[PATCH linux dev-4.13 05/16] fsi: gpio: Trace busy count

[Eddie James]

On 02/19/2018 10:18 PM, Andrew Jeffery wrote:
> An observation from trace output of the existing FSI tracepoints was that the
> remote device was sometimes reporting as busy. Add a new tracepoint reporting
> the busy count in order to get a better grip on how often this is the case,

Acked-by: Eddie James <eajames at linux.vnet.ibm.com>

>
> Signed-off-by: Andrew Jeffery <andrew at aj.id.au>
> ---
>   drivers/fsi/fsi-master-gpio.c          |  3 +++
>   include/trace/events/fsi_master_gpio.h | 16 ++++++++++++++++
>   2 files changed, 19 insertions(+)
>
> diff --git a/drivers/fsi/fsi-master-gpio.c b/drivers/fsi/fsi-master-gpio.c
> index 3f487449a277..2a49b167effe 100644
> --- a/drivers/fsi/fsi-master-gpio.c
> +++ b/drivers/fsi/fsi-master-gpio.c
> @@ -401,6 +401,9 @@ static int poll_for_response(struct fsi_master_gpio *master,
>   		break;
>   	}
>
> +	if (busy_count > 0)
> +		trace_fsi_master_gpio_poll_response_busy(master, busy_count);
> +
>   	/* Clock the slave enough to be ready for next operation */
>   	clock_zeros(master, FSI_GPIO_PRIME_SLAVE_CLOCKS);
>   	return rc;
> diff --git a/include/trace/events/fsi_master_gpio.h b/include/trace/events/fsi_master_gpio.h
> index 11b36c119048..48e83e5755f4 100644
> --- a/include/trace/events/fsi_master_gpio.h
> +++ b/include/trace/events/fsi_master_gpio.h
> @@ -63,6 +63,22 @@ TRACE_EVENT(fsi_master_gpio_break,
>   	)
>   );
>
> +
> +TRACE_EVENT(fsi_master_gpio_poll_response_busy,
> +	TP_PROTO(const struct fsi_master_gpio *master, int busy),
> +	TP_ARGS(master, busy),
> +	TP_STRUCT__entry(
> +		__field(int,	master_idx)
> +		__field(int,	busy)
> +	),
> +	TP_fast_assign(
> +		__entry->master_idx = master->master.idx;
> +		__entry->busy = busy;
> +	),
> +	TP_printk("fsi-gpio%d: device reported busy %d times",
> +		__entry->master_idx, __entry->busy)
> +);
> +
>   #endif /* _TRACE_FSI_MASTER_GPIO_H */
>
>   #include <trace/define_trace.h>