diff options
| author | yiancar <yiangosyiangou@cytanet.com.cy> | 2019-11-25 20:33:52 +0000 |
|---|---|---|
| committer | Joel Challis <git@zvecr.com> | 2019-11-25 20:33:52 +0000 |
| commit | c0fe8dbfb4ea7b36cc2c5ba65d943c2cbec84244 (patch) | |
| tree | 3fe648b223538ce54ad1f322f760a8610bd7a388 /autocomplete.sh | |
| parent | f0f161e5724375b7a289699703d86d6de2adae8d (diff) | |
| download | qmk_firmware-c0fe8dbfb4ea7b36cc2c5ba65d943c2cbec84244.tar.gz qmk_firmware-c0fe8dbfb4ea7b36cc2c5ba65d943c2cbec84244.zip | |
IS31FL3733 Dirty page fix (#7079)
* IS31FL3733 Dirty page fix
Function IS31FL3733_update_led_control_registers was never setting update register to false. As a result the led on/off page was being written every transaction even when it was not modified. This is ineficient and causes lots of bandwidth use.
-> Fix the IS31FL3733_update_led_control_registers.
-> After testing it was evident that failed I2C transactions could corrupt the Led on/off register.
-> Update IS31FL3733_write_pwm_buffer and IS31FL3733_write_register functions to return 0 upon succesful tranmission and 1 if any of the transmitions within the function fail.
-> Modify IS31FL3733_update_pwm_buffers function so if any of the IS31FL3733_write_pwm_buffer transuction fails, the g_led_control_registers_update_required register is set to true forcing a rewrite of the led on/off register in case it was corrupted.
* Minor comment update
* Upsie:)
* Update is31fl3733.c
* Return fix
* more return fix
* type change
* more boolian logic reversal:)
Diffstat (limited to 'autocomplete.sh')
0 files changed, 0 insertions, 0 deletions