mbox series

[v2,0/2] i2c:imx: Deliver a timely stop on slave side, fix recv

Message ID 20211112133956.655179-1-minyard@acm.org
Headers show
Series i2c:imx: Deliver a timely stop on slave side, fix recv | expand

Message

Corey Minyard Nov. 12, 2021, 1:39 p.m. UTC 
I was working on an application that needs the stop condition
immediately.  So this adds a timer after each byte is received/sent and
if the bus is idle at the timeout, send the stop.

Also, I noticed when you use the i2c-slave-eeprom, if you read some data
and then read some data again, the last byte of the first read will be
the first byte of the second read.  This is because i2c-slave-eeprom
expects a read-ahead.  That's what the documentation says, at least.

Thanks to Uwe Kleine-König and Oleksij Rempel for reviewing.

-corey

Changes since v1:

* Added a comment on the hrtimer cancel on why the return value didn't
  need to be checked.

* Combine the hrtimer patch into the main timer patch.