lkml.org 
[lkml]   [2013]   [Dec]   [11]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH v0 13/71] perf tools: Add machine__get_thread_pid()
On 12/11/13, 2:18 PM, Andi Kleen wrote:
> David Ahern <dsahern@gmail.com> writes:
>>
>> Are you looking up the current or next task? If the former why not use
>> sample->pid rather than parsing the sched_switch tracepoint?
>
> The itrace stream doesn't have a pid field, and it needs the exact
> time stamp of the switch. There may not actually be any samples
> before decoding.
>
> -Andi
>

What I meant is this:

perf record -e sched:sched_switch -a -- sleep 1 | perf script -f
comm,tid,pid,event,trace


qemu-system-x86 8688/8692 sched:sched_switch:
prev_comm=qemu-system-x86 prev_pid=8692 prev_prio=120 prev_state=S ==>
next_comm=swapper/15 next_pid=0 next_prio=120

8688/8692 are the pid and tid of the running task. If you are monitoring
sched_switch events and looking at running task -- the one getting
scheduled out -- you don't need to parse the tracepoint. But, if you
want to know next task then you do need to parse it. I was wondering
which task is getting looked up.

David


\
 
 \ /
  Last update: 2013-12-11 23:01    [W:0.489 / U:0.084 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site