Re: [PATCH] trace: change CPU ring buffer state from tracing_cpumask

[Vaibhav Nagarnaik <vnagarnaik@google.com>]

On Wed, Jul 20, 2011 at 6:04 PM, Vaibhav Nagarnaik
<vnagarnaik@google.com> wrote:
> According to Documentation/trace/ftrace.txt:
>
> tracing_cpumask:
>
>        This is a mask that lets the user only trace
>        on specified CPUS. The format is a hex string
>        representing the CPUS.
>
> The tracing_cpumask currently doesn't affect the tracing state of
> per-CPU ring buffers.
>
> This patch enables/disables CPU recording as its corresponding bit in
> tracing_cpumask is set/unset.
>
> Signed-off-by: Vaibhav Nagarnaik <vnagarnaik@google.com>
> ---
>  kernel/trace/trace.c |    2 ++
>  1 files changed, 2 insertions(+), 0 deletions(-)
>
> diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c
> index e8aa324..d8ed798 100644
> --- a/kernel/trace/trace.c
> +++ b/kernel/trace/trace.c
> @@ -2433,10 +2433,12 @@ tracing_cpumask_write(struct file *filp, const char __user *ubuf,
>                if (cpumask_test_cpu(cpu, tracing_cpumask) &&
>                                !cpumask_test_cpu(cpu, tracing_cpumask_new)) {
>                        atomic_inc(&global_trace.data[cpu]->disabled);
> +                       ring_buffer_record_disable_cpu(global_trace.buffer, cpu);
>                }
>                if (!cpumask_test_cpu(cpu, tracing_cpumask) &&
>                                cpumask_test_cpu(cpu, tracing_cpumask_new)) {
>                        atomic_dec(&global_trace.data[cpu]->disabled);
> +                       ring_buffer_record_enable_cpu(global_trace.buffer, cpu);
>                }
>        }
>        arch_spin_unlock(&ftrace_max_lock);

Hi Steve

What do you think about this patch?

Vaibhav Nagarnaik