[PATCH v5 00/20] Introduce threaded trace streaming for basic perf record operation

[PATCH v5 00/20] Introduce threaded trace streaming for basic perf record operation

[Alexey Bayduraev]

Changes in v5:
- fixed leaks in record__init_thread_masks_spec()
- fixed leaks after failed realloc
- replaced "%m" to strerror()
- added masks examples to the documentation
- captured Acked-by: tags by Andi Kleen
- do not allow --thread option for full_auxtrace mode 
- split patch 06/12 to 06/20 and 07/20
- split patch 08/12 to 09/20 and 10/20
- split patches 11/12 and 11/12 to 13/20-20/20

v4: https://lore.kernel.org/lkml/6c15adcb-6a9d-320e-70b5-957c4c8b6ff2@linux.intel.com/

Changes in v4:
- renamed 'comm' structure to 'pipes'
- moved thread fd/maps messages to verbose=2
- fixed leaks during allocation of thread_data structures
- fixed leaks during allocation of thread masks
- fixed possible fails when releasing thread masks

v3: https://lore.kernel.org/lkml/7d197a2d-56e2-896d-bf96-6de0a4db1fb8@linux.intel.com/

Changes in v3:
- avoided skipped redundant patch 3/15
- applied "data file" and "data directory" terms allover the patch set
- captured Acked-by: tags by Namhyung Kim
- avoided braces where don't needed
- employed thread local variable for serial trace streaming 
- added specs for --thread option - core, socket, numa and user defined
- added parallel loading of data directory files similar to the prototype [1]

v2: https://lore.kernel.org/lkml/1ec29ed6-0047-d22f-630b-a7f5ccee96b4@linux.intel.com/

Changes in v2:
- explicitly added credit tags to patches 6/15 and 15/15,
  additionally to cites [1], [2]
- updated description of 3/15 to explicitly mention the reason
  to open data directories in read access mode (e.g. for perf report)
- implemented fix for compilation error of 2/15
- explicitly elaborated on found issues to be resolved for
  threaded AUX trace capture

v1: https://lore.kernel.org/lkml/810f3a69-0004-9dff-a911-b7ff97220ae0@linux.intel.com/

Patch set provides parallel threaded trace streaming mode for basic
perf record operation. Provided mode mitigates profiling data losses
and resolves scalability issues of serial and asynchronous (--aio)
trace streaming modes on multicore server systems. The design and
implementation are based on the prototype [1], [2].

Parallel threaded mode executes trace streaming threads that read kernel
data buffers and write captured data into several data files located at
data directory. Layout of trace streaming threads and their mapping to data
buffers to read can be configured using a value of --thread command line
option. Specification value provides masks separated by colon so the masks
define cpus to be monitored by one thread and thread affinity mask is
separated by slash. <cpus mask 1>/<affinity mask 1>:<cpu mask 2>/<affinity mask 2>
specifies parallel threads layout that consists of two threads with
corresponding assigned cpus to be monitored. Specification value can be
a string e.g. "cpu", "core" or "socket" meaning creation of data streaming
thread for monitoring every cpu, whole core or socket. The option provided
with no or empty value defaults to "cpu" layout creating data streaming
thread for every cpu being monitored. Specification masks are filtered
by the mask provided via -C option.

Parallel streaming mode is compatible with Zstd compression/decompression
(--compression-level) and external control commands (--control). The mode
is not enabled for pipe mode. The mode is not enabled for AUX area tracing,
related and derived modes like --snapshot or --aux-sample. --switch-output-*
and --timestamp-filename options are not enabled for parallel streaming.
Initial intent to enable AUX area tracing faced the need to define some
optimal way to store index data in data directory. --switch-output-* and
--timestamp-filename use cases are not clear for data directories.
Asynchronous(--aio) trace streaming and affinity (--affinity) modes are
mutually exclusive to parallel streaming mode.

Basic analysis of data directories is provided in perf report mode.
Raw dump and aggregated reports are available for data directories,
still with no memory consumption optimizations.

Tested:

tools/perf/perf record -o prof.data --threads -- matrix.gcc.g.O3
tools/perf/perf record -o prof.data --threads= -- matrix.gcc.g.O3
tools/perf/perf record -o prof.data --threads=cpu -- matrix.gcc.g.O3
tools/perf/perf record -o prof.data --threads=core -- matrix.gcc.g.O3
tools/perf/perf record -o prof.data --threads=socket -- matrix.gcc.g.O3
tools/perf/perf record -o prof.data --threads=numa -- matrix.gcc.g.O3
tools/perf/perf record -o prof.data --threads=0-3/3:4-7/4 -- matrix.gcc.g.O3
tools/perf/perf record -o prof.data -C 2,5 --threads=0-3/3:4-7/4 -- matrix.gcc.g.O3
tools/perf/perf record -o prof.data -C 3,4 --threads=0-3/3:4-7/4 -- matrix.gcc.g.O3
tools/perf/perf record -o prof.data -C 0,4,2,6 --threads=core -- matrix.gcc.g.O3
tools/perf/perf record -o prof.data -C 0,4,2,6 --threads=numa -- matrix.gcc.g.O3
tools/perf/perf record -o prof.data --threads -g --call-graph dwarf,4096 -- matrix.gcc.g.O3
tools/perf/perf record -o prof.data --threads -g --call-graph dwarf,4096 --compression-level=3 -- matrix.gcc.g.O3
tools/perf/perf record -o prof.data --threads -a
tools/perf/perf record -D -1 -e cpu-cycles -a --control fd:10,11 -- sleep 30
tools/perf/perf record --threads -D -1 -e cpu-cycles -a --control fd:10,11 -- sleep 30

tools/perf/perf report -i prof.data
tools/perf/perf report -i prof.data --call-graph=callee
tools/perf/perf report -i prof.data --stdio --header
tools/perf/perf report -i prof.data -D --header

[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-jolsa@kernel.org/

---

Alexey Bayduraev (20):
  perf record: introduce thread affinity and mmap masks
  perf record: introduce thread specific data array
  perf record: introduce thread local variable
  perf record: stop threads in the end of trace streaming
  perf record: start threads in the beginning of trace streaming
  perf record: introduce data file at mmap buffer object
  perf record: introduce data transferred and compressed stats
  perf record: init data file at mmap buffer object
  tools lib: introduce bitmap_intersects() operation
  perf record: introduce --threads=<spec> command line option
  perf record: document parallel data streaming mode
  perf report: output data file name in raw trace dump
  perf session: move reader structure to the top
  perf session: introduce reader_state in reader object
  perf session: introduce reader objects in session object
  perf session: introduce decompressor into trace reader object
  perf session: move init into reader__init function
  perf session: move map/unmap into reader__mmap function
  perf session: load single file for analysis
  perf session: load data directory files for analysis

 tools/include/linux/bitmap.h             |   11 +
 tools/lib/api/fd/array.c                 |   17 +
 tools/lib/api/fd/array.h                 |    1 +
 tools/lib/bitmap.c                       |   14 +
 tools/perf/Documentation/perf-record.txt |   30 +
 tools/perf/builtin-inject.c              |    3 +-
 tools/perf/builtin-record.c              | 1066 ++++++++++++++++++++--
 tools/perf/util/evlist.c                 |   16 +
 tools/perf/util/evlist.h                 |    1 +
 tools/perf/util/mmap.c                   |    6 +
 tools/perf/util/mmap.h                   |    6 +
 tools/perf/util/ordered-events.h         |    1 +
 tools/perf/util/record.h                 |    2 +
 tools/perf/util/session.c                |  491 +++++++---
 tools/perf/util/session.h                |    5 +
 tools/perf/util/tool.h                   |    3 +-
 16 files changed, 1474 insertions(+), 199 deletions(-)

-- 
2.19.0

[PATCH v5 01/20] perf record: introduce thread affinity and mmap masks

[Alexey Bayduraev]

Introduce affinity and mmap thread masks. Thread affinity mask
defines cpus that a thread is allowed to run on. Thread maps
mask defines mmap data buffers the thread serves to stream
profiling data from.

Acked-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Alexey Bayduraev <alexey.v.bayduraev@linux.intel.com>
---
 tools/perf/builtin-record.c | 117 ++++++++++++++++++++++++++++++++++++
 1 file changed, 117 insertions(+)

diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c
index 3337b5f93336..4701aec4279c 100644
--- a/tools/perf/builtin-record.c
+++ b/tools/perf/builtin-record.c
@@ -87,6 +87,11 @@ struct switch_output {
 	int		 cur_file;
 };
 
+struct thread_mask {
+	struct mmap_cpu_mask	maps;
+	struct mmap_cpu_mask	affinity;
+};
+
 struct record {
 	struct perf_tool	tool;
 	struct record_opts	opts;
@@ -111,6 +116,8 @@ struct record {
 	unsigned long long	samples;
 	struct mmap_cpu_mask	affinity_mask;
 	unsigned long		output_max_size;	/* = 0: unlimited */
+	struct thread_mask	*thread_masks;
+	int			nr_threads;
 };
 
 static volatile int done;
@@ -2210,6 +2217,45 @@ static int record__parse_affinity(const struct option *opt, const char *str, int
 	return 0;
 }
 
+static int record__mmap_cpu_mask_alloc(struct mmap_cpu_mask *mask, int nr_bits)
+{
+	mask->nbits = nr_bits;
+	mask->bits = bitmap_alloc(mask->nbits);
+	if (!mask->bits) {
+		pr_err("Failed to allocate mmap_cpu mask\n");
+		return -ENOMEM;
+	}
+
+	return 0;
+}
+
+static void record__mmap_cpu_mask_free(struct mmap_cpu_mask *mask)
+{
+	bitmap_free(mask->bits);
+	mask->nbits = 0;
+}
+
+static void record__thread_mask_clear(struct thread_mask *mask)
+{
+	bitmap_zero(mask->maps.bits, mask->maps.nbits);
+	bitmap_zero(mask->affinity.bits, mask->affinity.nbits);
+}
+
+static int record__thread_mask_alloc(struct thread_mask *mask, int nr_bits)
+{
+	if (record__mmap_cpu_mask_alloc(&mask->maps, nr_bits) ||
+	    record__mmap_cpu_mask_alloc(&mask->affinity, nr_bits))
+		return -ENOMEM;
+
+	return 0;
+}
+
+static void record__thread_mask_free(struct thread_mask *mask)
+{
+	record__mmap_cpu_mask_free(&mask->maps);
+	record__mmap_cpu_mask_free(&mask->affinity);
+}
+
 static int parse_output_max_size(const struct option *opt,
 				 const char *str, int unset)
 {
@@ -2645,6 +2691,70 @@ static struct option __record_options[] = {
 
 struct option *record_options = __record_options;
 
+static void record__mmap_cpu_mask_init(struct mmap_cpu_mask *mask, struct perf_cpu_map *cpus)
+{
+	int c;
+
+	for (c = 0; c < cpus->nr; c++)
+		set_bit(cpus->map[c], mask->bits);
+}
+
+static int record__alloc_thread_masks(struct record *rec, int nr_threads, int nr_bits)
+{
+	int t, ret;
+
+	rec->thread_masks = zalloc(nr_threads * sizeof(*(rec->thread_masks)));
+	if (!rec->thread_masks) {
+		pr_err("Failed to allocate thread masks\n");
+		return -ENOMEM;
+	}
+
+	for (t = 0; t < nr_threads; t++) {
+		ret = record__thread_mask_alloc(&rec->thread_masks[t], nr_bits);
+		if (ret)
+			return ret;
+		record__thread_mask_clear(&rec->thread_masks[t]);
+	}
+
+	return 0;
+}
+static int record__init_thread_default_masks(struct record *rec, struct perf_cpu_map *cpus)
+{
+	int ret;
+
+	ret = record__alloc_thread_masks(rec, 1, cpu__max_cpu());
+	if (ret)
+		return ret;
+
+	record__mmap_cpu_mask_init(&rec->thread_masks->maps, cpus);
+
+	rec->nr_threads = 1;
+
+	return 0;
+}
+
+static int record__init_thread_masks(struct record *rec)
+{
+	struct perf_cpu_map *cpus = rec->evlist->core.cpus;
+
+	return record__init_thread_default_masks(rec, cpus);
+}
+
+static int record__fini_thread_masks(struct record *rec)
+{
+	int t;
+
+	if (rec->thread_masks)
+		for (t = 0; t < rec->nr_threads; t++)
+			record__thread_mask_free(&rec->thread_masks[t]);
+
+	zfree(&rec->thread_masks);
+
+	rec->nr_threads = 0;
+
+	return 0;
+}
+
 int cmd_record(int argc, const char **argv)
 {
 	int err;
@@ -2887,6 +2997,12 @@ int cmd_record(int argc, const char **argv)
 		goto out;
 	}
 
+	err = record__init_thread_masks(rec);
+	if (err) {
+		pr_err("record__init_thread_masks failed, error %d\n", err);
+		goto out;
+	}
+
 	if (rec->opts.nr_cblocks > nr_cblocks_max)
 		rec->opts.nr_cblocks = nr_cblocks_max;
 	pr_debug("nr_cblocks: %d\n", rec->opts.nr_cblocks);
@@ -2905,6 +3021,7 @@ int cmd_record(int argc, const char **argv)
 	symbol__exit();
 	auxtrace_record__free(rec->itr);
 out_opts:
+	record__fini_thread_masks(rec);
 	evlist__close_control(rec->opts.ctl_fd, rec->opts.ctl_fd_ack, &rec->opts.ctl_fd_close);
 	return err;
 }
-- 
2.19.0

[PATCH v5 02/20] perf record: introduce thread specific data array

[Alexey Bayduraev]

Introduce thread specific data object and array of such objects
to store and manage thread local data. Implement functions to
allocate, initialize, finalize and release thread specific data.

Thread local maps and overwrite_maps arrays keep pointers to
mmap buffer objects to serve according to maps thread mask.
Thread local pollfd array keeps event fds connected to mmaps
buffers according to maps thread mask.

Thread control commands are delivered via thread local comm pipes
and ctlfd_pos fd. External control commands (--control option)
are delivered via evlist ctlfd_pos fd and handled by the main
tool thread.

Signed-off-by: Alexey Bayduraev <alexey.v.bayduraev@linux.intel.com>
---
 tools/lib/api/fd/array.c    |  17 ++++
 tools/lib/api/fd/array.h    |   1 +
 tools/perf/builtin-record.c | 196 +++++++++++++++++++++++++++++++++++-
 3 files changed, 211 insertions(+), 3 deletions(-)

diff --git a/tools/lib/api/fd/array.c b/tools/lib/api/fd/array.c
index 5e6cb9debe37..de8bcbaea3f1 100644
--- a/tools/lib/api/fd/array.c
+++ b/tools/lib/api/fd/array.c
@@ -88,6 +88,23 @@ int fdarray__add(struct fdarray *fda, int fd, short revents, enum fdarray_flags
 	return pos;
 }
 
+int fdarray__clone(struct fdarray *fda, int pos, struct fdarray *base)
+{
+	struct pollfd *entry;
+	int npos;
+
+	if (pos >= base->nr)
+		return -EINVAL;
+
+	entry = &base->entries[pos];
+
+	npos = fdarray__add(fda, entry->fd, entry->events, base->priv[pos].flags);
+	if (npos >= 0)
+		fda->priv[npos] = base->priv[pos];
+
+	return npos;
+}
+
 int fdarray__filter(struct fdarray *fda, short revents,
 		    void (*entry_destructor)(struct fdarray *fda, int fd, void *arg),
 		    void *arg)
diff --git a/tools/lib/api/fd/array.h b/tools/lib/api/fd/array.h
index 7fcf21a33c0c..4a03da7f1fc1 100644
--- a/tools/lib/api/fd/array.h
+++ b/tools/lib/api/fd/array.h
@@ -42,6 +42,7 @@ struct fdarray *fdarray__new(int nr_alloc, int nr_autogrow);
 void fdarray__delete(struct fdarray *fda);
 
 int fdarray__add(struct fdarray *fda, int fd, short revents, enum fdarray_flags flags);
+int fdarray__clone(struct fdarray *fda, int pos, struct fdarray *base);
 int fdarray__poll(struct fdarray *fda, int timeout);
 int fdarray__filter(struct fdarray *fda, short revents,
 		    void (*entry_destructor)(struct fdarray *fda, int fd, void *arg),
diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c
index 4701aec4279c..a73c28051b79 100644
--- a/tools/perf/builtin-record.c
+++ b/tools/perf/builtin-record.c
@@ -58,6 +58,7 @@
 #include <poll.h>
 #include <pthread.h>
 #include <unistd.h>
+#include <sys/syscall.h>
 #include <sched.h>
 #include <signal.h>
 #ifdef HAVE_EVENTFD_SUPPORT
@@ -92,6 +93,23 @@ struct thread_mask {
 	struct mmap_cpu_mask	affinity;
 };
 
