M24C16-RMN6TP in approved EEPROM list, but addressing appears incorrect

[gsuberland]

The M24C16-RMN6TP is in the list of approved EEPROMs here.

The datasheet shows that the M24C16-R series of EEPROMs uses the address format 0b1010xxxx, with the bottom bits used for page select and R/W.

This means that the base address of this EEPROM series is 0xA0.

However, the badge firmware uses 0x50 or 0x57 for addressing EEPROMs:

badge-2024-software/modules/system/hexpansion/util.py

Lines 9 to 15 in e8d716a

	def detect_eeprom_addr(i2c):
	devices = i2c.scan()
	if 0x57 in devices and 0x50 not in devices:
	return 0x57
	if 0x50 in devices:
	return 0x50
	return None

As such, unless I am missing something critical, it is not possible for any hexpansions which use the M24C16-R series EEPROMs to be detected and read from with the current firmware.

[BartDeWaal]

The rightmost bit is R/W, so it doesn't count for the address. So you need to shift the address right by 1.

That means the address is 1010000, so 0x50. This is correct.

[hairymnstr]

There are certainly issues with this EEPROM which need addressing, see #129. However, I am certain that addressing is not the issue so I'm going to close this particular issue.

[gsuberland]

No worries, it does seem that the address thing was a red herring.