Re: [PATCH] perf inject: Flush ordered events on FINISHED_ROUND

2020-10-11 Thread Namhyung Kim
Hi Jiri, On Thu, Oct 8, 2020 at 6:07 PM Jiri Olsa wrote: > > On Tue, Oct 06, 2020 at 02:40:32PM +0900, namhy...@kernel.org wrote: > > This is the perf stat result: > > > > * Before > > > > 7,167,414,019 L1-dcache-loads > >337,471,761 L1-dcache-read-misses #4.71% of

Re: [PATCH] perf inject: Flush ordered events on FINISHED_ROUND

2020-10-08 Thread Jiri Olsa
On Tue, Oct 06, 2020 at 02:40:32PM +0900, namhy...@kernel.org wrote: > On Tue, Oct 06, 2020 at 11:39:49AM +0900, namhy...@kernel.org wrote: > > > > On Fri, Oct 02, 2020 at 10:03:17PM +0900, Namhyung Kim wrote: > > > > > Below measures time and memory usage during the perf inject and > > > > >

Re: [PATCH] perf inject: Flush ordered events on FINISHED_ROUND

2020-10-05 Thread namhyung
On Tue, Oct 06, 2020 at 11:39:49AM +0900, namhy...@kernel.org wrote: > > > On Fri, Oct 02, 2020 at 10:03:17PM +0900, Namhyung Kim wrote: > > > > Below measures time and memory usage during the perf inject and > > > > report using ~190MB data file. > > > > > > > > Before: > > > > perf inject:

Re: [PATCH] perf inject: Flush ordered events on FINISHED_ROUND

2020-10-05 Thread namhyung
> > On Fri, Oct 02, 2020 at 10:03:17PM +0900, Namhyung Kim wrote: > > > Below measures time and memory usage during the perf inject and > > > report using ~190MB data file. > > > > > > Before: > > > perf inject: 11.09 s, 382148 KB > > > perf report: 8.05 s, 397440 KB > > > > > > After: >

Re: [PATCH] perf inject: Flush ordered events on FINISHED_ROUND

2020-10-05 Thread Namhyung Kim
Hi Jiri, On Mon, Oct 5, 2020 at 4:52 AM Jiri Olsa wrote: > > On Fri, Oct 02, 2020 at 10:03:17PM +0900, Namhyung Kim wrote: > > Currently perf inject just repipes the event without any flush. It > > makes an issue that it changes the order of events processed. > > > > Normally we want to process

Re: [PATCH] perf inject: Flush ordered events on FINISHED_ROUND

2020-10-04 Thread Jiri Olsa
On Fri, Oct 02, 2020 at 10:03:17PM +0900, Namhyung Kim wrote: > Currently perf inject just repipes the event without any flush. It > makes an issue that it changes the order of events processed. > > Normally we want to process events in time order, but events are > recorded from several cpus and

[PATCH] perf inject: Flush ordered events on FINISHED_ROUND

2020-10-02 Thread Namhyung Kim
Currently perf inject just repipes the event without any flush. It makes an issue that it changes the order of events processed. Normally we want to process events in time order, but events are recorded from several cpus and they weren't sorted with each other. So we put them into the ordered