+struct thread_data {
+	pid_t			tid;
+	struct thread_mask	*mask;
+	struct {
+		int		msg[2];
+		int		ack[2];
+	} pipes;
+	struct fdarray		pollfd;
+	int			ctlfd_pos;
+	struct mmap		**maps;
+	struct mmap		**overwrite_maps;
+	int			nr_mmaps;
+	struct record		*rec;
+	unsigned long long	samples;
+	unsigned long		waking;
+};
+
 struct record {
 	struct perf_tool	tool;
 	struct record_opts	opts;
@@ -117,6 +135,7 @@ struct record {
 	struct mmap_cpu_mask	affinity_mask;
 	unsigned long		output_max_size;	/* = 0: unlimited */
 	struct thread_mask	*thread_masks;
+	struct thread_data	*thread_data;
 	int			nr_threads;
 };
 
@@ -847,9 +866,174 @@ static int record__kcore_copy(struct machine *machine, struct perf_data *data)
 	return kcore_copy(from_dir, kcore_dir);
 }
 
+static int record__thread_data_init_pipes(struct thread_data *thread_data)
+{
+	if (pipe(thread_data->pipes.msg) || pipe(thread_data->pipes.ack)) {
+		pr_err("Failed to create thread communication pipes: %s\n", strerror(errno));
+		return -ENOMEM;
+	}
+
+	pr_debug2("thread_data[%p]: msg=[%d,%d], ack=[%d,%d]\n", thread_data,
+		 thread_data->pipes.msg[0], thread_data->pipes.msg[1],
+		 thread_data->pipes.ack[0], thread_data->pipes.ack[1]);
+
+	return 0;
+}
+
+static int record__thread_data_init_maps(struct thread_data *thread_data, struct evlist *evlist)
+{
+	int m, tm, nr_mmaps = evlist->core.nr_mmaps;
+	struct mmap *mmap = evlist->mmap;
+	struct mmap *overwrite_mmap = evlist->overwrite_mmap;
+	struct perf_cpu_map *cpus = evlist->core.cpus;
+
+	thread_data->nr_mmaps = bitmap_weight(thread_data->mask->maps.bits,
+					      thread_data->mask->maps.nbits);
+	if (mmap) {
+		thread_data->maps = zalloc(thread_data->nr_mmaps * sizeof(struct mmap *));
+		if (!thread_data->maps) {
+			pr_err("Failed to allocate maps thread data\n");
+			return -ENOMEM;
+		}
+	}
+	if (overwrite_mmap) {
+		thread_data->overwrite_maps = zalloc(thread_data->nr_mmaps * sizeof(struct mmap *));
+		if (!thread_data->overwrite_maps) {
+			pr_err("Failed to allocate overwrite maps thread data\n");
+			return -ENOMEM;
+		}
+	}
+	pr_debug2("thread_data[%p]: nr_mmaps=%d, maps=%p, overwrite_maps=%p\n", thread_data,
+		 thread_data->nr_mmaps, thread_data->maps, thread_data->overwrite_maps);
+
+	for (m = 0, tm = 0; m < nr_mmaps && tm < thread_data->nr_mmaps; m++) {
+		if (test_bit(cpus->map[m], thread_data->mask->maps.bits)) {
+			if (thread_data->maps) {
+				thread_data->maps[tm] = &mmap[m];
+				pr_debug2("thread_data[%p]: maps[%d] -> mmap[%d], cpus[%d]\n",
+					  thread_data, tm, m, cpus->map[m]);
+			}
+			if (thread_data->overwrite_maps) {
+				thread_data->overwrite_maps[tm] = &overwrite_mmap[m];
+				pr_debug2("thread_data[%p]: ov_maps[%d] -> ov_mmap[%d], cpus[%d]\n",
+					  thread_data, tm, m, cpus->map[m]);
+			}
+			tm++;
+		}
+	}
+
+	return 0;
+}
+
+static int record__thread_data_init_pollfd(struct thread_data *thread_data, struct evlist *evlist)
+{
+	int f, tm, pos;
+	struct mmap *map, *overwrite_map;
+
+	fdarray__init(&thread_data->pollfd, 64);
+
+	for (tm = 0; tm < thread_data->nr_mmaps; tm++) {
+		map = thread_data->maps ? thread_data->maps[tm] : NULL;
+		overwrite_map = thread_data->overwrite_maps ?
+				thread_data->overwrite_maps[tm] : NULL;
+
+		for (f = 0; f < evlist->core.pollfd.nr; f++) {
+			void *ptr = evlist->core.pollfd.priv[f].ptr;
+
+			if ((map && ptr == map) || (overwrite_map && ptr == overwrite_map)) {
+				pos = fdarray__clone(&thread_data->pollfd, f, &evlist->core.pollfd);
+				if (pos < 0)
+					return pos;
+				pr_debug2("thread_data[%p]: pollfd[%d] <- event_fd=%d\n",
+					 thread_data, pos, evlist->core.pollfd.entries[f].fd);
+			}
+		}
+	}
+
+	return 0;
+}
+
+static int record__alloc_thread_data(struct record *rec, struct evlist *evlist)
+{
+	int t, ret;
+	struct thread_data *thread_data;
+
+	rec->thread_data = zalloc(rec->nr_threads * sizeof(*(rec->thread_data)));
+	if (!rec->thread_data) {
+		pr_err("Failed to allocate thread data\n");
+		return -ENOMEM;
+	}
+	thread_data = rec->thread_data;
+
+	for (t = 0; t < rec->nr_threads; t++) {
+		thread_data[t].rec = rec;
+		thread_data[t].mask = &rec->thread_masks[t];
+		ret = record__thread_data_init_maps(&thread_data[t], evlist);
+		if (ret)
+			return ret;
+		ret = record__thread_data_init_pollfd(&thread_data[t], evlist);
+		if (ret)
+			return ret;
+		if (t) {
+			thread_data[t].tid = -1;
+			ret = record__thread_data_init_pipes(&thread_data[t]);
+			if (ret)
+				return ret;
+			thread_data[t].ctlfd_pos = fdarray__add(&thread_data[t].pollfd,
+								thread_data[t].pipes.msg[0],
+								POLLIN | POLLERR | POLLHUP,
+								fdarray_flag__nonfilterable);
+			if (thread_data[t].ctlfd_pos < 0)
+				return -ENOMEM;
+			pr_debug2("thread_data[%p]: pollfd[%d] <- ctl_fd=%d\n",
+				 thread_data, thread_data[t].ctlfd_pos,
+				 thread_data[t].pipes.msg[0]);
+		} else {
+			thread_data[t].tid = syscall(SYS_gettid);
+			if (evlist->ctl_fd.pos == -1)
+				continue;
+			thread_data[t].ctlfd_pos = fdarray__clone(&thread_data[t].pollfd,
+								  evlist->ctl_fd.pos,
+								  &evlist->core.pollfd);
+			if (thread_data[t].ctlfd_pos < 0)
+				return -ENOMEM;
+			pr_debug2("thread_data[%p]: pollfd[%d] <- ctl_fd=%d\n",
+				 thread_data, thread_data[t].ctlfd_pos,
+				 evlist->core.pollfd.entries[evlist->ctl_fd.pos].fd);
+		}
+	}
+
+	return 0;
+}
+
+static void record__free_thread_data(struct record *rec)
+{
+	int t;
+
+	if (rec->thread_data == NULL)
+		return;
+
+	for (t = 0; t < rec->nr_threads; t++) {
+		if (rec->thread_data[t].pipes.msg[0])
+			close(rec->thread_data[t].pipes.msg[0]);
+		if (rec->thread_data[t].pipes.msg[1])
+			close(rec->thread_data[t].pipes.msg[1]);
+		if (rec->thread_data[t].pipes.ack[0])
+			close(rec->thread_data[t].pipes.ack[0]);
+		if (rec->thread_data[t].pipes.ack[1])
+			close(rec->thread_data[t].pipes.ack[1]);
+		zfree(&rec->thread_data[t].maps);
+		zfree(&rec->thread_data[t].overwrite_maps);
+		fdarray__exit(&rec->thread_data[t].pollfd);
+	}
+
+	zfree(&rec->thread_data);
+}
+
 static int record__mmap_evlist(struct record *rec,
 			       struct evlist *evlist)
 {
+	int ret;
 	struct record_opts *opts = &rec->opts;
 	bool auxtrace_overwrite = opts->auxtrace_snapshot_mode ||
 				  opts->auxtrace_sample_mode;
@@ -880,6 +1064,14 @@ static int record__mmap_evlist(struct record *rec,
 				return -EINVAL;
 		}
 	}
+
+	if (evlist__initialize_ctlfd(evlist, opts->ctl_fd, opts->ctl_fd_ack))
+		return -1;
+
+	ret = record__alloc_thread_data(rec, evlist);
+	if (ret)
+		return ret;
+
 	return 0;
 }
 
@@ -1874,9 +2066,6 @@ static int __cmd_record(struct record *rec, int argc, const char **argv)
 		evlist__start_workload(rec->evlist);
 	}
 
-	if (evlist__initialize_ctlfd(rec->evlist, opts->ctl_fd, opts->ctl_fd_ack))
-		goto out_child;
-
 	if (opts->initial_delay) {
 		pr_info(EVLIST_DISABLED_MSG);
 		if (opts->initial_delay > 0) {
@@ -2032,6 +2221,7 @@ static int __cmd_record(struct record *rec, int argc, const char **argv)
 		record__synthesize_workload(rec, true);
 
 out_child:
+	record__free_thread_data(rec);
 	evlist__finalize_ctlfd(rec->evlist);
 	record__mmap_read_all(rec, true);
 	record__aio_mmap_read_sync(rec);
-- 
2.19.0

[PATCH v5 03/20] perf record: introduce thread local variable

[Alexey Bayduraev]

Introduce thread local variable and use it for threaded trace streaming.
Use thread affinity mask instead or record affinity mask in affinity
modes.
Introduce and use evlist__ctlfd_update() function to propagate external
control commands to global evlist object.

Acked-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Alexey Bayduraev <alexey.v.bayduraev@linux.intel.com>
---
 tools/perf/builtin-record.c | 133 ++++++++++++++++++++++++------------
 tools/perf/util/evlist.c    |  16 +++++
 tools/perf/util/evlist.h    |   1 +
 3 files changed, 106 insertions(+), 44 deletions(-)

diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c
index a73c28051b79..2a3dac894aa3 100644
--- a/tools/perf/builtin-record.c
+++ b/tools/perf/builtin-record.c
@@ -110,6 +110,8 @@ struct thread_data {
 	unsigned long		waking;
 };
 
+static __thread struct thread_data *thread;
+
 struct record {
 	struct perf_tool	tool;
 	struct record_opts	opts;
@@ -132,7 +134,6 @@ struct record {
 	bool			timestamp_boundary;
 	struct switch_output	switch_output;
 	unsigned long long	samples;
-	struct mmap_cpu_mask	affinity_mask;
 	unsigned long		output_max_size;	/* = 0: unlimited */
 	struct thread_mask	*thread_masks;
 	struct thread_data	*thread_data;
@@ -567,7 +568,7 @@ static int record__pushfn(struct mmap *map, void *to, void *bf, size_t size)
 		bf   = map->data;
 	}
 
-	rec->samples++;
+	thread->samples++;
 	return record__write(rec, map, bf, size);
 }
 
@@ -1254,16 +1255,24 @@ static struct perf_event_header finished_round_event = {
 
 static void record__adjust_affinity(struct record *rec, struct mmap *map)
 {
+	int ret = 0;
+
 	if (rec->opts.affinity != PERF_AFFINITY_SYS &&
-	    !bitmap_equal(rec->affinity_mask.bits, map->affinity_mask.bits,
-			  rec->affinity_mask.nbits)) {
-		bitmap_zero(rec->affinity_mask.bits, rec->affinity_mask.nbits);
-		bitmap_or(rec->affinity_mask.bits, rec->affinity_mask.bits,
-			  map->affinity_mask.bits, rec->affinity_mask.nbits);
-		sched_setaffinity(0, MMAP_CPU_MASK_BYTES(&rec->affinity_mask),
-				  (cpu_set_t *)rec->affinity_mask.bits);
-		if (verbose == 2)
-			mmap_cpu_mask__scnprintf(&rec->affinity_mask, "thread");
+	    !bitmap_equal(thread->mask->affinity.bits, map->affinity_mask.bits,
+			  thread->mask->affinity.nbits)) {
+		bitmap_zero(thread->mask->affinity.bits, thread->mask->affinity.nbits);
+		bitmap_or(thread->mask->affinity.bits, thread->mask->affinity.bits,
+			  map->affinity_mask.bits, thread->mask->affinity.nbits);
+		ret = sched_setaffinity(0, MMAP_CPU_MASK_BYTES(&thread->mask->affinity),
+					(cpu_set_t *)thread->mask->affinity.bits);
+		if (ret)
+			pr_err("threads[%d]: sched_setaffinity() call failed: %s\n",
+			       thread->tid, strerror(errno));
+		if (verbose == 2) {
+			pr_debug("threads[%d]: addr=", thread->tid);
+			mmap_cpu_mask__scnprintf(&thread->mask->affinity, "thread");
+			pr_debug("threads[%d]: on cpu=%d\n", thread->tid, sched_getcpu());
+		}
 	}
 }
 
@@ -1304,14 +1313,17 @@ static int record__mmap_read_evlist(struct record *rec, struct evlist *evlist,
 	u64 bytes_written = rec->bytes_written;
 	int i;
 	int rc = 0;
-	struct mmap *maps;
+	int nr_mmaps;
+	struct mmap **maps;
 	int trace_fd = rec->data.file.fd;
 	off_t off = 0;
 
 	if (!evlist)
 		return 0;
 
-	maps = overwrite ? evlist->overwrite_mmap : evlist->mmap;
+	nr_mmaps = thread->nr_mmaps;
+	maps = overwrite ? thread->overwrite_maps : thread->maps;
+
 	if (!maps)
 		return 0;
 
@@ -1321,9 +1333,9 @@ static int record__mmap_read_evlist(struct record *rec, struct evlist *evlist,
 	if (record__aio_enabled(rec))
 		off = record__aio_get_pos(trace_fd);
 
-	for (i = 0; i < evlist->core.nr_mmaps; i++) {
+	for (i = 0; i < nr_mmaps; i++) {
 		u64 flush = 0;
-		struct mmap *map = &maps[i];
+		struct mmap *map = maps[i];
 
 		if (map->core.base) {
 			record__adjust_affinity(rec, map);
@@ -1386,6 +1398,15 @@ static int record__mmap_read_all(struct record *rec, bool synch)
 	return record__mmap_read_evlist(rec, rec->evlist, true, synch);
 }
 
+static void record__thread_munmap_filtered(struct fdarray *fda, int fd,
+					   void *arg __maybe_unused)
+{
+	struct perf_mmap *map = fda->priv[fd].ptr;
+
+	if (map)
+		perf_mmap__put(map);
+}
+
 static void record__init_features(struct record *rec)
 {
 	struct perf_session *session = rec->session;
@@ -1830,6 +1851,33 @@ static void record__uniquify_name(struct record *rec)
 	}
 }
 
+static int record__start_threads(struct record *rec)
+{
+	struct thread_data *thread_data = rec->thread_data;
+
+	thread = &thread_data[0];
+
+	pr_debug("threads[%d]: started on cpu=%d\n", thread->tid, sched_getcpu());
+
+	return 0;
+}
+
+static int record__stop_threads(struct record *rec, unsigned long *waking)
+{
+	int t;
+	struct thread_data *thread_data = rec->thread_data;
+
+	for (t = 0; t < rec->nr_threads; t++) {
+		rec->samples += thread_data[t].samples;
+		*waking += thread_data[t].waking;
+		pr_debug("threads[%d]: samples=%lld, wakes=%ld, trasferred=%ld, compressed=%ld\n",
+			 thread_data[t].tid, thread_data[t].samples, thread_data[t].waking,
+			 rec->session->bytes_transferred, rec->session->bytes_compressed);
+	}
+
+	return 0;
+}
+
 static int __cmd_record(struct record *rec, int argc, const char **argv)
 {
 	int err;
@@ -1938,7 +1986,7 @@ static int __cmd_record(struct record *rec, int argc, const char **argv)
 
 	if (record__open(rec) != 0) {
 		err = -1;
-		goto out_child;
+		goto out_free_threads;
 	}
 	session->header.env.comp_mmap_len = session->evlist->core.mmap_len;
 
@@ -1946,7 +1994,7 @@ static int __cmd_record(struct record *rec, int argc, const char **argv)
 		err = record__kcore_copy(&session->machines.host, data);
 		if (err) {
 			pr_err("ERROR: Failed to copy kcore\n");
-			goto out_child;
+			goto out_free_threads;
 		}
 	}
 
@@ -1957,7 +2005,7 @@ static int __cmd_record(struct record *rec, int argc, const char **argv)
 		bpf__strerror_apply_obj_config(err, errbuf, sizeof(errbuf));
 		pr_err("ERROR: Apply config to BPF failed: %s\n",
 			 errbuf);
-		goto out_child;
+		goto out_free_threads;
 	}
 
