Messages in this thread | | | Date | Tue, 26 Feb 2013 15:38:38 +0100 | From | Oleg Nesterov <> | Subject | Re: [PATCH 1/5] perf tools: Fix -C option for record command |
| |
On 02/25, Jiri Olsa wrote: > > On Mon, Feb 25, 2013 at 11:56:38AM -0700, David Ahern wrote: > > > If a target is given (-a, -C, -p or -t) that is what the data is > > collected for -- all cpus, a cpu, or one or more task ids. The > > workload in that case becomes a means for bounding the data > > collection (start and end points).
OK, so I misunderstood the intent and misread the code. Thanks David.
> yes, that's what actually this patch fixies.. now you can run: > > perf record -C 0 ls > > and that will attach to cpu 0 only (same as '-a -C 0' before)
Yes, and thus this is consistent with -a and correct. Thanks and sorry for noise.
But,
> maybe we could consider having: > > perf record -C 0 ls > - attaching to CPU 0 and ls workload pid > > perf record -a -C 0 ls > - attaching to CPU 0 globaly
Can't really comment since I am not a perf user, but imho looks good.
And useful. And least this behaviour was useful to me when I tried to test the perf/uprobe changes.
Oleg.
| |