Re: [PATCH v7 1/4] spinlock: A new lockref structure for locklessupdate of refcount

[Sedat Dilek]

On Sun, Sep 1, 2013 at 5:55 PM, Linus Torvalds
<torvalds@xxxxxxxxxxxxxxxxxxxx> wrote:
> On Sun, Sep 1, 2013 at 8:45 AM, Sedat Dilek <sedat.dilek@xxxxxxxxx> wrote:
>>
>> Samples: 160K of event 'cycles:pp', Event count (approx.): 77003901089
>> +  12,46%  t_lockref_from-  [kernel.kallsyms]     [k] irq_return
>> +   4,86%  t_lockref_from-  [kernel.kallsyms]     [k] lockref_get_or_lock
>> +   4,42%  t_lockref_from-  [kernel.kallsyms]     [k] __ticket_spin_lock
>> +   4,28%  t_lockref_from-  [kernel.kallsyms]     [k] __acct_update_integrals
>
> You need to go into __ticket_spin_lock to see who the callers are.
>
> Just go down to it and press enter to expand it (and then you need to
> go and expand that entry too to get the callers)
>

I am new to perf usage.

  4,60%  t_lockref_from-  [kernel.kallsyms]     [k] __ticket_spin_lock

Which entry to select?

Annotate __ticket_spin_lock
Zoom into t_lockref_from-(3962) thread
Zoom into the Kernel DSO
Browse map details
Run scripts for samples of thread [t_lockref_from-]
Run scripts for samples of symbol [__ticket_spin_lock]
Run scripts for all samples
Switch to another data file in PWD
Exit

> I still don't know how you get to irq_return. It should use sysret. Odd.

- Sedat -
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/