 	/*
@@ -1975,11 +2023,11 @@ static int __cmd_record(struct record *rec, int argc, const char **argv)
 	if (data->is_pipe) {
 		err = perf_header__write_pipe(fd);
 		if (err < 0)
-			goto out_child;
+			goto out_free_threads;
 	} else {
 		err = perf_session__write_header(session, rec->evlist, fd, false);
 		if (err < 0)
-			goto out_child;
+			goto out_free_threads;
 	}
 
 	err = -1;
@@ -1987,16 +2035,16 @@ static int __cmd_record(struct record *rec, int argc, const char **argv)
 	    && !perf_header__has_feat(&session->header, HEADER_BUILD_ID)) {
 		pr_err("Couldn't generate buildids. "
 		       "Use --no-buildid to profile anyway.\n");
-		goto out_child;
+		goto out_free_threads;
 	}
 
 	err = record__setup_sb_evlist(rec);
 	if (err)
-		goto out_child;
+		goto out_free_threads;
 
 	err = record__synthesize(rec, false);
 	if (err < 0)
-		goto out_child;
+		goto out_free_threads;
 
 	if (rec->realtime_prio) {
 		struct sched_param param;
@@ -2005,10 +2053,13 @@ static int __cmd_record(struct record *rec, int argc, const char **argv)
 		if (sched_setscheduler(0, SCHED_FIFO, &param)) {
 			pr_err("Could not set realtime priority.\n");
 			err = -1;
-			goto out_child;
+			goto out_free_threads;
 		}
 	}
 
+	if (record__start_threads(rec))
+		goto out_free_threads;
+
 	/*
 	 * When perf is starting the traced process, all the events
 	 * (apart from group members) have enable_on_exec=1 set,
@@ -2079,7 +2130,7 @@ static int __cmd_record(struct record *rec, int argc, const char **argv)
 	trigger_ready(&switch_output_trigger);
 	perf_hooks__invoke_record_start();
 	for (;;) {
-		unsigned long long hits = rec->samples;
+		unsigned long long hits = thread->samples;
 
 		/*
 		 * rec->evlist->bkw_mmap_state is possible to be
@@ -2148,20 +2199,24 @@ static int __cmd_record(struct record *rec, int argc, const char **argv)
 				alarm(rec->switch_output.time);
 		}
 
-		if (hits == rec->samples) {
+		if (hits == thread->samples) {
 			if (done || draining)
 				break;
-			err = evlist__poll(rec->evlist, -1);
+			err = fdarray__poll(&thread->pollfd, -1);
 			/*
 			 * Propagate error, only if there's any. Ignore positive
 			 * number of returned events and interrupt error.
 			 */
 			if (err > 0 || (err < 0 && errno == EINTR))
 				err = 0;
-			waking++;
+			thread->waking++;
 
-			if (evlist__filter_pollfd(rec->evlist, POLLERR | POLLHUP) == 0)
+			if (fdarray__filter(&thread->pollfd, POLLERR | POLLHUP,
+					    record__thread_munmap_filtered, NULL) == 0)
 				draining = true;
+
+			evlist__ctlfd_update(rec->evlist,
+				&thread->pollfd.entries[thread->ctlfd_pos]);
 		}
 
 		if (evlist__ctlfd_process(rec->evlist, &cmd) > 0) {
@@ -2214,18 +2269,20 @@ static int __cmd_record(struct record *rec, int argc, const char **argv)
 		goto out_child;
 	}
 
-	if (!quiet)
-		fprintf(stderr, "[ perf record: Woken up %ld times to write data ]\n", waking);
-
 	if (target__none(&rec->opts.target))
 		record__synthesize_workload(rec, true);
 
 out_child:
+	record__stop_threads(rec, &waking);
+out_free_threads:
 	record__free_thread_data(rec);
 	evlist__finalize_ctlfd(rec->evlist);
 	record__mmap_read_all(rec, true);
 	record__aio_mmap_read_sync(rec);
 
+	if (!quiet)
+		fprintf(stderr, "[ perf record: Woken up %ld times to write data ]\n", waking);
+
 	if (rec->session->bytes_transferred && rec->session->bytes_compressed) {
 		ratio = (float)rec->session->bytes_transferred/(float)rec->session->bytes_compressed;
 		session->header.env.comp_ratio = ratio + 0.5;
@@ -3058,17 +3115,6 @@ int cmd_record(int argc, const char **argv)
 
 	symbol__init(NULL);
 
-	if (rec->opts.affinity != PERF_AFFINITY_SYS) {
-		rec->affinity_mask.nbits = cpu__max_cpu();
-		rec->affinity_mask.bits = bitmap_alloc(rec->affinity_mask.nbits);
-		if (!rec->affinity_mask.bits) {
-			pr_err("Failed to allocate thread mask for %zd cpus\n", rec->affinity_mask.nbits);
-			err = -ENOMEM;
-			goto out_opts;
-		}
-		pr_debug2("thread mask[%zd]: empty\n", rec->affinity_mask.nbits);
-	}
-
 	err = record__auxtrace_init(rec);
 	if (err)
 		goto out;
@@ -3206,7 +3252,6 @@ int cmd_record(int argc, const char **argv)
 
 	err = __cmd_record(&record, argc, argv);
 out:
-	bitmap_free(rec->affinity_mask.bits);
 	evlist__delete(rec->evlist);
 	symbol__exit();
 	auxtrace_record__free(rec->itr);
diff --git a/tools/perf/util/evlist.c b/tools/perf/util/evlist.c
index 6e5c41528c7d..ac343a42408d 100644
--- a/tools/perf/util/evlist.c
+++ b/tools/perf/util/evlist.c
@@ -2135,6 +2135,22 @@ int evlist__ctlfd_process(struct evlist *evlist, enum evlist_ctl_cmd *cmd)
 	return err;
 }
 
+int evlist__ctlfd_update(struct evlist *evlist, struct pollfd *update)
+{
+	int ctlfd_pos = evlist->ctl_fd.pos;
+	struct pollfd *entries = evlist->core.pollfd.entries;
+
+	if (!evlist__ctlfd_initialized(evlist))
+		return 0;
+
+	if (entries[ctlfd_pos].fd != update->fd ||
+	    entries[ctlfd_pos].events != update->events)
+		return -1;
+
+	entries[ctlfd_pos].revents = update->revents;
+	return 0;
+}
+
 struct evsel *evlist__find_evsel(struct evlist *evlist, int idx)
 {
 	struct evsel *evsel;
diff --git a/tools/perf/util/evlist.h b/tools/perf/util/evlist.h
index a8b97b50cceb..f3d96e79f667 100644
--- a/tools/perf/util/evlist.h
+++ b/tools/perf/util/evlist.h
@@ -358,6 +358,7 @@ void evlist__close_control(int ctl_fd, int ctl_fd_ack, bool *ctl_fd_close);
 int evlist__initialize_ctlfd(struct evlist *evlist, int ctl_fd, int ctl_fd_ack);
 int evlist__finalize_ctlfd(struct evlist *evlist);
 bool evlist__ctlfd_initialized(struct evlist *evlist);
+int evlist__ctlfd_update(struct evlist *evlist, struct pollfd *update);
 int evlist__ctlfd_process(struct evlist *evlist, enum evlist_ctl_cmd *cmd);
 int evlist__ctlfd_ack(struct evlist *evlist);
 
-- 
2.19.0

[PATCH v5 04/20] perf record: stop threads in the end of trace streaming

[Alexey Bayduraev]

Signal thread to terminate by closing write fd of msg pipe.
Receive THREAD_MSG__READY message as the confirmation of the
thread's termination. Stop threads created for parallel trace
streaming prior their stats processing.

Acked-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Alexey Bayduraev <alexey.v.bayduraev@linux.intel.com>
---
 tools/perf/builtin-record.c | 30 ++++++++++++++++++++++++++++++
 1 file changed, 30 insertions(+)

diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c
index 2a3dac894aa3..52703eceb47f 100644
--- a/tools/perf/builtin-record.c
+++ b/tools/perf/builtin-record.c
@@ -112,6 +112,16 @@ struct thread_data {
 
 static __thread struct thread_data *thread;
 
+enum thread_msg {
+	THREAD_MSG__UNDEFINED = 0,
+	THREAD_MSG__READY,
+	THREAD_MSG__MAX,
+};
+
+static const char *thread_msg_tags[THREAD_MSG__MAX] = {
+	"UNDEFINED", "READY"
+};
+
 struct record {
 	struct perf_tool	tool;
 	struct record_opts	opts;
@@ -1851,6 +1861,23 @@ static void record__uniquify_name(struct record *rec)
 	}
 }
 
+static int record__terminate_thread(struct thread_data *thread_data)
+{
+	int res;
+	enum thread_msg ack = THREAD_MSG__UNDEFINED;
+	pid_t tid = thread_data->tid;
+
+	close(thread_data->pipes.msg[1]);
+	res = read(thread_data->pipes.ack[0], &ack, sizeof(ack));
+	if (res != -1)
+		pr_debug2("threads[%d]: sent %s\n", tid, thread_msg_tags[ack]);
+	else
+		pr_err("threads[%d]: failed to recv msg=%s from tid=%d\n",
+		       thread->tid, thread_msg_tags[ack], tid);
+
+	return 0;
+}
+
 static int record__start_threads(struct record *rec)
 {
 	struct thread_data *thread_data = rec->thread_data;
@@ -1867,6 +1894,9 @@ static int record__stop_threads(struct record *rec, unsigned long *waking)
 	int t;
 	struct thread_data *thread_data = rec->thread_data;
 
+	for (t = 1; t < rec->nr_threads; t++)
+		record__terminate_thread(&thread_data[t]);
+
 	for (t = 0; t < rec->nr_threads; t++) {
 		rec->samples += thread_data[t].samples;
 		*waking += thread_data[t].waking;
-- 
2.19.0

[PATCH v5 05/20] perf record: start threads in the beginning of trace streaming

[Alexey Bayduraev]

Start thread in detached state because its management is implemented
via messaging to avoid any scaling issues. Block signals prior thread
start so only main tool thread would be notified on external async
signals during data collection. Thread affinity mask is used to assign
eligible cpus for the thread to run. Wait and sync on thread start using
thread ack pipe.

Signed-off-by: Alexey Bayduraev <alexey.v.bayduraev@linux.intel.com>
---
 tools/perf/builtin-record.c | 105 +++++++++++++++++++++++++++++++++++-
 1 file changed, 104 insertions(+), 1 deletion(-)

diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c
index 52703eceb47f..baf9646c4ac9 100644
--- a/tools/perf/builtin-record.c
+++ b/tools/perf/builtin-record.c
@@ -1417,6 +1417,65 @@ static void record__thread_munmap_filtered(struct fdarray *fda, int fd,
 		perf_mmap__put(map);
 }
 
+static void *record__thread(void *arg)
+{
+	enum thread_msg msg = THREAD_MSG__READY;
+	bool terminate = false;
+	struct fdarray *pollfd;
+	int err, ctlfd_pos;
+
+	thread = arg;
+	thread->tid = syscall(SYS_gettid);
+
+	err = write(thread->pipes.ack[1], &msg, sizeof(msg));
+	if (err == -1)
+		pr_err("threads[%d]: failed to notify on start: %s", thread->tid, strerror(errno));
+
+	pr_debug("threads[%d]: started on cpu=%d\n", thread->tid, sched_getcpu());
+
+	pollfd = &thread->pollfd;
+	ctlfd_pos = thread->ctlfd_pos;
+
+	for (;;) {
+		unsigned long long hits = thread->samples;
+
+		if (record__mmap_read_all(thread->rec, false) < 0 || terminate)
+			break;
+
+		if (hits == thread->samples) {
+
+			err = fdarray__poll(pollfd, -1);
+			/*
+			 * Propagate error, only if there's any. Ignore positive
+			 * number of returned events and interrupt error.
+			 */
+			if (err > 0 || (err < 0 && errno == EINTR))
+				err = 0;
+			thread->waking++;
+
+			if (fdarray__filter(pollfd, POLLERR | POLLHUP,
+					    record__thread_munmap_filtered, NULL) == 0)
+				break;
+		}
+
+		if (pollfd->entries[ctlfd_pos].revents & POLLHUP) {
+			terminate = true;
+			close(thread->pipes.msg[0]);
+			pollfd->entries[ctlfd_pos].fd = -1;
+			pollfd->entries[ctlfd_pos].events = 0;
+		}
+
+		pollfd->entries[ctlfd_pos].revents = 0;
+	}
+
+	err = write(thread->pipes.ack[1], &msg, sizeof(msg));
+	if (err == -1)
+		pr_err("threads[%d]: failed to notify on termination: %s",
+		       thread->tid, strerror(errno));
+
+	return NULL;
+}
+
 static void record__init_features(struct record *rec)
 {
 	struct perf_session *session = rec->session;
@@ -1880,13 +1939,57 @@ static int record__terminate_thread(struct thread_data *thread_data)
 
 static int record__start_threads(struct record *rec)
 {
+	int t, tt, ret = 0, nr_threads = rec->nr_threads;
 	struct thread_data *thread_data = rec->thread_data;
+	sigset_t full, mask;
+	pthread_t handle;
+	pthread_attr_t attrs;
+
+	sigfillset(&full);
+	if (sigprocmask(SIG_SETMASK, &full, &mask)) {
+		pr_err("Failed to block signals on threads start: %s\n", strerror(errno));
+		return -1;
+	}
+
+	pthread_attr_init(&attrs);
+	pthread_attr_setdetachstate(&attrs, PTHREAD_CREATE_DETACHED);
+
+	for (t = 1; t < nr_threads; t++) {
+		enum thread_msg msg = THREAD_MSG__UNDEFINED;
+
+		pthread_attr_setaffinity_np(&attrs,
+					    MMAP_CPU_MASK_BYTES(&(thread_data[t].mask->affinity)),
+					    (cpu_set_t *)(thread_data[t].mask->affinity.bits));
+
+		if (pthread_create(&handle, &attrs, record__thread, &thread_data[t])) {
+			for (tt = 1; tt < t; tt++)
+				record__terminate_thread(&thread_data[t]);
+			pr_err("Failed to start threads: %s\n", strerror(errno));
+			ret = -1;
+			goto out_err;
+		}
+
+		if (read(thread_data[t].pipes.ack[0], &msg, sizeof(msg)) > 0)
+			pr_debug2("threads[%d]: sent %s\n", rec->thread_data[t].tid,
+				 thread_msg_tags[msg]);
+	}
+
+	if (nr_threads > 1) {
+		sched_setaffinity(0, MMAP_CPU_MASK_BYTES(&thread_data[0].mask->affinity),
+				  (cpu_set_t *)thread_data[0].mask->affinity.bits);
+	}
 
 	thread = &thread_data[0];
 
 	pr_debug("threads[%d]: started on cpu=%d\n", thread->tid, sched_getcpu());
 
-	return 0;
+out_err:
+	if (sigprocmask(SIG_SETMASK, &mask, NULL)) {
+		pr_err("Failed to unblock signals on threads start: %s\n", strerror(errno));
+		ret = -1;
+	}
+
+	return ret;
 }
 
 static int record__stop_threads(struct record *rec, unsigned long *waking)
-- 
2.19.0

[PATCH v5 06/20] perf record: introduce data file at mmap buffer object

[Alexey Bayduraev]

Introduce data file and compressor objects into mmap object so
they could be used to process and store data stream from the
corresponding kernel data buffer. Make use of the introduced
per mmap file and compressor when they are initialized and
available.

Acked-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Alexey Bayduraev <alexey.v.bayduraev@linux.intel.com>
---
 tools/perf/builtin-record.c | 3 +++
 tools/perf/util/mmap.c      | 6 ++++++
 tools/perf/util/mmap.h      | 3 +++
 3 files changed, 12 insertions(+)

diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c
index baf9646c4ac9..b0ca61f1d265 100644
--- a/tools/perf/builtin-record.c
+++ b/tools/perf/builtin-record.c
@@ -190,6 +190,9 @@ static int record__write(struct record *rec, struct mmap *map __maybe_unused,
 {
 	struct perf_data_file *file = &rec->session->data->file;
 
+	if (map && map->file)
+		file = map->file;
+
 	if (perf_data_file__write(file, bf, size) < 0) {
 		pr_err("failed to write perf data, error: %m\n");
 		return -1;
diff --git a/tools/perf/util/mmap.c b/tools/perf/util/mmap.c
index ab7108d22428..a2c5e4237592 100644
--- a/tools/perf/util/mmap.c
+++ b/tools/perf/util/mmap.c
@@ -230,6 +230,8 @@ void mmap__munmap(struct mmap *map)
 {
 	bitmap_free(map->affinity_mask.bits);
 
+	zstd_fini(&map->zstd_data);
+
 	perf_mmap__aio_munmap(map);
 	if (map->data != NULL) {
 		munmap(map->data, mmap__mmap_len(map));
@@ -291,6 +293,10 @@ int mmap__mmap(struct mmap *map, struct mmap_params *mp, int fd, int cpu)
 	map->core.flush = mp->flush;
 
 	map->comp_level = mp->comp_level;
+	if (zstd_init(&map->zstd_data, map->comp_level)) {
+		pr_debug2("failed to init mmap commpressor, error %d\n", errno);
+		return -1;
+	}
 
 	if (map->comp_level && !perf_mmap__aio_enabled(map)) {
 		map->data = mmap(NULL, mmap__mmap_len(map), PROT_READ|PROT_WRITE,
diff --git a/tools/perf/util/mmap.h b/tools/perf/util/mmap.h
index 9d5f589f02ae..c4aed6e89549 100644
--- a/tools/perf/util/mmap.h
+++ b/tools/perf/util/mmap.h
@@ -13,6 +13,7 @@
 #endif
 #include "auxtrace.h"
 #include "event.h"
+#include "util/compress.h"
 
 struct aiocb;
 
@@ -43,6 +44,8 @@ struct mmap {
 	struct mmap_cpu_mask	affinity_mask;
 	void		*data;
 	int		comp_level;
+	struct perf_data_file *file;
+	struct zstd_data      zstd_data;
 };
 
 struct mmap_params {
-- 
2.19.0

[PATCH v5 07/20] perf record: introduce data transferred and compressed stats

[Alexey Bayduraev]

Introduce bytes_transferred and bytes_compressed stats so they
would capture statistics for the related data buffer transfers.

