Re: [PATCH 5/5] I2C: DaVinci: initialize cmd_complete sooner

[Jean Delvare <khali-PUYAD+kWke1g9hUCZPvPmw@public.gmane.org>]

Hi Troy,

On Fri, 25 Apr 2008 09:58:14 -0700, Troy Kisky wrote:
> If an interrupt happens before an I2c master read/write,
> complete is called on uninitialized structure.

I'm curious why an interrupt would happen if no transaction has been
started. And even then, it seems to me that the interrupt handler
should be fixed to simply do nothing if it doesn't have anything to do.
Nevertheless...

> 
> Signed-off-by: Troy Kisky <troy.kisky-Q5RJGjKts06CY9SHAMCTRUEOCMrvLtNR@public.gmane.org>
> Signed-off-by: Kevin Hilman <khilman-Igf4POYTYCDQT0dZR+AlfA@public.gmane.org>
> ---
>  drivers/i2c/busses/i2c-davinci.c |    3 ++-
>  1 files changed, 2 insertions(+), 1 deletions(-)
> 
> diff --git a/drivers/i2c/busses/i2c-davinci.c b/drivers/i2c/busses/i2c-davinci.c
> index 6719cdb..7d8f1b0 100644
> --- a/drivers/i2c/busses/i2c-davinci.c
> +++ b/drivers/i2c/busses/i2c-davinci.c
> @@ -262,7 +262,7 @@ i2c_davinci_xfer_msg(struct i2c_adapter *adap, struct i2c_msg *msg, int stop)
>  
>  	davinci_i2c_write_reg(dev, DAVINCI_I2C_CNT_REG, dev->buf_len);
>  
> -	init_completion(&dev->cmd_complete);
> +	INIT_COMPLETION(dev->cmd_complete);
>  	dev->cmd_err = 0;
>  
>  	/* Take I2C out of reset, configure it as master and set the
> @@ -518,6 +518,7 @@ static int davinci_i2c_probe(struct platform_device *pdev)
>  		goto err_release_region;
>  	}
>  
> +	init_completion(&dev->cmd_complete);
>  	dev->dev = get_device(&pdev->dev);
>  	dev->irq = irq->start;
>  	platform_set_drvdata(pdev, dev);

... this patch is good to have from a performance point of view, so
I'll apply it.

-- 
Jean Delvare

_______________________________________________
i2c mailing list
i2c-GZX6beZjE8VD60Wz+7aTrA@public.gmane.org
http://lists.lm-sensors.org/mailman/listinfo/i2c