Fehler #558
Reading touch calibration is very unreliable
0%
Beschreibung
No idea whats going on here.
Devices are not really usable.
Write a loop, make retain-code verbose.
Historie
Von Maximilian Seesslen vor 5 Monaten aktualisiert
Von When device is ok, a soft-reset can not reproduce the problem.
When device is ok, a warm-reset via reset-button can reproduce the problem.
Looks like a software issue (not bootup timing of SPI-Flash). The logs show proper readout of the flash/retain-counter.
Von Maximilian Seesslen vor 5 Monaten aktualisiert
The retain-header address is different. No delays needed.
Von Maximilian Seesslen vor 5 Monaten aktualisiert
The initial reading of JDEC-ID does not work and so the erase-page-size is the default 8.
Does the default value make sense at all? A flash should use a proper one.
The 8 could change and then retain does not fit any more. CFlashI2c does already set it to 8.
The reset-line is not routed to the flash chip. Looks like the STM32 itself has some bootup time issue.
JDEC-ID is already read twice.
Von Maximilian Seesslen vor 5 Monaten aktualisiert
Before closing this defect, check if the status of SPI HAL command is "OK". Add result as a comment in the code.
Von Maximilian Seesslen vor 4 Monaten aktualisiert
I2C has exactly the same issue on main switch.
Most times it works.
But some times after reset-button on debug-board, the device needed up to an second till the first byte could be read.
Even the waitbusy was returning ok, but then data could not get read.
Soft-reset had no problems once the device was running.
The I2C-EEPROM does not see the reset of the reset button on the debug board.
Quite different Chip.