Acked-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Alexey Bayduraev <alexey.v.bayduraev@linux.intel.com>
---
 tools/perf/builtin-record.c | 64 +++++++++++++++++++++++++++++--------
 tools/perf/util/mmap.h      |  3 ++
 2 files changed, 54 insertions(+), 13 deletions(-)

diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c
index b0ca61f1d265..75cebec57357 100644
--- a/tools/perf/builtin-record.c
+++ b/tools/perf/builtin-record.c
@@ -198,6 +198,11 @@ static int record__write(struct record *rec, struct mmap *map __maybe_unused,
 		return -1;
 	}
 
+	if (map && map->file) {
+		map->bytes_written += size;
+		return 0;
+	}
+
 	rec->bytes_written += size;
 
 	if (record__output_max_size_exceeded(rec) && !done) {
@@ -215,8 +220,8 @@ static int record__write(struct record *rec, struct mmap *map __maybe_unused,
 
 static int record__aio_enabled(struct record *rec);
 static int record__comp_enabled(struct record *rec);
-static size_t zstd_compress(struct perf_session *session, void *dst, size_t dst_size,
-			    void *src, size_t src_size);
+static size_t zstd_compress(struct zstd_data *data,
+			    void *dst, size_t dst_size, void *src, size_t src_size);
 
 #ifdef HAVE_AIO_SUPPORT
 static int record__aio_write(struct aiocb *cblock, int trace_fd,
@@ -350,9 +355,13 @@ static int record__aio_pushfn(struct mmap *map, void *to, void *buf, size_t size
 	 */
 
 	if (record__comp_enabled(aio->rec)) {
-		size = zstd_compress(aio->rec->session, aio->data + aio->size,
-				     mmap__mmap_len(map) - aio->size,
+		struct zstd_data *zstd_data = &aio->rec->session->zstd_data;
+
+		aio->rec->session->bytes_transferred += size;
+		size = zstd_compress(zstd_data,
+				     aio->data + aio->size, mmap__mmap_len(map) - aio->size,
 				     buf, size);
+		aio->rec->session->bytes_compressed += size;
 	} else {
 		memcpy(aio->data + aio->size, buf, size);
 	}
@@ -577,8 +586,22 @@ static int record__pushfn(struct mmap *map, void *to, void *bf, size_t size)
 	struct record *rec = to;
 
 	if (record__comp_enabled(rec)) {
-		size = zstd_compress(rec->session, map->data, mmap__mmap_len(map), bf, size);
+		struct zstd_data *zstd_data = &rec->session->zstd_data;
+
+		if (map->file) {
+			zstd_data = &map->zstd_data;
+			map->bytes_transferred += size;
+		} else {
+			rec->session->bytes_transferred += size;
+		}
+
+		size = zstd_compress(zstd_data, map->data, mmap__mmap_len(map), bf, size);
 		bf   = map->data;
+
+		if (map->file)
+			map->bytes_compressed += size;
+		else
+			rec->session->bytes_compressed += size;
 	}
 
 	thread->samples++;
@@ -1305,18 +1328,15 @@ static size_t process_comp_header(void *record, size_t increment)
 	return size;
 }
 
-static size_t zstd_compress(struct perf_session *session, void *dst, size_t dst_size,
+static size_t zstd_compress(struct zstd_data *zstd_data, void *dst, size_t dst_size,
 			    void *src, size_t src_size)
 {
 	size_t compressed;
 	size_t max_record_size = PERF_SAMPLE_MAX_SIZE - sizeof(struct perf_record_compressed) - 1;
 
-	compressed = zstd_compress_stream_to_records(&session->zstd_data, dst, dst_size, src, src_size,
+	compressed = zstd_compress_stream_to_records(zstd_data, dst, dst_size, src, src_size,
 						     max_record_size, process_comp_header);
 
-	session->bytes_transferred += src_size;
-	session->bytes_compressed  += compressed;
-
 	return compressed;
 }
 
@@ -1997,8 +2017,10 @@ static int record__start_threads(struct record *rec)
 
 static int record__stop_threads(struct record *rec, unsigned long *waking)
 {
-	int t;
+	int t, tm;
+	struct mmap *map, *overwrite_map;
 	struct thread_data *thread_data = rec->thread_data;
+	u64 bytes_written = 0, bytes_transferred = 0, bytes_compressed = 0;
 
 	for (t = 1; t < rec->nr_threads; t++)
 		record__terminate_thread(&thread_data[t]);
@@ -2006,9 +2028,25 @@ static int record__stop_threads(struct record *rec, unsigned long *waking)
 	for (t = 0; t < rec->nr_threads; t++) {
 		rec->samples += thread_data[t].samples;
 		*waking += thread_data[t].waking;
-		pr_debug("threads[%d]: samples=%lld, wakes=%ld, trasferred=%ld, compressed=%ld\n",
+		for (tm = 0; tm < thread_data[t].nr_mmaps; tm++) {
+			if (thread_data[t].maps) {
+				map = thread_data[t].maps[tm];
+				bytes_transferred += map->bytes_transferred;
+				bytes_compressed += map->bytes_compressed;
+				bytes_written += map->bytes_written;
+			}
+			if (thread_data[t].overwrite_maps) {
+				overwrite_map = thread_data[t].overwrite_maps[tm];
+				bytes_transferred += overwrite_map->bytes_transferred;
+				bytes_compressed += overwrite_map->bytes_compressed;
+				bytes_written += overwrite_map->bytes_written;
+			}
+		}
+		rec->session->bytes_transferred += bytes_transferred;
+		rec->session->bytes_compressed += bytes_compressed;
+		pr_debug("threads[%d]: samples=%lld, wakes=%ld, trasferred=%ld, compressed=%ld, written=%ld\n",
 			 thread_data[t].tid, thread_data[t].samples, thread_data[t].waking,
-			 rec->session->bytes_transferred, rec->session->bytes_compressed);
+			 bytes_transferred, bytes_compressed, bytes_written);
 	}
 
 	return 0;
diff --git a/tools/perf/util/mmap.h b/tools/perf/util/mmap.h
index c4aed6e89549..c04ca4b5adf5 100644
--- a/tools/perf/util/mmap.h
+++ b/tools/perf/util/mmap.h
@@ -46,6 +46,9 @@ struct mmap {
 	int		comp_level;
 	struct perf_data_file *file;
 	struct zstd_data      zstd_data;
+	u64		      bytes_transferred;
+	u64		      bytes_compressed;
+	u64		      bytes_written;
 };
 
 struct mmap_params {
-- 
2.19.0

[PATCH v5 08/20] perf record: init data file at mmap buffer object

[Alexey Bayduraev]

Initialize data files located at mmap buffer objects so trace data
can be written into several data file located at data directory.

Acked-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Alexey Bayduraev <alexey.v.bayduraev@linux.intel.com>
---
 tools/perf/builtin-record.c | 41 ++++++++++++++++++++++++++++++-------
 tools/perf/util/record.h    |  1 +
 2 files changed, 35 insertions(+), 7 deletions(-)

diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c
index 75cebec57357..bf730e1220dc 100644
--- a/tools/perf/builtin-record.c
+++ b/tools/perf/builtin-record.c
@@ -160,6 +160,11 @@ static const char *affinity_tags[PERF_AFFINITY_MAX] = {
 	"SYS", "NODE", "CPU"
 };
 
+static int record__threads_enabled(struct record *rec)
+{
+	return rec->opts.threads_spec;
+}
+
 static bool switch_output_signal(struct record *rec)
 {
 	return rec->switch_output.signal &&
@@ -1070,7 +1075,7 @@ static void record__free_thread_data(struct record *rec)
 static int record__mmap_evlist(struct record *rec,
 			       struct evlist *evlist)
 {
-	int ret;
+	int i, ret;
 	struct record_opts *opts = &rec->opts;
 	bool auxtrace_overwrite = opts->auxtrace_snapshot_mode ||
 				  opts->auxtrace_sample_mode;
@@ -1109,6 +1114,18 @@ static int record__mmap_evlist(struct record *rec,
 	if (ret)
 		return ret;
 
+	if (record__threads_enabled(rec)) {
+		ret = perf_data__create_dir(&rec->data, evlist->core.nr_mmaps);
+		if (ret)
+			return ret;
+		for (i = 0; i < evlist->core.nr_mmaps; i++) {
+			if (evlist->mmap)
+				evlist->mmap[i].file = &rec->data.dir.files[i];
+			if (evlist->overwrite_mmap)
+				evlist->overwrite_mmap[i].file = &rec->data.dir.files[i];
+		}
+	}
+
 	return 0;
 }
 
@@ -1410,8 +1427,12 @@ static int record__mmap_read_evlist(struct record *rec, struct evlist *evlist,
 	/*
 	 * Mark the round finished in case we wrote
 	 * at least one event.
+	 *
+	 * No need for round events in directory mode,
+	 * because per-cpu maps and files have data
+	 * sorted by kernel.
 	 */
-	if (bytes_written != rec->bytes_written)
+	if (!record__threads_enabled(rec) && bytes_written != rec->bytes_written)
 		rc = record__write(rec, NULL, &finished_round_event, sizeof(finished_round_event));
 
 	if (overwrite)
@@ -1525,7 +1546,9 @@ static void record__init_features(struct record *rec)
 	if (!rec->opts.use_clockid)
 		perf_header__clear_feat(&session->header, HEADER_CLOCK_DATA);
 
-	perf_header__clear_feat(&session->header, HEADER_DIR_FORMAT);
+	if (!record__threads_enabled(rec))
+		perf_header__clear_feat(&session->header, HEADER_DIR_FORMAT);
+
 	if (!record__comp_enabled(rec))
 		perf_header__clear_feat(&session->header, HEADER_COMPRESSED);
 
@@ -1536,15 +1559,21 @@ static void
 record__finish_output(struct record *rec)
 {
 	struct perf_data *data = &rec->data;
-	int fd = perf_data__fd(data);
+	int i, fd = perf_data__fd(data);
 
 	if (data->is_pipe)
 		return;
 
 	rec->session->header.data_size += rec->bytes_written;
 	data->file.size = lseek(perf_data__fd(data), 0, SEEK_CUR);
+	if (record__threads_enabled(rec)) {
+		for (i = 0; i < data->dir.nr; i++)
+			data->dir.files[i].size = lseek(data->dir.files[i].fd, 0, SEEK_CUR);
+	}
 
 	if (!rec->no_buildid) {
+		/* this will be recalculated during process_buildids() */
+		rec->samples = 0;
 		process_buildids(rec);
 
 		if (rec->buildid_all)
@@ -2480,8 +2509,6 @@ static int __cmd_record(struct record *rec, int argc, const char **argv)
 		status = err;
 
 	record__synthesize(rec, true);
-	/* this will be recalculated during process_buildids() */
-	rec->samples = 0;
 
 	if (!err) {
 		if (!rec->timestamp_filename) {
@@ -3245,7 +3272,7 @@ int cmd_record(int argc, const char **argv)
 		rec->no_buildid = true;
 	}
 
-	if (rec->opts.kcore)
+	if (rec->opts.kcore || record__threads_enabled(rec))
 		rec->data.is_dir = true;
 
 	if (rec->opts.comp_level != 0) {
diff --git a/tools/perf/util/record.h b/tools/perf/util/record.h
index 68f471d9a88b..4d68b7e27272 100644
--- a/tools/perf/util/record.h
+++ b/tools/perf/util/record.h
@@ -77,6 +77,7 @@ struct record_opts {
 	int	      ctl_fd;
 	int	      ctl_fd_ack;
 	bool	      ctl_fd_close;
+	int	      threads_spec;
 };
 
 extern const char * const *record_usage;
-- 
2.19.0

[PATCH v5 09/20] tools lib: introduce bitmap_intersects() operation

[Alexey Bayduraev]

Introduce bitmap_intersects() routine that tests whether
bitmaps bitmap1 and bitmap2 intersects. This routine will
be used during thread masks initialization.

Acked-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Alexey Bayduraev <alexey.v.bayduraev@linux.intel.com>
---
 tools/include/linux/bitmap.h | 11 +++++++++++
 tools/lib/bitmap.c           | 14 ++++++++++++++
 2 files changed, 25 insertions(+)

diff --git a/tools/include/linux/bitmap.h b/tools/include/linux/bitmap.h
index 477a1cae513f..2eb1d1084543 100644
--- a/tools/include/linux/bitmap.h
+++ b/tools/include/linux/bitmap.h
@@ -18,6 +18,8 @@ int __bitmap_and(unsigned long *dst, const unsigned long *bitmap1,
 int __bitmap_equal(const unsigned long *bitmap1,
 		   const unsigned long *bitmap2, unsigned int bits);
 void bitmap_clear(unsigned long *map, unsigned int start, int len);
+int __bitmap_intersects(const unsigned long *bitmap1,
+			const unsigned long *bitmap2, unsigned int bits);
 
 #define BITMAP_FIRST_WORD_MASK(start) (~0UL << ((start) & (BITS_PER_LONG - 1)))
 
@@ -178,4 +180,13 @@ static inline int bitmap_equal(const unsigned long *src1,
 	return __bitmap_equal(src1, src2, nbits);
 }
 
+static inline int bitmap_intersects(const unsigned long *src1,
+			const unsigned long *src2, unsigned int nbits)
+{
+	if (small_const_nbits(nbits))
+		return ((*src1 & *src2) & BITMAP_LAST_WORD_MASK(nbits)) != 0;
+	else
+		return __bitmap_intersects(src1, src2, nbits);
+}
+
 #endif /* _PERF_BITOPS_H */
diff --git a/tools/lib/bitmap.c b/tools/lib/bitmap.c
index 5043747ef6c5..3cc3a5b43bb5 100644
--- a/tools/lib/bitmap.c
+++ b/tools/lib/bitmap.c
@@ -86,3 +86,17 @@ int __bitmap_equal(const unsigned long *bitmap1,
 
 	return 1;
 }
+
+int __bitmap_intersects(const unsigned long *bitmap1,
+			const unsigned long *bitmap2, unsigned int bits)
+{
+	unsigned int k, lim = bits/BITS_PER_LONG;
+	for (k = 0; k < lim; ++k)
+		if (bitmap1[k] & bitmap2[k])
+			return 1;
+
+	if (bits % BITS_PER_LONG)
+		if ((bitmap1[k] & bitmap2[k]) & BITMAP_LAST_WORD_MASK(bits))
+			return 1;
+	return 0;
+}
-- 
2.19.0

[PATCH v5 10/20] perf record: introduce --threads=<spec> command line option

[Alexey Bayduraev]

Provide --threads option in perf record command line interface.
The option can have a value in the form of masks that specify
cpus to be monitored with data streaming threads and its layout
in system topology. The masks can be filtered using cpu mask
provided via -C option.

The specification value can be user defined list of masks. Masks
separated by colon define cpus to be monitored by one thread and
affinity mask of that thread is separated by slash. For example:
<cpus mask 1>/<affinity mask 1>:<cpu mask 2>/<affinity mask 2>
specifies parallel threads layout that consists of two threads
with corresponding assigned cpus to be monitored.

The specification value can be a string e.g. "cpu", "core" or
"socket" meaning creation of data streaming thread for every
cpu or core or socket to monitor distinct cpus or cpus grouped
by core or socket.

The option provided with no or empty value defaults to per-cpu
parallel threads layout creating data streaming thread for every
cpu being monitored.

