Commit graph

6 commits

Author SHA1 Message Date
Simon Arlott
b829a1d264
Avoid 8-bit timer overflows in debounce algorithms (#12240)
* Add fast_timer_t that is 16-bit or 32-bit based on architecture

A 16-bit timer will overflow sooner but be faster to compare on AVR.

* Avoid 8-bit timer overflows in debounce algorithms

Count down remaining elapsed time instead of trying to do 8-bit timer
comparisons.

Add a "none" implementation that is automatically used if DEBOUNCE is
0 otherwise it will break the _pk/_pr count down.

* Avoid unnecessary polling of the entire matrix in sym_eager_pk

The matrix only needs to be updated when a debounce timer expires.

* Avoid unnecessary polling of the entire matrix in sym_eager_pr

The matrix only needs to be updated when a debounce timer expires.

The use of the "needed_update" variable is trying to do what
"matrix_need_update" was added to fix but didn't work because it only
applied when all keys finished debouncing.

* Fix sym_defer_g timing inconsistency compared to other debounce algorithms

DEBOUNCE=5 should process the key after 5ms, not 6ms

* Add debounce tests
2021-06-09 17:23:21 +10:00
Alex Ong
b5b119544a Removed #include "matrix.h" from debounce.h 2019-01-26 22:44:33 +11:00
alex-ong
5b7fc758d7 Restored debounce.h since there wasnt any real change. 2019-01-26 17:46:27 +11:00
alex-ong
123608fb31 DO NOT USE Revert back to original API to support split_keyboards. 2019-01-26 17:10:27 +11:00
alex-ong
d0b691df0e DO NOT USE - debounce successfully compiled. 2019-01-26 17:03:46 +11:00
James Churchill
28929ad017 Simplify split_common Code significantly (#4772)
* Eliminate separate slave loop

Both master and slave run the standard keyboard_task main loop now.

* Refactor i2c/serial specific code

Simplify some of the preprocessor mess by using common function names.

* Fix missing #endif

* Move direct pin mapping support from miniaxe to split_common

For boards with more pins than sense--sorry, switches.

* Reordering and reformatting only

* Don't run matrix_scan_quantum on slave side

* Clean up the offset/slaveOffset calculations

* Cut undebounced matrix size in half

* Refactor debouncing

* Minor fixups

* Split split_common transport and debounce code into their own files

Can now be replaced with custom versions per keyboard using
CUSTOM_TRANSPORT = yes and CUSTOM_DEBOUNCE = yes

* Refactor debounce for non-split keyboards too

* Update handwired/xealous to build using new split_common

* Fix debounce breaking basic test

* Dodgy method to allow a split kb to only include one of i2c/serial

SPLIT_TRANSPORT = serial or SPLIT_TRANSPORT = i2c will include only
that driver code in the binary.

SPLIT_TRANSPORT = custom (or anything else) will include neither, the
keyboard must supply it's own code

if SPLIT_TRANSPORT is not defined then the original behaviour (include
both avr i2c and serial code) is maintained.

This could be better but it would require explicitly updating all the
existing split keyboards.

* Enable LTO to get lets_split/sockets under the line

* Add docs for SPLIT_TRANSPORT, CUSTOM_MATRIX, CUSTOM_DEBOUNCE

* Remove avr-specific sei() from split matrix_setup

Not needed now that slave doesn't have a separate main loop.
Both sides (on avr) call sei() in lufa's main() after exiting
keyboard_setup().

* Fix QUANTUM_LIB_SRC references and simplify SPLIT_TRANSPORT.

* Add comments and fix formatting.
2019-01-17 10:08:14 -08:00