RE: [PATCH] i2c: aspeed: Update the stop sw state when the bus recovry occurs

[Tommy Huang]

Hi Andi,

> -----Original Message-----
> From: Andi Shyti <andi.shyti@xxxxxxxxxx>
> Sent: Thursday, June 6, 2024 9:27 AM
> To: Tommy Huang <tommy_huang@xxxxxxxxxxxxxx>
> Cc: brendan.higgins@xxxxxxxxx; benh@xxxxxxxxxxxxxxxxxxx; joel@xxxxxxxxx;
> andrew@xxxxxxxxxxxxxxxxxxxx; wsa@xxxxxxxxxx; linux-i2c@xxxxxxxxxxxxxxx;
> openbmc@xxxxxxxxxxxxxxxx; linux-arm-kernel@xxxxxxxxxxxxxxxxxxx;
> linux-aspeed@xxxxxxxxxxxxxxxx; linux-kernel@xxxxxxxxxxxxxxx; BMC-SW
> <BMC-SW@xxxxxxxxxxxxxx>
> Subject: Re: [PATCH] i2c: aspeed: Update the stop sw state when the bus
> recovry occurs
> 
> Hi Tommy,
> 
> On Thu, May 30, 2024 at 03:06:56PM +0800, Tommy Huang wrote:
> > When the i2c bus recovey occurs, driver will send i2c stop command in
> > the scl low condition. In this case the sw state will still keep
> > original situation. Under multi-master usage, i2c bus recovery will be
> > called when i2c transfer timeout occurs. Update the stop command
> > calling with aspeed_i2c_do_stop function to update master_state.
> >
> > Fixes: f327c686d3ba ("i2c: aspeed: added driver for Aspeed I2C")
> >
> > Signed-off-by: Tommy Huang <tommy_huang@xxxxxxxxxxxxxx>
> 
> Can you please add:
> 
> Cc: <stable@xxxxxxxxxxxxxxx> # v4.13+

Got it. I will add it.

> 
> > ---
> >  drivers/i2c/busses/i2c-aspeed.c | 3 ++-
> >  1 file changed, 2 insertions(+), 1 deletion(-)
> >
> > diff --git a/drivers/i2c/busses/i2c-aspeed.c
> > b/drivers/i2c/busses/i2c-aspeed.c index ce8c4846b7fa..32f8b0c1c174
> > 100644
> > --- a/drivers/i2c/busses/i2c-aspeed.c
> > +++ b/drivers/i2c/busses/i2c-aspeed.c
> > @@ -169,6 +169,7 @@ struct aspeed_i2c_bus {  };
> >
> >  static int aspeed_i2c_reset(struct aspeed_i2c_bus *bus);
> > +static void aspeed_i2c_do_stop(struct aspeed_i2c_bus *bus);
> 
> Can you please move aspeed_i2c_do_stop() on top? Doesn't make much sense
> to add the prototype here as there is no dependencies.

Sure. I will update it.

> 
> It's different the case of aspeed_i2c_reset() because it needs aspeed_i2c_init().
> 
> >  static int aspeed_i2c_recover_bus(struct aspeed_i2c_bus *bus)  { @@
> > -187,7 +188,7 @@ static int aspeed_i2c_recover_bus(struct aspeed_i2c_bus
> *bus)
> >  			command);
> >
> >  		reinit_completion(&bus->cmd_complete);
> > -		writel(ASPEED_I2CD_M_STOP_CMD, bus->base +
> ASPEED_I2C_CMD_REG);
> > +		aspeed_i2c_do_stop(bus);
> 
> The patch is good, though!
> 

Thanks for your commects.

> Thanks,
> Andi