Feature design and implementation are based on prototypes [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-jolsa@kernel.org/

Suggested-by: Jiri Olsa <jolsa@kernel.org>
Suggested-by: Namhyung Kim <namhyung@kernel.org>
Acked-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Alexey Bayduraev <alexey.v.bayduraev@linux.intel.com>
---
 tools/perf/builtin-record.c | 342 +++++++++++++++++++++++++++++++++++-
 tools/perf/util/record.h    |   1 +
 2 files changed, 341 insertions(+), 2 deletions(-)

diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c
index bf730e1220dc..e5ea1334cc7d 100644
--- a/tools/perf/builtin-record.c
+++ b/tools/perf/builtin-record.c
@@ -51,6 +51,7 @@
 #include "util/evlist-hybrid.h"
 #include "asm/bug.h"
 #include "perf.h"
+#include "cputopo.h"
 
 #include <errno.h>
 #include <inttypes.h>
@@ -122,6 +123,20 @@ static const char *thread_msg_tags[THREAD_MSG__MAX] = {
 	"UNDEFINED", "READY"
 };
 
+enum thread_spec {
+	THREAD_SPEC__UNDEFINED = 0,
+	THREAD_SPEC__CPU,
+	THREAD_SPEC__CORE,
+	THREAD_SPEC__SOCKET,
+	THREAD_SPEC__NUMA,
+	THREAD_SPEC__USER,
+	THREAD_SPEC__MAX,
+};
+
+static const char *thread_spec_tags[THREAD_SPEC__MAX] = {
+	"undefined", "cpu", "core", "socket", "numa", "user"
+};
+
 struct record {
 	struct perf_tool	tool;
 	struct record_opts	opts;
@@ -2704,6 +2719,70 @@ static void record__thread_mask_free(struct thread_mask *mask)
 	record__mmap_cpu_mask_free(&mask->affinity);
 }
 
+static int record__thread_mask_or(struct thread_mask *dest, struct thread_mask *src1,
+				   struct thread_mask *src2)
+{
+	if (src1->maps.nbits != src2->maps.nbits ||
+	    dest->maps.nbits != src1->maps.nbits ||
+	    src1->affinity.nbits != src2->affinity.nbits ||
+	    dest->affinity.nbits != src1->affinity.nbits)
+		return -EINVAL;
+
+	bitmap_or(dest->maps.bits, src1->maps.bits,
+		  src2->maps.bits, src1->maps.nbits);
+	bitmap_or(dest->affinity.bits, src1->affinity.bits,
+		  src2->affinity.bits, src1->affinity.nbits);
+
+	return 0;
+}
+
+static int record__thread_mask_intersects(struct thread_mask *mask_1, struct thread_mask *mask_2)
+{
+	int res1, res2;
+
+	if (mask_1->maps.nbits != mask_2->maps.nbits ||
+	    mask_1->affinity.nbits != mask_2->affinity.nbits)
+		return -EINVAL;
+
+	res1 = bitmap_intersects(mask_1->maps.bits, mask_2->maps.bits,
+				 mask_1->maps.nbits);
+	res2 = bitmap_intersects(mask_1->affinity.bits, mask_2->affinity.bits,
+				 mask_1->affinity.nbits);
+	if (res1 || res2)
+		return 1;
+
+	return 0;
+}
+
+static int record__parse_threads(const struct option *opt, const char *str, int unset)
+{
+	int s;
+	struct record_opts *opts = opt->value;
+
+	if (unset || !str || !strlen(str)) {
+		opts->threads_spec = THREAD_SPEC__CPU;
+	} else {
+		for (s = 1; s < THREAD_SPEC__MAX; s++) {
+			if (s == THREAD_SPEC__USER) {
+				opts->threads_user_spec = strdup(str);
+				opts->threads_spec = THREAD_SPEC__USER;
+				break;
+			}
+			if (!strncasecmp(str, thread_spec_tags[s], strlen(thread_spec_tags[s]))) {
+				opts->threads_spec = s;
+				break;
+			}
+		}
+	}
+
+	pr_debug("threads_spec: %s", thread_spec_tags[opts->threads_spec]);
+	if (opts->threads_spec == THREAD_SPEC__USER)
+		pr_debug("=[%s]", opts->threads_user_spec);
+	pr_debug("\n");
+
+	return 0;
+}
+
 static int parse_output_max_size(const struct option *opt,
 				 const char *str, int unset)
 {
@@ -3134,6 +3213,9 @@ static struct option __record_options[] = {
 		     "\t\t\t  Optionally send control command completion ('ack\\n') to ack-fd descriptor.\n"
 		     "\t\t\t  Alternatively, ctl-fifo / ack-fifo will be opened and used as ctl-fd / ack-fd.",
 		      parse_control_option),
+	OPT_CALLBACK_OPTARG(0, "threads", &record.opts, NULL, "spec",
+			    "write collected trace data into several data files using parallel threads",
+			    record__parse_threads),
 	OPT_END()
 };
 
@@ -3147,6 +3229,17 @@ static void record__mmap_cpu_mask_init(struct mmap_cpu_mask *mask, struct perf_c
 		set_bit(cpus->map[c], mask->bits);
 }
 
+static void record__mmap_cpu_mask_init_spec(struct mmap_cpu_mask *mask, char *mask_spec)
+{
+	struct perf_cpu_map *cpus;
+
+	cpus = perf_cpu_map__new(mask_spec);
+	if (cpus) {
+		record__mmap_cpu_mask_init(mask, cpus);
+		perf_cpu_map__put(cpus);
+	}
+}
+
 static int record__alloc_thread_masks(struct record *rec, int nr_threads, int nr_bits)
 {
 	int t, ret;
@@ -3166,6 +3259,224 @@ static int record__alloc_thread_masks(struct record *rec, int nr_threads, int nr
 
 	return 0;
 }
+
+static int record__init_thread_cpu_masks(struct record *rec, struct perf_cpu_map *cpus)
+{
+	int t, ret, nr_cpus = perf_cpu_map__nr(cpus);
+
+	ret = record__alloc_thread_masks(rec, nr_cpus, cpu__max_cpu());
+	if (ret)
+		return ret;
+
+	rec->nr_threads = nr_cpus;
+	pr_debug("threads: nr_threads=%d\n", rec->nr_threads);
+
+	for (t = 0; t < rec->nr_threads; t++) {
+		set_bit(cpus->map[t], rec->thread_masks[t].maps.bits);
+		pr_debug("thread_masks[%d]: maps mask [%d]\n", t, cpus->map[t]);
+		set_bit(cpus->map[t], rec->thread_masks[t].affinity.bits);
+		pr_debug("thread_masks[%d]: affinity mask [%d]\n", t, cpus->map[t]);
+	}
+
+	return 0;
+}
+
+static int record__init_thread_masks_spec(struct record *rec, struct perf_cpu_map *cpus,
+					  char **maps_spec, char **affinity_spec, u32 nr_spec)
+{
+	u32 s;
+	int ret = 0, nr_threads = 0;
+	struct mmap_cpu_mask cpus_mask;
+	struct thread_mask thread_mask, full_mask, *prev_masks;
+
+	ret = record__mmap_cpu_mask_alloc(&cpus_mask, cpu__max_cpu());
+	if (ret)
+		goto out;
+	record__mmap_cpu_mask_init(&cpus_mask, cpus);
+	ret = record__thread_mask_alloc(&thread_mask, cpu__max_cpu());
+	if (ret)
+		goto out_free_cpu_mask;
+	ret = record__thread_mask_alloc(&full_mask, cpu__max_cpu());
+	if (ret)
+		goto out_free_thread_mask;
+	record__thread_mask_clear(&full_mask);
+
+	for (s = 0; s < nr_spec; s++) {
+		record__thread_mask_clear(&thread_mask);
+
+		record__mmap_cpu_mask_init_spec(&thread_mask.maps, maps_spec[s]);
+		record__mmap_cpu_mask_init_spec(&thread_mask.affinity, affinity_spec[s]);
+
+		if (!bitmap_and(thread_mask.maps.bits, thread_mask.maps.bits,
+				cpus_mask.bits, thread_mask.maps.nbits) ||
+		    !bitmap_and(thread_mask.affinity.bits, thread_mask.affinity.bits,
+				cpus_mask.bits, thread_mask.affinity.nbits))
+			continue;
+
+		ret = record__thread_mask_intersects(&thread_mask, &full_mask);
+		if (ret)
+			goto out_free_full_mask;
+		record__thread_mask_or(&full_mask, &full_mask, &thread_mask);
+
+		prev_masks = rec->thread_masks;
+		rec->thread_masks = realloc(rec->thread_masks,
+					    (nr_threads + 1) * sizeof(struct thread_mask));
+		if (!rec->thread_masks) {
+			pr_err("Failed to allocate thread masks\n");
+			rec->thread_masks = prev_masks;
+			ret = -ENOMEM;
+			goto out_free_full_mask;
+		}
+		rec->thread_masks[nr_threads] = thread_mask;
+		pr_debug("thread_masks[%d]: addr=", nr_threads);
+		mmap_cpu_mask__scnprintf(&rec->thread_masks[nr_threads].maps, "maps");
+		pr_debug("thread_masks[%d]: addr=", nr_threads);
+		mmap_cpu_mask__scnprintf(&rec->thread_masks[nr_threads].affinity, "affinity");
+		nr_threads++;
+		ret = record__thread_mask_alloc(&thread_mask, cpu__max_cpu());
+		if (ret)
+			goto out_free_full_mask;
+	}
+
+	rec->nr_threads = nr_threads;
+	pr_debug("threads: nr_threads=%d\n", rec->nr_threads);
+
+out_free_full_mask:
+	record__thread_mask_free(&full_mask);
+out_free_thread_mask:
+	record__thread_mask_free(&thread_mask);
+out_free_cpu_mask:
+	record__mmap_cpu_mask_free(&cpus_mask);
+out:
+	return ret;
+}
+
+static int record__init_thread_core_masks(struct record *rec, struct perf_cpu_map *cpus)
+{
+	int ret;
+	struct cpu_topology *topo;
+
+	topo = cpu_topology__new();
+	if (!topo)
+		return -EINVAL;
+
+	ret = record__init_thread_masks_spec(rec, cpus, topo->thread_siblings,
+					     topo->thread_siblings, topo->thread_sib);
+	cpu_topology__delete(topo);
+
+	return ret;
+}
+
+static int record__init_thread_socket_masks(struct record *rec, struct perf_cpu_map *cpus)
+{
+	int ret;
+	struct cpu_topology *topo;
+
+	topo = cpu_topology__new();
+	if (!topo)
+		return -EINVAL;
+
+	ret = record__init_thread_masks_spec(rec, cpus, topo->core_siblings,
+					     topo->core_siblings, topo->core_sib);
+	cpu_topology__delete(topo);
+
+	return ret;
+}
+
+static int record__init_thread_numa_masks(struct record *rec, struct perf_cpu_map *cpus)
+{
+	u32 s;
+	int ret;
+	char **spec;
+	struct numa_topology *topo;
+
+	topo = numa_topology__new();
+	if (!topo)
+		return -EINVAL;
+	spec = zalloc(topo->nr * sizeof(char *));
+	if (!spec) {
+		ret = -ENOMEM;
+		goto out_delete_topo;
+	}
+	for (s = 0; s < topo->nr; s++)
+		spec[s] = topo->nodes[s].cpus;
+
+	ret = record__init_thread_masks_spec(rec, cpus, spec, spec, topo->nr);
+
+	zfree(&spec);
+
+out_delete_topo:
+	numa_topology__delete(topo);
+
+	return ret;
+}
+
+static int record__init_thread_user_masks(struct record *rec, struct perf_cpu_map *cpus)
+{
+	int t, ret;
+	u32 s, nr_spec = 0;
+	char **maps_spec = NULL, **affinity_spec = NULL, **prev_spec;
+	char *spec, *spec_ptr, *user_spec, *mask, *mask_ptr;
+
+	for (t = 0, user_spec = (char *)rec->opts.threads_user_spec; ; t++, user_spec = NULL) {
+		spec = strtok_r(user_spec, ":", &spec_ptr);
+		if (spec == NULL)
+			break;
+		pr_debug(" spec[%d]: %s\n", t, spec);
+		mask = strtok_r(spec, "/", &mask_ptr);
+		if (mask == NULL)
+			break;
+		pr_debug("  maps mask: %s\n", mask);
+		prev_spec = maps_spec;
+		maps_spec = realloc(maps_spec, (nr_spec + 1) * sizeof(char *));
+		if (!maps_spec) {
+			pr_err("Failed to realloc maps_spec\n");
+			maps_spec = prev_spec;
+			ret = -ENOMEM;
+			goto out_free_all_specs;
+		}
+		maps_spec[nr_spec] = strdup(mask);
+		if (!maps_spec[nr_spec]) {
+			pr_err("Failed to alloc maps_spec[%d]\n", nr_spec);
+			ret = -ENOMEM;
+			goto out_free_all_specs;
+		}
+		mask = strtok_r(NULL, "/", &mask_ptr);
+		if (mask == NULL)
+			break;
+		pr_debug("  affinity mask: %s\n", mask);
+		prev_spec = affinity_spec;
+		affinity_spec = realloc(affinity_spec, (nr_spec + 1) * sizeof(char *));
+		if (!affinity_spec) {
+			pr_err("Failed to realloc affinity_spec\n");
+			affinity_spec = prev_spec;
+			ret = -ENOMEM;
+			goto out_free_all_specs;
+		}
+		affinity_spec[nr_spec] = strdup(mask);
+		if (!affinity_spec[nr_spec]) {
+			pr_err("Failed to alloc affinity_spec[%d]\n", nr_spec);
+			ret = -ENOMEM;
+			goto out_free_all_specs;
+		}
+		nr_spec++;
+	}
+
+	ret = record__init_thread_masks_spec(rec, cpus, maps_spec, affinity_spec, nr_spec);
+
+out_free_all_specs:
+	for (s = 0; s < nr_spec; s++) {
+		if (maps_spec)
+			free(maps_spec[s]);
+		if (affinity_spec)
+			free(affinity_spec[s]);
+	}
+	free(affinity_spec);
+	free(maps_spec);
+
+	return ret;
+}
+
 static int record__init_thread_default_masks(struct record *rec, struct perf_cpu_map *cpus)
 {
 	int ret;
@@ -3183,9 +3494,33 @@ static int record__init_thread_default_masks(struct record *rec, struct perf_cpu
 
 static int record__init_thread_masks(struct record *rec)
 {
+	int ret = 0;
 	struct perf_cpu_map *cpus = rec->evlist->core.cpus;
 
-	return record__init_thread_default_masks(rec, cpus);
+	if (!record__threads_enabled(rec))
+		return record__init_thread_default_masks(rec, cpus);
+
+	switch (rec->opts.threads_spec) {
+	case THREAD_SPEC__CPU:
+		ret = record__init_thread_cpu_masks(rec, cpus);
+		break;
+	case THREAD_SPEC__CORE:
+		ret = record__init_thread_core_masks(rec, cpus);
+		break;
+	case THREAD_SPEC__SOCKET:
+		ret = record__init_thread_socket_masks(rec, cpus);
+		break;
+	case THREAD_SPEC__NUMA:
+		ret = record__init_thread_numa_masks(rec, cpus);
+		break;
+	case THREAD_SPEC__USER:
+		ret = record__init_thread_user_masks(rec, cpus);
+		break;
+	default:
+		break;
+	}
+
+	return ret;
 }
 
 static int record__fini_thread_masks(struct record *rec)
@@ -3436,7 +3771,10 @@ int cmd_record(int argc, const char **argv)
 
 	err = record__init_thread_masks(rec);
 	if (err) {
-		pr_err("record__init_thread_masks failed, error %d\n", err);
+		if (err > 0)
+			pr_err("ERROR: parallel data streaming masks (--threads) intersect.\n");
+		else
+			pr_err("record__init_thread_masks failed, error %d\n", err);
 		goto out;
 	}
 
diff --git a/tools/perf/util/record.h b/tools/perf/util/record.h
index 4d68b7e27272..3da156498f47 100644
--- a/tools/perf/util/record.h
+++ b/tools/perf/util/record.h
@@ -78,6 +78,7 @@ struct record_opts {
 	int	      ctl_fd_ack;
 	bool	      ctl_fd_close;
 	int	      threads_spec;
+	const char    *threads_user_spec;
 };
 
 extern const char * const *record_usage;
-- 
2.19.0

[PATCH v5 11/20] perf record: document parallel data streaming mode

[Alexey Bayduraev]

Document --threads option syntax and parallel data streaming modes
in Documentation/perf-record.txt. Implement compatibility checks for
other modes and related command line options: asynchronous(--aio)
trace streaming and affinity (--affinity) modes, pipe mode, AUX
area tracing --snapshot and --aux-sample options, --switch-output,
--switch-output-event, --switch-max-files and --timestamp-filename
options. Parallel data streaming is compatible with Zstd compression
(--compression-level) and external control commands (--control).
Cpu mask provided via -C option filters --threads specification masks.

Signed-off-by: Alexey Bayduraev <alexey.v.bayduraev@linux.intel.com>
---
 tools/perf/Documentation/perf-record.txt | 30 +++++++++++++++
 tools/perf/builtin-record.c              | 49 ++++++++++++++++++++++--
 2 files changed, 76 insertions(+), 3 deletions(-)

diff --git a/tools/perf/Documentation/perf-record.txt b/tools/perf/Documentation/perf-record.txt
index d71bac847936..2046b28d9822 100644
--- a/tools/perf/Documentation/perf-record.txt
+++ b/tools/perf/Documentation/perf-record.txt
@@ -695,6 +695,36 @@ measurements:
  wait -n ${perf_pid}
  exit $?
 
