Re: [PATCH v2 06/15] perf session: load data directory into tool process memory

[Jiri Olsa]

On Wed, Oct 28, 2020 at 04:22:49PM +0900, Namhyung Kim wrote:
> On Tue, Oct 27, 2020 at 11:43 PM Alexey Budankov
> <alexey.budankov@xxxxxxxxxxxxxxx> wrote:
> >
> >
> > On 27.10.2020 15:21, Jiri Olsa wrote:
> > > On Tue, Oct 27, 2020 at 10:37:58AM +0300, Alexey Budankov wrote:
> > >> I agree perf report OOM issue can exist on really-big servers but data
> > >> directories support for report mode for not-so-big servers and desktops
> > >> is already enabled with this smaller change. Also really-big-servers
> > >> come with really-big amount of memory and collection could possibly be
> > >> limited to only interesting phases of execution so the issue could likely
> > >> be avoided. At the same time threaded trace streaming could clarify on
> > >> real use cases that are blocked by perf report OOM issue and that would
> > >> clarify on exact required solution. So perf report OOM issue shouldn't
> > >> be the showstopper for upstream of threaded trace streaming.
> > >
> > > so the short answer is no, right? ;-)
> >
> > Answer to what question? Resolve OOM in perf report for data directories?
> > I don't see a simple solution for that. The next issue after OOM is resolved
> > is a very long processing of data directories. And again there is no simple
> > solution for that as well. But it still need progress in order to be resolved
> > eventually.
> 
> I think we should find a better way than just adding all events to the
> ordered events queue in memory then processing them one by one.
> 
> Separating tracking events (FORK/MMAP/...) might be the first step.

I recall seeing this change before for threaded perf report,
maybe even from you, right? ;-)

jirka

> 
> Thanks
> Namhyung
>