lkml.org 
[lkml]   [2009]   [Mar]   [12]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
Date
From
SubjectRe: [Patch 2/2] tracepoints for softirq entry/exit - tracepoints
On Thu, Mar 12, 2009 at 02:36:03PM -0400, Jason Baron wrote:
>
> introduce softirq entry/exit tracepoints. These are useful for
> augmenting existing tracers, and to figure out softirq frequencies and
> timings.
>
> Signed-off-by: Jason Baron <jbaron@redhat.com>
>
> ---
>
> include/trace/irq_event_types.h | 12 ++++++++++++
> kernel/softirq.c | 7 ++++++-
> 2 files changed, 18 insertions(+), 1 deletions(-)
>
>
> diff --git a/include/trace/irq_event_types.h b/include/trace/irq_event_types.h
> index 214bb92..38b4bdd 100644
> --- a/include/trace/irq_event_types.h
> +++ b/include/trace/irq_event_types.h
> @@ -40,4 +40,16 @@ TRACE_EVENT(irq_handler_exit,
> __entry->irq, __entry->ret ? "handled" : "unhandled")
> );
>
> +TRACE_FORMAT(irq_softirq_entry,
> + TP_PROTO(struct softirq_action *h, struct softirq_action *vec),
> + TP_ARGS(h, vec),
> + TP_FMT("softirq=%d action=%s", h - vec, softirq_to_name[h-vec])
> + );
> +
> +TRACE_FORMAT(irq_softirq_exit,
> + TP_PROTO(struct softirq_action *h, struct softirq_action *vec),
> + TP_ARGS(h, vec),
> + TP_FMT("softirq=%d action=%s", h - vec, softirq_to_name[h-vec])
> + );
> +
> #undef TRACE_SYSTEM
> diff --git a/kernel/softirq.c b/kernel/softirq.c
> index ba1511f..c378d53 100644
> --- a/kernel/softirq.c
> +++ b/kernel/softirq.c
> @@ -24,6 +24,7 @@
> #include <linux/ftrace.h>
> #include <linux/smp.h>
> #include <linux/tick.h>
> +#include <trace/irq.h>
>
> #include <asm/irq.h>
> /*
> @@ -186,6 +187,9 @@ EXPORT_SYMBOL(local_bh_enable_ip);
> */
> #define MAX_SOFTIRQ_RESTART 10
>
> +DEFINE_TRACE(irq_softirq_entry);
> +DEFINE_TRACE(irq_softirq_exit);


Just one nit here. The "irq_" prefix seems to me too much.
On the trace we have:

/* irq_softirq_entry: softirq=nb action=nb_to_logical_name */

It's even too much words that says the same things.
Moreover, we have the logical name, the number seems not very useful
because we have its logical translation just after.

I would suggest to have just:

/* softirq_entry: nb_to_logical_name */

ie:

/* softirq_entry: SCHED_SOFTIRQ */

Don't you think it's more clear and obvious?

Other than that, I think these tracepoints are a good idea.


> asmlinkage void __do_softirq(void)
> {
> struct softirq_action *h;
> @@ -212,8 +216,9 @@ restart:
> if (pending & 1) {
> int prev_count = preempt_count();
>
> + trace_irq_softirq_entry(h, softirq_vec);
> h->action(h);
> -
> + trace_irq_softirq_exit(h, softirq_vec);
> if (unlikely(prev_count != preempt_count())) {
> printk(KERN_ERR "huh, entered softirq %td %s %p"
> "with preempt_count %08x,"



\
 
 \ /
  Last update: 2009-03-12 20:05    [W:0.141 / U:0.340 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site