+--threads=<spec>::
+Write collected trace data into several data files using parallel threads.
+<spec> value can be user defined list of masks. Masks separated by colon
+define cpus to be monitored by a thread and affinity mask of that thread
+is separated by slash:
+
+    <cpus mask 1>/<affinity mask 1>:<cpus mask 2>/<affinity mask 2>:...
+
+For example user specification like the following:
+
+    0,2-4/2-4:1,5-7/5-7
+
+specifies parallel threads layout that consists of two threads,
+the first thread monitors cpus 0 and 2-4 with the affinity mask 2-4,
+the second monitors cpus 1 and 5-7 with the affinity mask 5-7.
+
+<spec> value can also be a string meaning predefined parallel threads
+layout:
+
+    cpu    - create new data streaming thread for every monitored cpu
+    core   - create new thread to monitor cpus grouped by a core
+    socket - create new thread to monitor cpus grouped by a socket
+    numa   - create new threed to monitor cpus grouped by a numa domain
+
+Predefined layouts can be used on systems with large number of cpus in
+order not to spawn multiple per-cpu streaming threads but still avoid LOST
+events in data directory files. Option specified with no or empty value
+defaults to cpu layout. Masks defined or provided by the option value are
+filtered through the mask provided by -C option.
+
 include::intel-hybrid.txt[]
 
 SEE ALSO
diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c
index e5ea1334cc7d..4d676a1be139 100644
--- a/tools/perf/builtin-record.c
+++ b/tools/perf/builtin-record.c
@@ -800,6 +800,12 @@ static int record__auxtrace_init(struct record *rec)
 {
 	int err;
 
+	if ((rec->opts.auxtrace_snapshot_opts || rec->opts.auxtrace_sample_opts)
+	    && record__threads_enabled(rec)) {
+		pr_err("AUX area tracing options are not available in parallel streaming mode.\n");
+		return -EINVAL;
+	}
+
 	if (!rec->itr) {
 		rec->itr = auxtrace_record__init(rec->evlist, &err);
 		if (err)
@@ -2145,6 +2151,17 @@ static int __cmd_record(struct record *rec, int argc, const char **argv)
 		return PTR_ERR(session);
 	}
 
+	if (record__threads_enabled(rec)) {
+		if (perf_data__is_pipe(&rec->data)) {
+			pr_err("Parallel trace streaming is not available in pipe mode.\n");
+			return -1;
+		}
+		if (rec->opts.full_auxtrace) {
+			pr_err("Parallel trace streaming is not available in AUX area tracing mode.\n");
+			return -1;
+		}
+	}
+
 	fd = perf_data__fd(data);
 	rec->session = session;
 
@@ -2903,12 +2920,22 @@ static int switch_output_setup(struct record *rec)
 	 * --switch-output=signal, as we'll send a SIGUSR2 from the side band
 	 *  thread to its parent.
 	 */
-	if (rec->switch_output_event_set)
+	if (rec->switch_output_event_set) {
+		if (record__threads_enabled(rec)) {
+			pr_warning("WARNING: --switch-output-event option is not available in parallel streaming mode.\n");
+			return 0;
+		}
 		goto do_signal;
+	}
 
 	if (!s->set)
 		return 0;
 
+	if (record__threads_enabled(rec)) {
+		pr_warning("WARNING: --switch-output option is not available in parallel streaming mode.\n");
+		return 0;
+	}
+
 	if (!strcmp(s->str, "signal")) {
 do_signal:
 		s->signal = true;
@@ -3193,8 +3220,8 @@ static struct option __record_options[] = {
 		     "Set affinity mask of trace reading thread to NUMA node cpu mask or cpu of processed mmap buffer",
 		     record__parse_affinity),
 #ifdef HAVE_ZSTD_SUPPORT
-	OPT_CALLBACK_OPTARG('z', "compression-level", &record.opts, &comp_level_default,
-			    "n", "Compressed records using specified level (default: 1 - fastest compression, 22 - greatest compression)",
+	OPT_CALLBACK_OPTARG('z', "compression-level", &record.opts, &comp_level_default, "n",
+			    "Compress records using specified level (default: 1 - fastest compression, 22 - greatest compression)",
 			    record__parse_comp_level),
 #endif
 	OPT_CALLBACK(0, "max-size", &record.output_max_size,
@@ -3610,6 +3637,17 @@ int cmd_record(int argc, const char **argv)
 	if (rec->opts.kcore || record__threads_enabled(rec))
 		rec->data.is_dir = true;
 
+	if (record__threads_enabled(rec)) {
+		if (rec->opts.affinity != PERF_AFFINITY_SYS) {
+			pr_err("--affinity option is mutually exclusive to parallel streaming mode.\n");
+			goto out_opts;
+		}
+		if (record__aio_enabled(rec)) {
+			pr_err("Asynchronous streaming mode (--aio) is mutually exclusive to parallel streaming mode.\n");
+			goto out_opts;
+		}
+	}
+
 	if (rec->opts.comp_level != 0) {
 		pr_debug("Compression enabled, disabling build id collection at the end of the session.\n");
 		rec->no_buildid = true;
@@ -3643,6 +3681,11 @@ int cmd_record(int argc, const char **argv)
 		}
 	}
 
+	if (rec->timestamp_filename && record__threads_enabled(rec)) {
+		rec->timestamp_filename = false;
+		pr_warning("WARNING: --timestamp-filename option is not available in parallel streaming mode.\n");
+	}
+
 	/*
 	 * Allow aliases to facilitate the lookup of symbols for address
 	 * filters. Refer to auxtrace_parse_filters().
-- 
2.19.0

[PATCH v5 12/20] perf report: output data file name in raw trace dump

[Alexey Bayduraev]

Print path and name of a data file into raw dump (-D)
<file_offset>@<path/file>. Print offset of PERF_RECORD_COMPRESSED
record instead of zero for decompressed records:
  0x2226a@perf.data [0x30]: event: 9
or
  0x15cc36@perf.data/data.7 [0x30]: event: 9

Acked-by: Namhyung Kim <namhyung@kernel.org>
Acked-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Alexey Bayduraev <alexey.v.bayduraev@linux.intel.com>
---
 tools/perf/builtin-inject.c      |  3 +-
 tools/perf/util/ordered-events.h |  1 +
 tools/perf/util/session.c        | 77 +++++++++++++++++++-------------
 tools/perf/util/session.h        |  1 +
 tools/perf/util/tool.h           |  3 +-
 5 files changed, 52 insertions(+), 33 deletions(-)

diff --git a/tools/perf/builtin-inject.c b/tools/perf/builtin-inject.c
index ddccc0eb7390..4f6d8e9fa4ca 100644
--- a/tools/perf/builtin-inject.c
+++ b/tools/perf/builtin-inject.c
@@ -106,7 +106,8 @@ static int perf_event__repipe_op2_synth(struct perf_session *session,
 
 static int perf_event__repipe_op4_synth(struct perf_session *session,
 					union perf_event *event,
-					u64 data __maybe_unused)
+					u64 data __maybe_unused,
+					const char *str __maybe_unused)
 {
 	return perf_event__repipe_synth(session->tool, event);
 }
diff --git a/tools/perf/util/ordered-events.h b/tools/perf/util/ordered-events.h
index 75345946c4b9..42c9764c6b5b 100644
--- a/tools/perf/util/ordered-events.h
+++ b/tools/perf/util/ordered-events.h
@@ -9,6 +9,7 @@ struct perf_sample;
 struct ordered_event {
 	u64			timestamp;
 	u64			file_offset;
+	const char		*file_path;
 	union perf_event	*event;
 	struct list_head	list;
 };
diff --git a/tools/perf/util/session.c b/tools/perf/util/session.c
index a12cf4f0e97a..573bdb2f446b 100644
--- a/tools/perf/util/session.c
+++ b/tools/perf/util/session.c
@@ -38,7 +38,8 @@
 
 #ifdef HAVE_ZSTD_SUPPORT
 static int perf_session__process_compressed_event(struct perf_session *session,
-						  union perf_event *event, u64 file_offset)
+						  union perf_event *event, u64 file_offset,
+						  const char *file_path)
 {
 	void *src;
 	size_t decomp_size, src_size;
@@ -60,6 +61,7 @@ static int perf_session__process_compressed_event(struct perf_session *session,
 	}
 
 	decomp->file_pos = file_offset;
+	decomp->file_path = file_path;
 	decomp->mmap_len = mmap_len;
 	decomp->head = 0;
 
@@ -100,7 +102,8 @@ static int perf_session__process_compressed_event(struct perf_session *session,
 static int perf_session__deliver_event(struct perf_session *session,
 				       union perf_event *event,
 				       struct perf_tool *tool,
-				       u64 file_offset);
+				       u64 file_offset,
+				       const char *file_path);
 
 static int perf_session__open(struct perf_session *session)
 {
@@ -182,7 +185,8 @@ static int ordered_events__deliver_event(struct ordered_events *oe,
 						    ordered_events);
 
 	return perf_session__deliver_event(session, event->event,
-					   session->tool, event->file_offset);
+					   session->tool, event->file_offset,
+					   event->file_path);
 }
 
 struct perf_session *perf_session__new(struct perf_data *data,
@@ -464,7 +468,8 @@ static int process_event_time_conv_stub(struct perf_session *perf_session __mayb
 
 static int perf_session__process_compressed_event_stub(struct perf_session *session __maybe_unused,
 						       union perf_event *event __maybe_unused,
-						       u64 file_offset __maybe_unused)
+						       u64 file_offset __maybe_unused,
+						       const char *file_path __maybe_unused)
 {
        dump_printf(": unhandled!\n");
        return 0;
@@ -1267,13 +1272,14 @@ static void sample_read__printf(struct perf_sample *sample, u64 read_format)
 }
 
 static void dump_event(struct evlist *evlist, union perf_event *event,
-		       u64 file_offset, struct perf_sample *sample)
+		       u64 file_offset, struct perf_sample *sample,
+		       const char *file_path)
 {
 	if (!dump_trace)
 		return;
 
-	printf("\n%#" PRIx64 " [%#x]: event: %d\n",
-	       file_offset, event->header.size, event->header.type);
+	printf("\n%#" PRIx64 "@%s [%#x]: event: %d\n",
+	       file_offset, file_path, event->header.size, event->header.type);
 
 	trace_event(event);
 	if (event->header.type == PERF_RECORD_SAMPLE && evlist->trace_event_sample_raw)
@@ -1476,12 +1482,13 @@ static int machines__deliver_event(struct machines *machines,
 				   struct evlist *evlist,
 				   union perf_event *event,
 				   struct perf_sample *sample,
-				   struct perf_tool *tool, u64 file_offset)
+				   struct perf_tool *tool, u64 file_offset,
+				   const char *file_path)
 {
 	struct evsel *evsel;
 	struct machine *machine;
 
-	dump_event(evlist, event, file_offset, sample);
+	dump_event(evlist, event, file_offset, sample, file_path);
 
 	evsel = evlist__id2evsel(evlist, sample->id);
 
@@ -1558,7 +1565,8 @@ static int machines__deliver_event(struct machines *machines,
 static int perf_session__deliver_event(struct perf_session *session,
 				       union perf_event *event,
 				       struct perf_tool *tool,
-				       u64 file_offset)
+				       u64 file_offset,
+				       const char *file_path)
 {
 	struct perf_sample sample;
 	int ret = evlist__parse_sample(session->evlist, event, &sample);
@@ -1575,7 +1583,7 @@ static int perf_session__deliver_event(struct perf_session *session,
 		return 0;
 
 	ret = machines__deliver_event(&session->machines, session->evlist,
-				      event, &sample, tool, file_offset);
+				      event, &sample, tool, file_offset, file_path);
 
 	if (dump_trace && sample.aux_sample.size)
 		auxtrace__dump_auxtrace_sample(session, &sample);
@@ -1585,7 +1593,8 @@ static int perf_session__deliver_event(struct perf_session *session,
 
 static s64 perf_session__process_user_event(struct perf_session *session,
 					    union perf_event *event,
-					    u64 file_offset)
+					    u64 file_offset,
+					    const char *file_path)
 {
 	struct ordered_events *oe = &session->ordered_events;
 	struct perf_tool *tool = session->tool;
@@ -1595,7 +1604,7 @@ static s64 perf_session__process_user_event(struct perf_session *session,
 
 	if (event->header.type != PERF_RECORD_COMPRESSED ||
 	    tool->compressed == perf_session__process_compressed_event_stub)
-		dump_event(session->evlist, event, file_offset, &sample);
+		dump_event(session->evlist, event, file_offset, &sample, file_path);
 
 	/* These events are processed right away */
 	switch (event->header.type) {
@@ -1654,9 +1663,9 @@ static s64 perf_session__process_user_event(struct perf_session *session,
 	case PERF_RECORD_HEADER_FEATURE:
 		return tool->feature(session, event);
 	case PERF_RECORD_COMPRESSED:
-		err = tool->compressed(session, event, file_offset);
+		err = tool->compressed(session, event, file_offset, file_path);
 		if (err)
-			dump_event(session->evlist, event, file_offset, &sample);
+			dump_event(session->evlist, event, file_offset, &sample, file_path);
 		return err;
 	default:
 		return -EINVAL;
@@ -1673,9 +1682,9 @@ int perf_session__deliver_synth_event(struct perf_session *session,
 	events_stats__inc(&evlist->stats, event->header.type);
 
 	if (event->header.type >= PERF_RECORD_USER_TYPE_START)
-		return perf_session__process_user_event(session, event, 0);
+		return perf_session__process_user_event(session, event, 0, NULL);
 
-	return machines__deliver_event(&session->machines, evlist, event, sample, tool, 0);
+	return machines__deliver_event(&session->machines, evlist, event, sample, tool, 0, NULL);
 }
 
 static void event_swap(union perf_event *event, bool sample_id_all)
@@ -1771,7 +1780,8 @@ int perf_session__peek_events(struct perf_session *session, u64 offset,
 }
 
 static s64 perf_session__process_event(struct perf_session *session,
-				       union perf_event *event, u64 file_offset)
+				       union perf_event *event, u64 file_offset,
+				       const char *file_path)
 {
 	struct evlist *evlist = session->evlist;
 	struct perf_tool *tool = session->tool;
@@ -1786,7 +1796,7 @@ static s64 perf_session__process_event(struct perf_session *session,
 	events_stats__inc(&evlist->stats, event->header.type);
 
 	if (event->header.type >= PERF_RECORD_USER_TYPE_START)
-		return perf_session__process_user_event(session, event, file_offset);
+		return perf_session__process_user_event(session, event, file_offset, file_path);
 
 	if (tool->ordered_events) {
 		u64 timestamp = -1ULL;
@@ -1800,7 +1810,7 @@ static s64 perf_session__process_event(struct perf_session *session,
 			return ret;
 	}
 
-	return perf_session__deliver_event(session, event, tool, file_offset);
+	return perf_session__deliver_event(session, event, tool, file_offset, file_path);
 }
 
 void perf_event_header__bswap(struct perf_event_header *hdr)
@@ -2020,7 +2030,8 @@ static int __perf_session__process_pipe_events(struct perf_session *session)
 		}
 	}
 
-	if ((skip = perf_session__process_event(session, event, head)) < 0) {
+	skip = perf_session__process_event(session, event, head, "pipe");
+	if (skip < 0) {
 		pr_err("%#" PRIx64 " [%#x]: failed to process type: %d\n",
 		       head, event->header.size, event->header.type);
 		err = -EINVAL;
@@ -2101,7 +2112,7 @@ fetch_decomp_event(u64 head, size_t mmap_size, char *buf, bool needs_swap)
 static int __perf_session__process_decomp_events(struct perf_session *session)
 {
 	s64 skip;
-	u64 size, file_pos = 0;
+	u64 size;
 	struct decomp *decomp = session->decomp_last;
 
 	if (!decomp)
@@ -2115,9 +2126,9 @@ static int __perf_session__process_decomp_events(struct perf_session *session)
 			break;
 
 		size = event->header.size;
-
-		if (size < sizeof(struct perf_event_header) ||
-		    (skip = perf_session__process_event(session, event, file_pos)) < 0) {
+		skip = perf_session__process_event(session, event, decomp->file_pos,
+						   decomp->file_path);
+		if (size < sizeof(struct perf_event_header) || skip < 0) {
 			pr_err("%#" PRIx64 " [%#x]: failed to process type: %d\n",
 				decomp->file_pos + decomp->head, event->header.size, event->header.type);
 			return -EINVAL;
@@ -2148,10 +2159,12 @@ struct reader;
 
 typedef s64 (*reader_cb_t)(struct perf_session *session,
 			   union perf_event *event,
-			   u64 file_offset);
+			   u64 file_offset,
+			   const char *file_path);
 
 struct reader {
 	int		 fd;
+	const char	 *path;
 	u64		 data_size;
 	u64		 data_offset;
 	reader_cb_t	 process;
@@ -2230,9 +2243,9 @@ reader__process_events(struct reader *rd, struct perf_session *session,
 	skip = -EINVAL;
 
 	if (size < sizeof(struct perf_event_header) ||
-	    (skip = rd->process(session, event, file_pos)) < 0) {
-		pr_err("%#" PRIx64 " [%#x]: failed to process type: %d [%s]\n",
-		       file_offset + head, event->header.size,
+	    (skip = rd->process(session, event, file_pos, rd->path)) < 0) {
+		pr_err("%#" PRIx64 " [%s] [%#x]: failed to process type: %d [%s]\n",
+		       file_offset + head, rd->path, event->header.size,
 		       event->header.type, strerror(-skip));
 		err = skip;
 		goto out;
@@ -2262,9 +2275,10 @@ reader__process_events(struct reader *rd, struct perf_session *session,
 
 static s64 process_simple(struct perf_session *session,
 			  union perf_event *event,
-			  u64 file_offset)
+			  u64 file_offset,
+			  const char *file_path)
 {
-	return perf_session__process_event(session, event, file_offset);
+	return perf_session__process_event(session, event, file_offset, file_path);
 }
 
 static int __perf_session__process_events(struct perf_session *session)
@@ -2274,6 +2288,7 @@ static int __perf_session__process_events(struct perf_session *session)
 		.data_size	= session->header.data_size,
 		.data_offset	= session->header.data_offset,
 		.process	= process_simple,
+		.path		= session->data->file.path,
 	};
 	struct ordered_events *oe = &session->ordered_events;
 	struct perf_tool *tool = session->tool;
diff --git a/tools/perf/util/session.h b/tools/perf/util/session.h
index e31ba4c92a6c..6895a22ab5b7 100644
--- a/tools/perf/util/session.h
+++ b/tools/perf/util/session.h
@@ -46,6 +46,7 @@ struct perf_session {
 struct decomp {
 	struct decomp *next;
 	u64 file_pos;
+	const char *file_path;
 	size_t mmap_len;
 	u64 head;
 	size_t size;
diff --git a/tools/perf/util/tool.h b/tools/perf/util/tool.h
index bbbc0dcd461f..c966531d3eca 100644
--- a/tools/perf/util/tool.h
+++ b/tools/perf/util/tool.h
@@ -28,7 +28,8 @@ typedef int (*event_attr_op)(struct perf_tool *tool,
 
 typedef int (*event_op2)(struct perf_session *session, union perf_event *event);
 typedef s64 (*event_op3)(struct perf_session *session, union perf_event *event);
-typedef int (*event_op4)(struct perf_session *session, union perf_event *event, u64 data);
+typedef int (*event_op4)(struct perf_session *session, union perf_event *event, u64 data,
+			 const char *str);
 
 typedef int (*event_oe)(struct perf_tool *tool, union perf_event *event,
 			struct ordered_events *oe);
-- 
2.19.0

[PATCH v5 13/20] perf session: move reader structure to the top

[Alexey Bayduraev]

Moving reader structure to the top of the file. This is necessary
for the further use of this structure in the decompressor.

