As pointed out by davidefer, the lookahead pointer modular arithmetic
does not work around integer overflow when the pointer size is not a
multiple of the block count.
To avoid overflow problems, the easy solution is to stop trying to
work around integer overflows and keep the lookahead offset inside the
block device. To make this work, the ack was modified into a resetable
counter that is decremented every block allocation.
As a plus, quite a bit of the allocation logic ended up simplified.
One of the big simplifications in littlefs's implementation is the
complete lack of tracking free blocks, allowing operations to simply
drop blocks that are no longer in use.
However, this means the lookahead buffer can easily contain outdated
blocks that were previously deleted. This is usually fine, as littlefs
will rescan the storage if it can't find a free block in the lookahead
buffer, but after changes that caused littlefs to more conservatively
respect the alloc acks (e611cf5), any scanned blocks after an ack would
be incorrectly trusted.
The fix is to eagerly scan ahead in the lookahead when we allocate so
that alloc acks are better able to discredit old lookahead blocks. Since
usually alloc acks are tightly coupled to allocations of one or two blocks,
this allows littlefs to properly rescan every set of allocations.
This may still be a concern if there is a long series of worn out
blocks, but in the worst case littlefs will conservatively avoid using
blocks it's not sure about.
Found by davidefer
Tests have shown that there is hich-up on MSI clock during the setup phase.
If this stabilization phase happens when application has restarted again
this can have side effects, like grambled UART characters typically.
So we're adding a delay before hading-over back to application.
With this modification, on NCULEO_L476RG, the wake-up time is increased
from 2ms to 2,5ms.
If possible this should be improved in the future to save 500 microseconds
of wak-up time. See TODO
There are cases where HW registers are found in unpexcepted state when
exiting Deep Sleep only few micro-seconds after it was entered.
By using an internal clock that does not depend on anythin and clocking
the system without using PLL, this allows SetSysClock default configuration
to run fine whatever possible configuration we find the HW in when
exiting Deep Sleep.
Also we shall restore interrupts only after all cloks are back to
expected running state.
Christopher Haster
Cruz Monrreal
Martin Kojtal
Netanel Gonen
Jimmy Brisson
Kevin Bracey
Teemu Kultala
Antti Kauppila
Laurent MEUNIER
Thomas Sailer
Thomas Sailer