya... All I can do is blame your weird issues on the build you're using! ;)

Re: [TESTING] Updated 1/30/11 New Kernel (100% FIXED Panel Wake)
 

New kernel posted guys.

This one is ONLY for debugging that nasty SoD issue where the phone is completely down - you cannot wake it no matter what, waiting 10 minutes, calling it, pressing every button - nothing.

I've seen this with the green LED stuck on, and sometimes with the orange LED stuck on. Sometimes I've been able to slide the kbd and see the kbd backlight turn on/off, but nothing else would work. Any of these situations I would call SoD.

Please read the first post carefully. I changed it quite a bit, as we need a few things to change to debug this SoD issue. Please feel free to ask questions, but read that first post thoroughly - I think I included everything you will need in that post, if not I will modify it. Thanks!

Re: [TESTING] Updated 1/30/11 New Kernel (100% FIXED Panel Wake)
 

This is awesome, I almost feel like this new last_kmesg libc.so should have its own thread. Should be great for debugging all sorts of SoD, freezes, and reboots.

Re: [TESTING] Updated 1/30/11 New Kernel (100% FIXED Panel Wake)
 

I was going to make a new thread, but decided to just tack it onto this one...

Re: [TESTING] Updated 1/30/11 New Kernel (100% FIXED Panel Wake)
 

arrrghhh, I'm assuming this is for stock speed only, correct? If I OC and set pm.sleep_mode=2, I will get SOD.

Re: [TESTING] Updated 1/30/11 New Kernel (100% FIXED Panel Wake)
 

You should never be setting sleep_mode=2, assuming you're running a fairly new kernel with bzo's PLL2 commit...

sleep_mode=1 should basically always be used. I'm OC'd and sleep_mode=1, runnin just fine.

Re: [TESTING] Updated 1/30/11 New Kernel (100% FIXED Panel Wake)
 

Crap guys. F22 just pointed out that I'm dumb, and you don't need to do that libc crap.... there's a MUCH better method, as these are already logged and compressed in this folder:Has nothing to do with the service 'Dropbox' FYI ;).

So, going forward you guys don't need to do that libc.so mounting mess... Sorry for the confusion, I really had no idea it was already being logged and compressed.

Thanks again F22 for pointing this out!

Re: [TESTING] Updated 1/30/11 New Kernel (100% FIXED Panel Wake)
 

OK, I guess I mis-understood. The one I was talking about above is a SOD related to OCing but the one you're trying to debug is something else, related to panel waking up with this test kernel. I haven't had any SOD related to the panel wake; once in a while it goes into deep sleep (green LED), it may take me like 15-20 seconds to wake it up, repeatedly pushing the power button; I haven't had it gone dead sleep on me; how do you reproduce this?