Messages in this thread | | | Date | Wed, 25 Mar 2015 15:20:09 -0400 | From | Don Zickus <> | Subject | Re: [PATCH] perf tool: Fix ppid for synthesized fork events |
| |
On Wed, Mar 25, 2015 at 09:03:38AM -0600, David Ahern wrote: > On 3/25/15 7:24 AM, Arnaldo Carvalho de Melo wrote: > >So it starts when there are tons of threads in the system, for which > >synthezing from /proc will have to take place, without looking again at > >that patch, I can't think about what would be a problem :-\ > > 3 extra lines are read from /proc/pid/status: > > Name: bash > State: S (sleeping) > Tgid: 6046 > > < current patch breaks here> > > Ngid: 0 > Pid: 6046 > PPid: 6045 > > < my patch reads these 3 lines, repeats the memcmp and does an atoi > on the PPid: value >
Ah, I read my email out of order. So you figured out the extra latency. Nice. In that case I think I am ok with your V2, though I still think split it out would make sense.
Sorry about that. Thanks for quickly debugging that!!
Cheers, Don
> > Let me remove that loop by reading in 4k at a time and making a > single pass. That should bring down the overhead, but filling in the > ppid will add some. > > David >
| |