Was thinking to do some experimenting with the N9 Lockscreen (Lockcode enabled), my idea was to get NFC working, to unlock the phone.
Doh! I cant even enter a lockcode in the settings (no lockcode ATM set). The device keeps telling me 'wrong lock code' when I enter a code (and yes, I do enter the code exactly the same in the confirmation field).
May be this is related to having flashed the open-mode kernel!
EDIT: Yes, the problem was Openmode kernel, reflash stock, and the problem is no more. Oh well, bettter choice to have open-mode
Yes, the problem was Openmode kernel, reflash stock, and the problem is no more. Oh well, bettter choice to have open-mode
I ran into the same problem just yesterday.
I have been using openmode kernel since the beginning when I got my new N9, which is about a month back now.
Immediately when I got the phone, the first thing I did was to reflash it clean with PR1.2 and open kernel. Just about everything I have done works correctly except for the creation of the security lock code.
I was under the impression that it should work if the phone is not used in normal mode before flashing open kernel into it.
Now, how to fix this?
I guess I should backup my current installed apps, settings & everything, then reflash it with open mode kernel and some kind of super-clearing option and finally restore backups. Is that workable solution?
Openmode kernel has CAL area locked always. So currently only way to get lock code working is _not_ to use Openmode kernel.
OK, I see.
That is quite a limitation, though. So, does device locking/unlocking require some HW property that resides on CAL area or what is the reason it cannot be mapped elsewhere in memory?
I assume it could be possible to write own device locking that could be used instead of the default/real device locking.
For most purposes locking anyway is just to prevent accidental or incompetent tampering with the device.
It can easily be mapped outside from CAL area, for example into aegis-fs dir. But nobody has just released such a devlock yet. Hopefully someone will do sometime in the future.
It can easily be mapped outside from CAL area, for example into aegis-fs dir. But nobody has just released such a devlock yet. Hopefully someone will do sometime in the future.
Well yes, I would like to have it, so I might as well do something about it. I haven't really done any development on harmattan yet so this is quite new for me but I do have a solid unix/C background.
Besides, I propably should have some devproject outside my work area anyways
Do you have any useful information on the existing locking mechanism/framewok that could serve as a starting point?
Well yes, I would like to have it, so I might as well do something about it. I haven't really done any development on harmattan yet so this is quite new for me but I do have a solid unix/C background.
Besides, I propably should have some devproject outside my work area anyways
Do you have any useful information on the existing locking mechanism/framewok that could serve as a starting point?
I know that someone has already done a version which doesn't use CAL at all, but he might be waiting for next possible pr-release before releasing, as I haven't seen that package in public yet.
I know that someone has already done a version which doesn't use CAL at all, but he might be waiting for next possible pr-release before releasing, as I haven't seen that package in public yet.
Allright, some time has passed and we all have PR1.3 on our phones, right?
Do you know where might I get the non-CAL-area compatible lockcode solution, or take part in creating one if it does not exist yet?
On another note, if it is possible to replace the existing lockcode mechanism I would like to have an option of using a gesture pattern lock, similar as I have seen on some android devices.
I guess it would not be difficult to implement on top of the open lockcode solution.
