* [QUESTION] trace_function_graph: consumed event may be invalid
@ 2009-06-30 8:14 Lai Jiangshan
2009-06-30 10:54 ` Frederic Weisbecker
0 siblings, 1 reply; 2+ messages in thread
From: Lai Jiangshan @ 2009-06-30 8:14 UTC (permalink / raw)
To: Frederic Weisbecker; +Cc: Steven Rostedt, LKML
In get_return_for_leaf() {
......
ring_buffer_consume()
......
access to @curr->ent.pid
}
@curr is saved at consumed event, if this event is the last event
in the page, the ring_buffer may reuse the page. It will become invalid.
Is there any other code to ensure it valid when we `cat trace_pipe`?
Lai.
^ permalink raw reply [flat|nested] 2+ messages in thread* Re: [QUESTION] trace_function_graph: consumed event may be invalid
2009-06-30 8:14 [QUESTION] trace_function_graph: consumed event may be invalid Lai Jiangshan
@ 2009-06-30 10:54 ` Frederic Weisbecker
0 siblings, 0 replies; 2+ messages in thread
From: Frederic Weisbecker @ 2009-06-30 10:54 UTC (permalink / raw)
To: Lai Jiangshan; +Cc: Steven Rostedt, LKML
On Tue, Jun 30, 2009 at 04:14:36PM +0800, Lai Jiangshan wrote:
>
> In get_return_for_leaf() {
> ......
> ring_buffer_consume()
> ......
> access to @curr->ent.pid
> }
>
> @curr is saved at consumed event, if this event is the last event
> in the page, the ring_buffer may reuse the page. It will become invalid.
Indeed!
> Is there any other code to ensure it valid when we `cat trace_pipe`?
>
> Lai.
The only solution I see is to do a copy of the entry before consuming it.
I'll try that, thanks for your report!
Frederic.
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2009-06-30 10:54 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2009-06-30 8:14 [QUESTION] trace_function_graph: consumed event may be invalid Lai Jiangshan
2009-06-30 10:54 ` Frederic Weisbecker
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox