[Q] KERNEL PANICS - what are the relations and causes? [not a simple question] - Optimus One, P500, V General

I have made a lot of searches and read a lot about battery calibration... and of course i know what the hell is kernel panic, its definition and its symptoms.
However I still have some confusions and I hope some of you may be able to solve them
Maybe some of them are noobie questions but I honestly sorry for that also this may help others who has these confusions too
My situation:
The system stats are shown in my signature.
on 5/9 I have flashed franco.Kernel v19.4 bfs and then started a lot of OC tests,
finally I was stable on stress test (SetCPU) 729mhz for 2 hours non-stop,
and also I used the phone stably for over 24 hours without reboot.
However I faced a problem which is related to Darktremor scripts.
I used 2 weeks to solve it and finally it is gone now.
I started battery calibration which is the final process before I use my phone normally in my daily lives.
But in these two days, my kernel panics after ~4 hours of usage of 729mhz max CPU frequency.
without the need of advice, surely i have downclocked to 710mhz max and its all good. but however i am super curious about why 729 becomes from stable to unstable before and after.
1) Does battery stats related to kernel panics in ANY extent? if yes, how?
Ignore all questions under 1 if the answer is No.
1.1) Does wiping battery stats before flashing a kernel helps you OC higher?
1.2) Does wiping battery stats before flashing a ROM helps you OC higher?
1.3) Does wiping battery stats (for calibration purposes) has possibilities to produce kernel panics?​
2) Why is there a sudden drop of battery level (~4-5%) after kernel panics?
3) Is there any other possible causes of kernel panics BESIDES overclocking too high?
3.1) *2sd scripts?
3.2) ext4 barrier = 0?
3.3) inputting EXT4 data writeback?
(tune2fs -o journal_data_writeback /dev/block/mmcblk0p2)​
4) Is SetCPU the problem?
Some says "dont use SetCPU when you can overclock using CM7 settings."
Is that true?​
5) BESIDES the max speed, does it also depends on Governors?
Ignore powersave governor or others which forces the clock to the minimum frequencies.
So, is there any difference between setting the same maximum speed, different governors?
(for example) Would smartass gives a panic at 729mhz but ondemand doesnt?​
6) Does the "change of frequencies" gives easier kernel panics?
Situation A: 729min, 729 max
Situation B: 480min, 729max
In situation B, the system makes the frequencies change due to load, means it may switch from 480 -> 600 -> 729 -> 710 -> 691 -> 480 etc.
so does this "change of frequencies" makes the kernel loads more and finally gets it to panic?​
Thanks for your time reading and answering my questions!
Please just quote one of them if you would just like to answer one of the above
Thank you very much!

---

1) no.
2) It happens when u remove the battery and put it back.
3) There are some strange reports of kernel panic at 600Mhz which are weird. Sometimes Roms/kernel play a part(unknown).
4) it is because you can overclock in Cm7 settings menu, so no need to install another app such as SetCpu
5) Maybe, a governor who stays more on the highest frequencies i think are easy to panic because hardware can't handle it.
6) for me, i would say situation A would be easier to kernel panic, as it is constantly on the highest frequency and stressing your cpu too much that it can't handle it.
Since you get stable at 729mhz for 2h and panicked when 4h of use, that means the cpu can't handle 729mhz for above 4h.

---

@terratrix
thanks for your reply!! at least im not using my stupidity to wipe battery stats and trying to avoid kernel panics, lol.
@info update and conclusion after 1 more day of testing
And there, after some readings on the forum and the experiences myself, i think i have found the critical point already - QUESTION NO.5
Conclusion : kernel panics are related to governors, at least, our new amazing profiled one - SmartassV2!!
On the thread of Franco's .35 v1 kernel, there are some users reporting the panic issues faced with ONLY SmartassV2.
And me myself, has joined the party as well
Smartass(v1) has been my best bud all the time at 729mhz or even higher for more than 6 months. And when I left it and went to choose his next generation, my phone got a generation gap with Mr.V2, lol.
On my phone: (Franco.kernel 19.4bfs and no any reflash or stuffs between testings)
SmartassV2 ---
very unstable panics AND random shutdowns at random times
729mhz - sometimes 2 hours, sometimes 4 hours
710mhz - yeah I justnow kernel paniced with this also, you know how long after boot? 30minutes!!!
Why 30minutes? I suspect that it.is related to franco's network speed hack. I'm on EDGE data plan and because of Franco, I can enable HSDPA with 4x more speed.
On 710mhz, my phone was stable for over 8 hours with EDGE on (I intended to save battery), but after a reboot (manual one for some other purposes) I started to use (the hacked) HSDPA. After 30 minutes, panic. I remembered when I was at 729 on smartassv2 I panicked after 3 hours and at that time I was using HSDPA too.
Smartass (v1) ---
Well now I'm on smartass v1 at 729mhz stably typing this message. Funny huh...?
Results:
Franco's network speed hack
OR
3G network/HSDPA network
OR
Even just because of specific phone hardwares
IS NOT COMPATIBLE WITH THE ALMIGHTY SMARTASS-V2 GOVERNOR AND WOULD RESULT IN KERNEL PANICS AT THE FREQUENCIES WHERE THE PHONE CAN BE STABLE WITH OTHER COMMON GOVERNORS.
If I am wrong or you can confirm this by testing yourself, please let me know!!! thanks and hope I have provided some useful information
Goodnight and I'm going to continue working with my old friend smartass-v1 despite v2 has better profiling tweaks, lol
Sent from my LG-P500 using XDA Premium App

---

Dude, as i had mentioned earlier in franco's thread, that it had to do with the governors... cause i usually try and find out the cause of the issue myself before posting a bug..
SO all i tried were playing around with the performance settings... and i start off with changing the governors... the best targeted experimental ones are MINMAX (my personal fav), SMARTASS, SMOOTHASS (now removed but one AWESOME governor), and INTERACTIVE-X (rewritten by franco i guess).... About the frequencies... from what i have observed so far apart from reading stuffs is that, at 729 Mhz our optimus DOES pretty much stay stable even in the ****TIEST devices.. LOL i can say that cause we indians get the crappiest hardware of all... the point is... if you got a kernel panic at 729, it is most likely that you would get it again at 600 Mhz too cause its not overclock related..!!!
Bottom-line: Avoid SmartassV2 unless your phone is made of the finest hardware i.e. you can easily overclock it to 806 Mhz (ive seen some users using it at this frequency without any issues at all .. :O ) Use Smartass older or Minmax instead...!! And avoid overclocking above 729..!!!
P.s.: mine works fine even at 768 BTW....

---

We have to wait for Franco's next release for more smartass2 updates
Sent from my LG-P500

---

SmartassV2 is kind of strange, at least in my phone. I notice my battery draining faster while I use it, and it's probably because it stays in the max frequency longer for some reason, despite screen off. I use NoFrills for my CPU setting and it has the ability to view stats, and when I'm using V2 it stays longer at the max than at the min (65 vs 35%). The opposite is true for Smartass v1, which to me makes more sense because my phone stays at standby more often than being used.
As comparison data, my phone with smartass v1 can typically be OCed up to 787 MHz. 748 MHz on v2 is stable, but I haven't extensively tested higher freqs.

---

I have a Indian phone , smartass v2 oc to 768 stable xD
Sent from my LG-P500