Signed-off-by: Alexey Bayduraev <alexey.v.bayduraev@linux.intel.com>
---
 tools/perf/util/session.c | 54 +++++++++++++++++++--------------------
 1 file changed, 27 insertions(+), 27 deletions(-)

diff --git a/tools/perf/util/session.c b/tools/perf/util/session.c
index 573bdb2f446b..a4d225e0569c 100644
--- a/tools/perf/util/session.c
+++ b/tools/perf/util/session.c
@@ -36,6 +36,33 @@
 #include "units.h"
 #include <internal/lib.h>
 
+/*
+ * On 64bit we can mmap the data file in one go. No need for tiny mmap
+ * slices. On 32bit we use 32MB.
+ */
+#if BITS_PER_LONG == 64
+#define MMAP_SIZE ULLONG_MAX
+#define NUM_MMAPS 1
+#else
+#define MMAP_SIZE (32 * 1024 * 1024ULL)
+#define NUM_MMAPS 128
+#endif
+
+struct reader;
+
+typedef s64 (*reader_cb_t)(struct perf_session *session,
+			   union perf_event *event,
+			   u64 file_offset,
+			   const char *file_path);
+
+struct reader {
+	int		 fd;
+	const char	 *path;
+	u64		 data_size;
+	u64		 data_offset;
+	reader_cb_t	 process;
+};
+
 #ifdef HAVE_ZSTD_SUPPORT
 static int perf_session__process_compressed_event(struct perf_session *session,
 						  union perf_event *event, u64 file_offset,
@@ -2143,33 +2170,6 @@ static int __perf_session__process_decomp_events(struct perf_session *session)
 	return 0;
 }
 
-/*
- * On 64bit we can mmap the data file in one go. No need for tiny mmap
- * slices. On 32bit we use 32MB.
- */
-#if BITS_PER_LONG == 64
-#define MMAP_SIZE ULLONG_MAX
-#define NUM_MMAPS 1
-#else
-#define MMAP_SIZE (32 * 1024 * 1024ULL)
-#define NUM_MMAPS 128
-#endif
-
-struct reader;
-
-typedef s64 (*reader_cb_t)(struct perf_session *session,
-			   union perf_event *event,
-			   u64 file_offset,
-			   const char *file_path);
-
-struct reader {
-	int		 fd;
-	const char	 *path;
-	u64		 data_size;
-	u64		 data_offset;
-	reader_cb_t	 process;
-};
-
 static int
 reader__process_events(struct reader *rd, struct perf_session *session,
 		       struct ui_progress *prog)
-- 
2.19.0

[PATCH v5 14/20] perf session: introduce reader_state in reader object

[Alexey Bayduraev]

We need all the state info about reader in separate object to load data
from multiple files, so we can keep multiple readers at the same time.
Adding struct reader_state and adding all items that need to be kept.

Design and implementation are 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-jolsa@kernel.org/

Suggested-by: Jiri Olsa <jolsa@kernel.org>
Signed-off-by: Alexey Bayduraev <alexey.v.bayduraev@linux.intel.com>
---
 tools/perf/util/session.c | 74 +++++++++++++++++++++++----------------
 1 file changed, 43 insertions(+), 31 deletions(-)

diff --git a/tools/perf/util/session.c b/tools/perf/util/session.c
index a4d225e0569c..054f4d04eea9 100644
--- a/tools/perf/util/session.c
+++ b/tools/perf/util/session.c
@@ -55,12 +55,24 @@ typedef s64 (*reader_cb_t)(struct perf_session *session,
 			   u64 file_offset,
 			   const char *file_path);
 
+struct reader_state {
+	char	*mmaps[NUM_MMAPS];
+	size_t	 mmap_size;
+	int	 mmap_idx;
+	char	*mmap_cur;
+	u64	 file_pos;
+	u64	 file_offset;
+	u64	 data_size;
+	u64	 head;
+};
+
 struct reader {
 	int		 fd;
 	const char	 *path;
 	u64		 data_size;
 	u64		 data_offset;
 	reader_cb_t	 process;
+	struct reader_state state;
 };
 
 #ifdef HAVE_ZSTD_SUPPORT
@@ -2174,29 +2186,28 @@ static int
 reader__process_events(struct reader *rd, struct perf_session *session,
 		       struct ui_progress *prog)
 {
-	u64 data_size = rd->data_size;
-	u64 head, page_offset, file_offset, file_pos, size;
-	int err = 0, mmap_prot, mmap_flags, map_idx = 0;
-	size_t	mmap_size;
-	char *buf, *mmaps[NUM_MMAPS];
+	struct reader_state *st = &rd->state;
+	u64 page_offset, size;
+	int err = 0, mmap_prot, mmap_flags;
+	char *buf, **mmaps = st->mmaps;
 	union perf_event *event;
 	s64 skip;
 
 	page_offset = page_size * (rd->data_offset / page_size);
-	file_offset = page_offset;
-	head = rd->data_offset - page_offset;
+	st->file_offset = page_offset;
+	st->head = rd->data_offset - page_offset;
 
-	ui_progress__init_size(prog, data_size, "Processing events...");
+	ui_progress__init_size(prog, rd->data_size, "Processing events...");
 
-	data_size += rd->data_offset;
+	st->data_size = rd->data_size + rd->data_offset;
 
-	mmap_size = MMAP_SIZE;
-	if (mmap_size > data_size) {
-		mmap_size = data_size;
+	st->mmap_size = MMAP_SIZE;
+	if (st->mmap_size > st->data_size) {
+		st->mmap_size = st->data_size;
 		session->one_mmap = true;
 	}
 
-	memset(mmaps, 0, sizeof(mmaps));
+	memset(mmaps, 0, sizeof(st->mmaps));
 
 	mmap_prot  = PROT_READ;
 	mmap_flags = MAP_SHARED;
@@ -2206,35 +2217,36 @@ reader__process_events(struct reader *rd, struct perf_session *session,
 		mmap_flags = MAP_PRIVATE;
 	}
 remap:
-	buf = mmap(NULL, mmap_size, mmap_prot, mmap_flags, rd->fd,
-		   file_offset);
+	buf = mmap(NULL, st->mmap_size, mmap_prot, mmap_flags, rd->fd,
+		   st->file_offset);
 	if (buf == MAP_FAILED) {
 		pr_err("failed to mmap file\n");
 		err = -errno;
 		goto out;
 	}
-	mmaps[map_idx] = buf;
-	map_idx = (map_idx + 1) & (ARRAY_SIZE(mmaps) - 1);
-	file_pos = file_offset + head;
+	mmaps[st->mmap_idx] = st->mmap_cur = buf;
+	st->mmap_idx = (st->mmap_idx + 1) & (ARRAY_SIZE(st->mmaps) - 1);
+	st->file_pos = st->file_offset + st->head;
 	if (session->one_mmap) {
 		session->one_mmap_addr = buf;
-		session->one_mmap_offset = file_offset;
+		session->one_mmap_offset = st->file_offset;
 	}
 
 more:
-	event = fetch_mmaped_event(head, mmap_size, buf, session->header.needs_swap);
+	event = fetch_mmaped_event(st->head, st->mmap_size, st->mmap_cur,
+				   session->header.needs_swap);
 	if (IS_ERR(event))
 		return PTR_ERR(event);
 
 	if (!event) {
-		if (mmaps[map_idx]) {
-			munmap(mmaps[map_idx], mmap_size);
-			mmaps[map_idx] = NULL;
+		if (mmaps[st->mmap_idx]) {
+			munmap(mmaps[st->mmap_idx], st->mmap_size);
+			mmaps[st->mmap_idx] = NULL;
 		}
 
-		page_offset = page_size * (head / page_size);
-		file_offset += page_offset;
-		head -= page_offset;
+		page_offset = page_size * (st->head / page_size);
+		st->file_offset += page_offset;
+		st->head -= page_offset;
 		goto remap;
 	}
 
@@ -2243,9 +2255,9 @@ reader__process_events(struct reader *rd, struct perf_session *session,
 	skip = -EINVAL;
 
 	if (size < sizeof(struct perf_event_header) ||
-	    (skip = rd->process(session, event, file_pos, rd->path)) < 0) {
+	    (skip = rd->process(session, event, st->file_pos, rd->path)) < 0) {
 		pr_err("%#" PRIx64 " [%s] [%#x]: failed to process type: %d [%s]\n",
-		       file_offset + head, rd->path, event->header.size,
+		       st->file_offset + st->head, rd->path, event->header.size,
 		       event->header.type, strerror(-skip));
 		err = skip;
 		goto out;
@@ -2254,8 +2266,8 @@ reader__process_events(struct reader *rd, struct perf_session *session,
 	if (skip)
 		size += skip;
 
-	head += size;
-	file_pos += size;
+	st->head += size;
+	st->file_pos += size;
 
 	err = __perf_session__process_decomp_events(session);
 	if (err)
@@ -2266,7 +2278,7 @@ reader__process_events(struct reader *rd, struct perf_session *session,
 	if (session_done())
 		goto out;
 
-	if (file_pos < data_size)
+	if (st->file_pos < st->data_size)
 		goto more;
 
 out:
-- 
2.19.0

[PATCH v5 15/20] perf session: introduce reader objects in session object

[Alexey Bayduraev]

Allow to allocate multiple reader objects, so we could load multiple
data files located in data directory at the same time.

Design and implementation are 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-jolsa@kernel.org/

Suggested-by: Jiri Olsa <jolsa@kernel.org>
Signed-off-by: Alexey Bayduraev <alexey.v.bayduraev@linux.intel.com>
---
 tools/perf/util/session.c | 31 ++++++++++++++++++++-----------
 tools/perf/util/session.h |  3 +++
 2 files changed, 23 insertions(+), 11 deletions(-)

diff --git a/tools/perf/util/session.c b/tools/perf/util/session.c
index 054f4d04eea9..65ce798eb27d 100644
--- a/tools/perf/util/session.c
+++ b/tools/perf/util/session.c
@@ -341,6 +341,10 @@ void perf_session__delete(struct perf_session *session)
 	auxtrace_index__free(&session->auxtrace_index);
 	perf_session__destroy_kernel_maps(session);
 	perf_session__delete_threads(session);
+	if (session->readers) {
+		zfree(&session->readers);
+		session->nr_readers = 0;
+	}
 	perf_session__release_decomp_events(session);
 	perf_env__exit(&session->header.env);
 	machines__exit(&session->machines);
@@ -2295,13 +2299,7 @@ static s64 process_simple(struct perf_session *session,
 
 static int __perf_session__process_events(struct perf_session *session)
 {
-	struct reader rd = {
-		.fd		= perf_data__fd(session->data),
-		.data_size	= session->header.data_size,
-		.data_offset	= session->header.data_offset,
-		.process	= process_simple,
-		.path		= session->data->file.path,
-	};
+	struct reader *rd;
 	struct ordered_events *oe = &session->ordered_events;
 	struct perf_tool *tool = session->tool;
 	struct ui_progress prog;
@@ -2309,12 +2307,23 @@ static int __perf_session__process_events(struct perf_session *session)
 
 	perf_tool__fill_defaults(tool);
 
-	if (rd.data_size == 0)
-		return -1;
+	rd = session->readers = zalloc(sizeof(struct reader));
+	if (!rd)
+		return -ENOMEM;
+
+	session->nr_readers = 1;
+
+	*rd = (struct reader) {
+		.fd		= perf_data__fd(session->data),
+		.data_size	= session->header.data_size,
+		.data_offset	= session->header.data_offset,
+		.process	= process_simple,
+		.path		= session->data->file.path,
+	};
 
-	ui_progress__init_size(&prog, rd.data_size, "Processing events...");
+	ui_progress__init_size(&prog, rd->data_size, "Processing events...");
 
-	err = reader__process_events(&rd, session, &prog);
+	err = reader__process_events(rd, session, &prog);
 	if (err)
 		goto out_err;
 	/* do the final flush for ordered samples */
diff --git a/tools/perf/util/session.h b/tools/perf/util/session.h
index 6895a22ab5b7..2815d00b5467 100644
--- a/tools/perf/util/session.h
+++ b/tools/perf/util/session.h
@@ -19,6 +19,7 @@ struct thread;
 
 struct auxtrace;
 struct itrace_synth_opts;
+struct reader;
 
 struct perf_session {
 	struct perf_header	header;
@@ -41,6 +42,8 @@ struct perf_session {
 	struct zstd_data	zstd_data;
 	struct decomp		*decomp;
 	struct decomp		*decomp_last;
+	struct reader		*readers;
+	int			nr_readers;
 };
 
 struct decomp {
-- 
2.19.0

[PATCH v5 16/20] perf session: introduce decompressor into trace reader object

[Alexey Bayduraev]

Introduce decompressor into trace reader object so that decompression
could be executed on per data file basis separately for every data
file located in data directory.

Signed-off-by: Alexey Bayduraev <alexey.v.bayduraev@linux.intel.com>
---
 tools/perf/util/session.c | 51 +++++++++++++++++++++++++++++----------
 tools/perf/util/session.h |  1 +
 2 files changed, 39 insertions(+), 13 deletions(-)

diff --git a/tools/perf/util/session.c b/tools/perf/util/session.c
index 65ce798eb27d..81214052b2e3 100644
--- a/tools/perf/util/session.c
+++ b/tools/perf/util/session.c
@@ -72,6 +72,9 @@ struct reader {
 	u64		 data_size;
 	u64		 data_offset;
 	reader_cb_t	 process;
+	struct zstd_data zstd_data;
+	struct decomp	 *decomp;
+	struct decomp	 *decomp_last;
 	struct reader_state state;
 };
 
@@ -84,7 +87,10 @@ static int perf_session__process_compressed_event(struct perf_session *session,
 	size_t decomp_size, src_size;
 	u64 decomp_last_rem = 0;
 	size_t mmap_len, decomp_len = session->header.env.comp_mmap_len;
-	struct decomp *decomp, *decomp_last = session->decomp_last;
+	struct decomp *decomp, *decomp_last = session->active_reader ?
+		session->active_reader->decomp_last : session->decomp_last;
+	struct zstd_data *zstd_data = session->active_reader ?
+		&session->active_reader->zstd_data : &session->zstd_data;
 
 	if (decomp_last) {
 		decomp_last_rem = decomp_last->size - decomp_last->head;
@@ -112,7 +118,7 @@ static int perf_session__process_compressed_event(struct perf_session *session,
 	src = (void *)event + sizeof(struct perf_record_compressed);
 	src_size = event->pack.header.size - sizeof(struct perf_record_compressed);
 
-	decomp_size = zstd_decompress_stream(&(session->zstd_data), src, src_size,
+	decomp_size = zstd_decompress_stream(zstd_data, src, src_size,
 				&(decomp->data[decomp_last_rem]), decomp_len - decomp_last_rem);
 	if (!decomp_size) {
 		munmap(decomp, mmap_len);
@@ -122,12 +128,22 @@ static int perf_session__process_compressed_event(struct perf_session *session,
 
 	decomp->size += decomp_size;
 
