On Wed, Oct 21, 2020 at 07:01:19PM +0300, Alexey Budankov wrote: > > Read trace files located in data directory into tool process memory. > Basic analysis support of data directories is provided for report > mode. Raw dump (-D) and aggregated reports are available for data > directories, still with no memory consumption optimizations. However > data directories collected with --compression-level option enabled > can be analyzed with little less memory because trace files are > unmaped from tool process memory after loading collected data. > The implementation is based on the prototype [1], [2]. > > [1] git clone https://git.kernel.org/pub/scm/linux/kernel/git/jolsa/perf.git > -b perf/record_threads > [2] https://lore.kernel.org/lkml/20180913125450.21342-1-jo...@kernel.org/ > > Suggested-by: Jiri Olsa <jo...@kernel.org>
very loosely ;-) so there was a reason for all that reader refactoring, so we could have __perf_session__process_dir_events function: https://git.kernel.org/pub/scm/linux/kernel/git/jolsa/perf.git/commit/?h=perf/record_threads&id=308aa7cff1fed335401cfc02c7bac1a4644af68e when reporting the threaded record data on really big servers, you will run out of memory, so you need to read and flush all the files together by smaller pieces IMO we need to have this change before we allow threaded record jirka > Suggested-by: Namhyung Kim <namhy...@kernel.org> > Signed-off-by: Alexey Budankov <alexey.budan...@linux.intel.com> > --- > tools/perf/util/session.c | 48 +++++++++++++++++++++++++++++++++++++++ > tools/perf/util/session.h | 1 + > 2 files changed, 49 insertions(+) > > diff --git a/tools/perf/util/session.c b/tools/perf/util/session.c > index 6afc670fdf0c..0752eec19813 100644 > --- a/tools/perf/util/session.c > +++ b/tools/perf/util/session.c > @@ -2212,6 +2212,17 @@ reader__process_events(struct reader *rd, struct > perf_session *session, > goto more; > > out: > + if (rd->unmap_file) { > + int i; > + > + for (i = 0; i < NUM_MMAPS; i++) { > + if (mmaps[i]) { > + munmap(mmaps[i], mmap_size); > + mmaps[i] = NULL; > + } > + } > + } > + > return err; > } > > @@ -2231,6 +2242,7 @@ static int __perf_session__process_events(struct > perf_session *session) > .data_offset = session->header.data_offset, > .process = process_simple, > .path = session->data->file.path, > + .unmap_file = false, > }; > struct ordered_events *oe = &session->ordered_events; > struct perf_tool *tool = session->tool; > @@ -2247,6 +2259,42 @@ static int __perf_session__process_events(struct > perf_session *session) > err = reader__process_events(&rd, session, &prog); > if (err) > goto out_err; > + > + if (perf_data__is_dir(session->data)) { > + int i, nr = session->data->dir.nr; > + struct reader file_rd[nr]; > + u64 total_size = perf_data__size(session->data); > + > + total_size -= session->data->file.size; > + ui_progress__init_size(&prog, total_size, "Sorting events..."); > + > + memset(&file_rd, 0, nr * sizeof(file_rd[0])); > + > + for (i = 0; i < nr ; i++) { > + struct perf_data_file *file; > + > + file = &session->data->dir.files[i]; > + file_rd[i] = (struct reader) { > + .fd = file->fd, > + .path = file->path, > + .data_size = file->size, > + .data_offset = 0, > + .process = process_simple, > + }; > + file_rd[i].unmap_file = > perf_header__has_feat(&session->header, > + > HEADER_COMPRESSED); > + session->reader = &file_rd[i]; > + > + if (zstd_init(&(file_rd[i].zstd_data), 0)) > + goto out_err; > + err = reader__process_events(&file_rd[i], session, > &prog); > + zstd_fini(&(file_rd[i].zstd_data)); > + session->reader = NULL; > + if (err) > + goto out_err; > + } > + } > + > /* do the final flush for ordered samples */ > err = ordered_events__flush(oe, OE_FLUSH__FINAL); > if (err) > diff --git a/tools/perf/util/session.h b/tools/perf/util/session.h > index 4fc9ccdf7970..d428f3eaf7fd 100644 > --- a/tools/perf/util/session.h > +++ b/tools/perf/util/session.h > @@ -43,6 +43,7 @@ struct reader { > u64 data_offset; > reader_cb_t process; > struct zstd_data zstd_data; > + bool unmap_file; > }; > > struct perf_session { > -- > 2.24.1 >
