Re: [PATCH v4 2/3] tracing: Add trace_irqsoff tracepoints

[Daniel Wagner]

Hi Binoy,

On 08/30/2016 12:28 PM, Binoy Jayan wrote:
> +static inline void trace_latency_preempt_mark_ts(enum latency_type ltype)
> +{
> +	this_cpu_write(lat_ts[ltype], (cycle_t) trace_clock_local());
> +}
> +
> +static inline void latency_trace(enum latency_type type)
> +{
> +	trace_latency_preempt(type,
> +		(cycle_t) trace_clock_local() - this_cpu_read(lat_ts[type]));
> +}

As Masami has pointed out, the prefix trace_ should not be used. Also 
having trace_latency_ and latency_trace_ is kind of confusing. What 
about {start|stop}_latency_timing()? It would match the existing 
{start|stop}_critical_timing(). Or is it too close and it leads to 
confusion?

Another idea is {start|stop}_latency_preempt(). This matches the 
trace_latency_preempt_enable() function.

cheers,
daniel