On 23/03/21 5:33 pm, Chris Packham wrote: > The "meat" of this series is in the last patch which is the change that > actually starts making use of the interrupts to drive a state machine. > The dt-bindings patches can probably go in at any time. The rest of the > series isn't dependent on them. > > I've tested it on a T2081 based system with a number of i2c and smbus > devices. Its the end of my work day so I figured I'd get this out now > but I'll do some more testing on a P2041 board and a few different i2c > devices tomorrow.
I've done more testing on a T2081 and P2041 board. Both look good. I've had some feedback from Rob on the dt-bindings which I think I've got sorted now. I've got a couple of minor cosmetic changes to 6/6 but I'll hold fire on sending a v2 to give people a chance to look at the functional changes. > Chris Packham (6): > dt-bindings: i2c-mpc: Document interrupt property as required > dt-bindings: i2c: convert i2c-mpc to json-schema > i2c: mpc: Make use of i2c_recover_bus() > i2c: mpc: make interrupt mandatory and remove polling code > i2c: mpc: use device managed APIs > i2c: mpc: Interrupt driven transfer > > .../devicetree/bindings/i2c/i2c-mpc.txt | 62 --- > .../devicetree/bindings/i2c/i2c-mpc.yaml | 99 ++++ > drivers/i2c/busses/i2c-mpc.c | 513 ++++++++++-------- > 3 files changed, 373 insertions(+), 301 deletions(-) > delete mode 100644 Documentation/devicetree/bindings/i2c/i2c-mpc.txt > create mode 100644 Documentation/devicetree/bindings/i2c/i2c-mpc.yaml >
