Custom matrix lite support for split keyboards

[jaygreco]

Description

Until recently, split boards with a custom matrix needed to use CUSTOM_MATRIX = yes rather than the simpler lite option.
The new unified split common + regular matrix introduced in #13330 is really close to supporting CUSTOM_MATRIX = lite. If the debounce code is omitted from quantum/matrix_common/matrix_scan() when the SPLIT_KEYBOARD and CUSTOM_MATRIX options are set, the lite option can be used.

Tested with split board that will be added in a follow-on PR and also on a NIBBLE, which uses CUSTOM_MATRIX = lite but is not a split board in order to make sure the changes didn't break existing non-split boards.

Types of Changes

• Core
• Bugfix
• New feature
• Enhancement/optimization
• Keyboard (addition or update)
• Keymap/layout/userspace (addition or update)
• Documentation

Issues Fixed or Closed by This PR

Checklist

• My code follows the code style of this project: C, Python
• I have read the PR Checklist document and have made the appropriate changes.
• My change requires a change to the documentation.
• I have updated the documentation accordingly.
• I have read the CONTRIBUTING document.
• I have added tests to cover my changes.
• I have tested the changes and verified that they work and don't break anything (as well as I can manage).

[drashna]

seems to compile as is.

Also, it may make more sense to move even more of the split code into matrix_common, so that matrix.c is mostly just handling the matrix read, and matrix_common is doing all the heavy lifting?

[jaygreco]

Great suggestion. I agree. I was considering that but was apprehensive about moving too much functionality across without discussing it. I'll make and test that change & update the PR.

[jaygreco]

Alright, I tested the changes as suggested. The order of the debounce matters for split keyboards. The debounce routine needs to be run before the slave matrix is read in, or it will double-debounce and toggle held keys each read, which is what I'm seeing.

quantum/matrix.c:362

#ifdef SPLIT_KEYBOARD
    debounce(raw_matrix, matrix + thisHand, ROWS_PER_HAND, changed);
    changed = (changed || matrix_post_scan());
#else
    debounce(raw_matrix, matrix, ROWS_PER_HAND, changed);
    matrix_scan_quantum();
#endif

Any thoughts?

[jaygreco]

@drashna I pushed a commit that takes a hack at this. If nothing else, it's a starting point. It adds the debouncing in the right sequence. It also adds a new weak function prototype for matrix_post_scan_custom. A question that's probably worth asking is if it's worth it (and/or appropriate) to move the standard matrix_post_scan over to matrix_common entirely, since right now my matrix_post_scan_custom is just a carbon copy of that code.

I also made a gist that has the meat of my custom_matrix function just to add some context: https://gist.github.com/jaygreco/56319ffc5b15a567f125fabfff796649

[jaygreco]

I don't love doing this with an #ifdef, but IMO doing something like setting thisHand to 0 and passing that for a non-split seems like it would be confusing.

[drashna]

The post scan function should be probably moved wholesale, without any custom option, as its sole purpose is to handle the transport code, and doesn't actually touch the matrix, directly.

[jaygreco]

Shuffled things to make matrix_post_scan live in matrix_common, which cleans things up considerably on the keyboard side.

[drashna]

Merge conflicts.

[jaygreco]

Checking in on this (I thought it had been merged before the breaking changes but was mistaken). If I resolve the conflicts, can it be merged into the dev branch?

[drashna]

This is cause errors during compile, on ARM split boards.

build_keyboard.mk:461: *** [.build/obj_handwired_tractyl_manuform_5x6_right_f411_drashna/quantum/matrix_common.o] error 1

#0  .build/obj_handwired_tractyl_manuform_5x6_right_f411_drashna/quantum/process_keycode/process_tap_dance.o at /Users/drashna/repos/qmk_firmware/build_keyboard.mk:457
#1  .build/handwired_tractyl_manuform_5x6_right_f411_drashna.elf at /Users/drashna/repos/qmk_firmware/tmk_core/rules.mk:362
#2  elf at /Users/drashna/repos/qmk_firmware/tmk_core/rules.mk:283
#3  build at /Users/drashna/repos/qmk_firmware/build_keyboard.mk:445
#4  all at /Users/drashna/repos/qmk_firmware/build_keyboard.mk:439
Compiling: quantum/matrix_common.c                                                                 In file included from quantum/logging/debug.h:21,
                 from quantum/keymap.h:34,
                 from quantum/quantum.h:21,
                 from quantum/matrix_common.c:1:
quantum/matrix_common.c: In function 'matrix_print':
quantum/logging/print.h:85:39: error: unknown conversion type character 'b' in format [-Werror=format=]
   85 | #define print_bin_reverse8(i) xprintf("%08b", bitrev(i))
      |                                       ^~~~~~
quantum/matrix_common.c:63:35: note: in expansion of macro 'print_bin_reverse8'
   63 | #    define print_matrix_row(row) print_bin_reverse8(matrix_get_row(row))
      |                                   ^~~~~~~~~~~~~~~~~~
quantum/matrix_common.c:81:9: note: in expansion of macro 'print_matrix_row'
   81 |         print_matrix_row(row);
      |         ^~~~~~~~~~~~~~~~
quantum/logging/print.h:85:43: note: format string is defined here
   85 | #define print_bin_reverse8(i) xprintf("%08b", bitrev(i))
      |                                           ^
quantum/logging/print.h:85:39: error: too many arguments for format [-Werror=format-extra-args]
   85 | #define print_bin_reverse8(i) xprintf("%08b", bitrev(i))
      |                                       ^~~~~~
quantum/matrix_common.c:63:35: note: in expansion of macro 'print_bin_reverse8'
   63 | #    define print_matrix_row(row) print_bin_reverse8(matrix_get_row(row))
      |                                   ^~~~~~~~~~~~~~~~~~
quantum/matrix_common.c:81:9: note: in expansion of macro 'print_matrix_row'
   81 |         print_matrix_row(row);
      |         ^~~~~~~~~~~~~~~~
cc1: all warnings being treated as errors
 [ERRORS]
`