storage: Add the I2C EEPROM block device #3564
Conversation
geky
changed the title
|
I think this PR is currently blocked pending resolution of the #3671 review. |
| for (int i = 0; i < I2CEEPROM_TIMEOUT; i++) { | ||
| _i2c.start(); | ||
| int ack = _i2c.write(_i2c_address | 0); | ||
| _i2c.stop(); |
There was a problem hiding this comment.
_i2c.write(_i2c_address | 0, 0, 0) could be used instead here.
In case of start / write(byte), there is no way for the driver to know whether a another byte will be written next.
With the write(address, buf, length) API, this is clear that this is a check for availability of the device (length is 0).
We need this info for some of the stm32 device to have the check working ok..
There was a problem hiding this comment.
Hi @LMESTM, thanks for pointing this out.
I think I may be misunderstanding a part of the I2C protocol. Doesn't every byte that is sent over I2C get an ack?
|
marking as do not merge until dependent PR is merged |
|
@LMESTM are you happy with this now ? |
|
This has been moved out to here: Feel free to raise any remaining issues on that repo, thanks 👍 |
Adds support for I2CEepromBlockDevice, a block device driver to support I2C EEPROMs like Microchip's 24LC range and AMTELS AT24C range.
This is a fork of the I2CEeprom driver by Robin Hourahane on the developer site:
https://developer.mbed.org/users/rhourahane/code/I2CEeprom/
This provides support for small I2C EEPROM devices around the 32k byte range. Note, these chips may be too small to support the FATFileSystem which requires a minimum of 64k bytes.
This driver should follow the life-cycle of the SDCardBlockDevice, and move into its own repo before this feature-branch is merged.
dependent on #3671
relevant commit range 3440cd7...geky:bd-i2ceeprom
cc @simonqhughes, @c1728p9, @theotherjimmy