-	if (session->decomp == NULL) {
-		session->decomp = decomp;
-		session->decomp_last = decomp;
+	if (session->active_reader) {
+		if (session->active_reader->decomp == NULL) {
+			session->active_reader->decomp = decomp;
+			session->active_reader->decomp_last = decomp;
+		} else {
+			session->active_reader->decomp_last->next = decomp;
+			session->active_reader->decomp_last = decomp;
+		}
 	} else {
-		session->decomp_last->next = decomp;
-		session->decomp_last = decomp;
+		if (session->decomp == NULL) {
+			session->decomp = decomp;
+			session->decomp_last = decomp;
+		} else {
+			session->decomp_last->next = decomp;
+			session->decomp_last = decomp;
+		}
 	}
 
 	pr_debug("decomp (B): %zd to %zd\n", src_size, decomp_size);
@@ -318,11 +334,10 @@ static void perf_session__delete_threads(struct perf_session *session)
 	machine__delete_threads(&session->machines.host);
 }
 
-static void perf_session__release_decomp_events(struct perf_session *session)
+static void perf_decomp__release_events(struct decomp *next)
 {
-	struct decomp *next, *decomp;
+	struct decomp *decomp;
 	size_t mmap_len;
-	next = session->decomp;
 	do {
 		decomp = next;
 		if (decomp == NULL)
@@ -335,6 +350,8 @@ static void perf_session__release_decomp_events(struct perf_session *session)
 
 void perf_session__delete(struct perf_session *session)
 {
+	int r;
+
 	if (session == NULL)
 		return;
 	auxtrace__free(session);
@@ -342,10 +359,12 @@ void perf_session__delete(struct perf_session *session)
 	perf_session__destroy_kernel_maps(session);
 	perf_session__delete_threads(session);
 	if (session->readers) {
+		for (r = 0; r < session->nr_readers; r++)
+			perf_decomp__release_events(session->readers[r].decomp);
 		zfree(&session->readers);
 		session->nr_readers = 0;
 	}
-	perf_session__release_decomp_events(session);
+	perf_decomp__release_events(session->decomp);
 	perf_env__exit(&session->header.env);
 	machines__exit(&session->machines);
 	if (session->data)
@@ -2156,7 +2175,8 @@ static int __perf_session__process_decomp_events(struct perf_session *session)
 {
 	s64 skip;
 	u64 size;
-	struct decomp *decomp = session->decomp_last;
+	struct decomp *decomp = session->active_reader ?
+		session->active_reader->decomp_last : session->decomp_last;
 
 	if (!decomp)
 		return 0;
@@ -2213,6 +2233,9 @@ reader__process_events(struct reader *rd, struct perf_session *session,
 
 	memset(mmaps, 0, sizeof(st->mmaps));
 
+	if (zstd_init(&rd->zstd_data, 0))
+		return -1;
+
 	mmap_prot  = PROT_READ;
 	mmap_flags = MAP_SHARED;
 
@@ -2254,12 +2277,13 @@ reader__process_events(struct reader *rd, struct perf_session *session,
 		goto remap;
 	}
 
+	session->active_reader = rd;
 	size = event->header.size;
 
 	skip = -EINVAL;
 
 	if (size < sizeof(struct perf_event_header) ||
-	    (skip = rd->process(session, event, st->file_pos, rd->path)) < 0) {
+	    (skip = perf_session__process_event(session, event, st->file_pos, rd->path)) < 0) {
 		pr_err("%#" PRIx64 " [%s] [%#x]: failed to process type: %d [%s]\n",
 		       st->file_offset + st->head, rd->path, event->header.size,
 		       event->header.type, strerror(-skip));
@@ -2286,6 +2310,7 @@ reader__process_events(struct reader *rd, struct perf_session *session,
 		goto more;
 
 out:
+	session->active_reader = NULL;
 	return err;
 }
 
diff --git a/tools/perf/util/session.h b/tools/perf/util/session.h
index 2815d00b5467..e0a8712f8770 100644
--- a/tools/perf/util/session.h
+++ b/tools/perf/util/session.h
@@ -44,6 +44,7 @@ struct perf_session {
 	struct decomp		*decomp_last;
 	struct reader		*readers;
 	int			nr_readers;
+	struct reader		*active_reader;
 };
 
 struct decomp {
-- 
2.19.0

[PATCH v5 17/20] perf session: move init into reader__init function

[Alexey Bayduraev]

Separating initialization code into reader__init function.

Design and implementation are 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-jolsa@kernel.org/

Suggested-by: Jiri Olsa <jolsa@kernel.org>
Signed-off-by: Alexey Bayduraev <alexey.v.bayduraev@linux.intel.com>
---
 tools/perf/util/session.c | 34 ++++++++++++++++++++++++----------
 1 file changed, 24 insertions(+), 10 deletions(-)

diff --git a/tools/perf/util/session.c b/tools/perf/util/session.c
index 81214052b2e3..08447376fa92 100644
--- a/tools/perf/util/session.c
+++ b/tools/perf/util/session.c
@@ -2207,28 +2207,25 @@ static int __perf_session__process_decomp_events(struct perf_session *session)
 }
 
 static int
-reader__process_events(struct reader *rd, struct perf_session *session,
-		       struct ui_progress *prog)
+reader__init(struct reader *rd, bool *one_mmap)
 {
 	struct reader_state *st = &rd->state;
-	u64 page_offset, size;
-	int err = 0, mmap_prot, mmap_flags;
-	char *buf, **mmaps = st->mmaps;
-	union perf_event *event;
-	s64 skip;
+	u64 page_offset;
+	char **mmaps = st->mmaps;
+
+	pr_debug("reader processing %s\n", rd->path);
 
 	page_offset = page_size * (rd->data_offset / page_size);
 	st->file_offset = page_offset;
 	st->head = rd->data_offset - page_offset;
 
-	ui_progress__init_size(prog, rd->data_size, "Processing events...");
-
 	st->data_size = rd->data_size + rd->data_offset;
 
 	st->mmap_size = MMAP_SIZE;
 	if (st->mmap_size > st->data_size) {
 		st->mmap_size = st->data_size;
-		session->one_mmap = true;
+		if (one_mmap)
+			*one_mmap = true;
 	}
 
 	memset(mmaps, 0, sizeof(st->mmaps));
@@ -2236,6 +2233,20 @@ reader__process_events(struct reader *rd, struct perf_session *session,
 	if (zstd_init(&rd->zstd_data, 0))
 		return -1;
 
+	return 0;
+}
+
+static int
+reader__process_events(struct reader *rd, struct perf_session *session,
+		       struct ui_progress *prog)
+{
+	struct reader_state *st = &rd->state;
+	u64 page_offset, size;
+	int err = 0, mmap_prot, mmap_flags;
+	char *buf, **mmaps = st->mmaps;
+	union perf_event *event;
+	s64 skip;
+
 	mmap_prot  = PROT_READ;
 	mmap_flags = MAP_SHARED;
 
@@ -2348,6 +2359,9 @@ static int __perf_session__process_events(struct perf_session *session)
 
 	ui_progress__init_size(&prog, rd->data_size, "Processing events...");
 
+	err = reader__init(rd, &session->one_mmap);
+	if (err)
+		goto out_err;
 	err = reader__process_events(rd, session, &prog);
 	if (err)
 		goto out_err;
-- 
2.19.0

[PATCH v5 18/20] perf session: move map/unmap into reader__mmap function

[Alexey Bayduraev]

Moving mapping and unmapping code into reader__mmap, so the mmap
code is located together. Moving head/file_offset computation into
reader__mmap function, so all the offset computation is located
together and in one place only.

Design and implementation are 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-jolsa@kernel.org/

Suggested-by: Jiri Olsa <jolsa@kernel.org>
Signed-off-by: Alexey Bayduraev <alexey.v.bayduraev@linux.intel.com>
---
 tools/perf/util/session.c | 60 +++++++++++++++++++++++----------------
 1 file changed, 35 insertions(+), 25 deletions(-)

diff --git a/tools/perf/util/session.c b/tools/perf/util/session.c
index 08447376fa92..5bb3babeeb94 100644
--- a/tools/perf/util/session.c
+++ b/tools/perf/util/session.c
@@ -2210,14 +2210,11 @@ static int
 reader__init(struct reader *rd, bool *one_mmap)
 {
 	struct reader_state *st = &rd->state;
-	u64 page_offset;
 	char **mmaps = st->mmaps;
 
 	pr_debug("reader processing %s\n", rd->path);
 
-	page_offset = page_size * (rd->data_offset / page_size);
-	st->file_offset = page_offset;
-	st->head = rd->data_offset - page_offset;
+	st->head = rd->data_offset;
 
 	st->data_size = rd->data_size + rd->data_offset;
 
@@ -2237,15 +2234,12 @@ reader__init(struct reader *rd, bool *one_mmap)
 }
 
 static int
-reader__process_events(struct reader *rd, struct perf_session *session,
-		       struct ui_progress *prog)
+reader__mmap(struct reader *rd, struct perf_session *session)
 {
 	struct reader_state *st = &rd->state;
-	u64 page_offset, size;
-	int err = 0, mmap_prot, mmap_flags;
+	int mmap_prot, mmap_flags;
 	char *buf, **mmaps = st->mmaps;
-	union perf_event *event;
-	s64 skip;
+	u64 page_offset;
 
 	mmap_prot  = PROT_READ;
 	mmap_flags = MAP_SHARED;
@@ -2254,20 +2248,45 @@ reader__process_events(struct reader *rd, struct perf_session *session,
 		mmap_prot  |= PROT_WRITE;
 		mmap_flags = MAP_PRIVATE;
 	}
-remap:
+
+	if (mmaps[st->mmap_idx]) {
+		munmap(mmaps[st->mmap_idx], st->mmap_size);
+		mmaps[st->mmap_idx] = NULL;
+	}
+
+	page_offset = page_size * (st->head / page_size);
+	st->file_offset += page_offset;
+	st->head -= page_offset;
+
 	buf = mmap(NULL, st->mmap_size, mmap_prot, mmap_flags, rd->fd,
 		   st->file_offset);
 	if (buf == MAP_FAILED) {
 		pr_err("failed to mmap file\n");
-		err = -errno;
-		goto out;
+		return -errno;
 	}
 	mmaps[st->mmap_idx] = st->mmap_cur = buf;
 	st->mmap_idx = (st->mmap_idx + 1) & (ARRAY_SIZE(st->mmaps) - 1);
 	st->file_pos = st->file_offset + st->head;
+	return 0;
+}
+
+static int
+reader__process_events(struct reader *rd, struct perf_session *session,
+		       struct ui_progress *prog)
+{
+	struct reader_state *st = &rd->state;
+	u64 size;
+	int err = 0;
+	union perf_event *event;
+	s64 skip;
+
+remap:
+	err = reader__mmap(rd, session);
+	if (err)
+		goto out;
 	if (session->one_mmap) {
-		session->one_mmap_addr = buf;
-		session->one_mmap_offset = st->file_offset;
+		session->one_mmap_addr   = rd->state.mmap_cur;
+		session->one_mmap_offset = rd->state.file_offset;
 	}
 
 more:
@@ -2276,17 +2295,8 @@ reader__process_events(struct reader *rd, struct perf_session *session,
 	if (IS_ERR(event))
 		return PTR_ERR(event);
 
-	if (!event) {
-		if (mmaps[st->mmap_idx]) {
-			munmap(mmaps[st->mmap_idx], st->mmap_size);
-			mmaps[st->mmap_idx] = NULL;
-		}
-
-		page_offset = page_size * (st->head / page_size);
-		st->file_offset += page_offset;
-		st->head -= page_offset;
+	if (!event)
 		goto remap;
-	}
 
 	session->active_reader = rd;
 	size = event->header.size;
-- 
2.19.0

[PATCH v5 19/20] perf session: load single file for analysis

[Alexey Bayduraev]

Adding eof flag to reader state and moving the check to reader__mmap.
Separating reading code of single event into reader__read_event function.
Adding basic reader return codes to simplify the code and introducing
reader remmap/read_event loop based on them.

Design and implementation are 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-jolsa@kernel.org/

Suggested-by: Jiri Olsa <jolsa@kernel.org>
Signed-off-by: Alexey Bayduraev <alexey.v.bayduraev@linux.intel.com>
---
 tools/perf/util/session.c | 71 ++++++++++++++++++++++++---------------
 1 file changed, 44 insertions(+), 27 deletions(-)

diff --git a/tools/perf/util/session.c b/tools/perf/util/session.c
index 5bb3babeeb94..47a414016510 100644
--- a/tools/perf/util/session.c
+++ b/tools/perf/util/session.c
@@ -64,6 +64,12 @@ struct reader_state {
 	u64	 file_offset;
 	u64	 data_size;
 	u64	 head;
+	bool	 eof;
+};
+
+enum {
+	READER_EOF	=  0,
+	READER_OK	=  1,
 };
 
 struct reader {
@@ -2241,6 +2247,11 @@ reader__mmap(struct reader *rd, struct perf_session *session)
 	char *buf, **mmaps = st->mmaps;
 	u64 page_offset;
 
+	if (st->file_pos >= st->data_size) {
+		st->eof = true;
+		return READER_EOF;
+	}
+
 	mmap_prot  = PROT_READ;
 	mmap_flags = MAP_SHARED;
 
@@ -2267,36 +2278,26 @@ reader__mmap(struct reader *rd, struct perf_session *session)
 	mmaps[st->mmap_idx] = st->mmap_cur = buf;
 	st->mmap_idx = (st->mmap_idx + 1) & (ARRAY_SIZE(st->mmaps) - 1);
 	st->file_pos = st->file_offset + st->head;
-	return 0;
+	return READER_OK;
 }
 
 static int
-reader__process_events(struct reader *rd, struct perf_session *session,
-		       struct ui_progress *prog)
+reader__read_event(struct reader *rd, struct perf_session *session,
+		   struct ui_progress *prog)
 {
 	struct reader_state *st = &rd->state;
-	u64 size;
-	int err = 0;
+	int err = READER_OK;
 	union perf_event *event;
+	u64 size;
 	s64 skip;
 
-remap:
-	err = reader__mmap(rd, session);
-	if (err)
-		goto out;
-	if (session->one_mmap) {
-		session->one_mmap_addr   = rd->state.mmap_cur;
-		session->one_mmap_offset = rd->state.file_offset;
-	}
-
-more:
 	event = fetch_mmaped_event(st->head, st->mmap_size, st->mmap_cur,
 				   session->header.needs_swap);
 	if (IS_ERR(event))
 		return PTR_ERR(event);
 
 	if (!event)
-		goto remap;
+		return READER_EOF;
 
 	session->active_reader = rd;
 	size = event->header.size;
@@ -2318,18 +2319,12 @@ reader__process_events(struct reader *rd, struct perf_session *session,
 	st->head += size;
 	st->file_pos += size;
 
-	err = __perf_session__process_decomp_events(session);
-	if (err)
-		goto out;
+	skip = __perf_session__process_decomp_events(session);
+	if (skip)
+		err = skip;
 
 	ui_progress__update(prog, size);
 
-	if (session_done())
-		goto out;
-
-	if (st->file_pos < st->data_size)
-		goto more;
-
 out:
 	session->active_reader = NULL;
 	return err;
@@ -2372,9 +2367,31 @@ static int __perf_session__process_events(struct perf_session *session)
 	err = reader__init(rd, &session->one_mmap);
 	if (err)
 		goto out_err;
-	err = reader__process_events(rd, session, &prog);
-	if (err)
+	err = reader__mmap(rd, session);
+	if (err != READER_OK) {
+		if (err == READER_EOF)
+			err = -EINVAL;
 		goto out_err;
+	}
+	if (session->one_mmap) {
+		session->one_mmap_addr   = rd->state.mmap_cur;
+		session->one_mmap_offset = rd->state.file_offset;
+	}
+
+	while (true) {
+		if (session_done())
+			break;
+
+		err = reader__read_event(rd, session, &prog);
+		if (err < 0)
+			break;
+		if (err == READER_EOF) {
+			err = reader__mmap(rd, session);
+			if (err <= 0)
+				break;
+		}
+	}
+
 	/* do the final flush for ordered samples */
 	err = ordered_events__flush(oe, OE_FLUSH__FINAL);
 	if (err)
-- 
2.19.0

[PATCH v5 20/20] perf session: load data directory files for analysis

[Alexey Bayduraev]

Load data directory files and provide basic raw dump and aggregated
analysis support of data directories in report mode, still with no
memory consumption optimizations.

Design and implementation are 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-jolsa@kernel.org/

Suggested-by: Jiri Olsa <jolsa@kernel.org>
Signed-off-by: Alexey Bayduraev <alexey.v.bayduraev@linux.intel.com>
---
 tools/perf/util/session.c | 127 ++++++++++++++++++++++++++++++++++++++
 1 file changed, 127 insertions(+)

diff --git a/tools/perf/util/session.c b/tools/perf/util/session.c
index 47a414016510..1a741e6fc35b 100644
--- a/tools/perf/util/session.c
+++ b/tools/perf/util/session.c
@@ -65,6 +65,7 @@ struct reader_state {
 	u64	 data_size;
 	u64	 head;
 	bool	 eof;
+	u64	 size;
 };
 
 enum {
@@ -2316,6 +2317,7 @@ reader__read_event(struct reader *rd, struct perf_session *session,
 	if (skip)
 		size += skip;
 
+	st->size += size;
 	st->head += size;
 	st->file_pos += size;
 
@@ -2414,6 +2416,128 @@ static int __perf_session__process_events(struct perf_session *session)
 	return err;
 }
 
+/*
+ * This function reads, merge and process directory data.
+ * It assumens the version 1 of directory data, where each
+ * data file holds per-cpu data, already sorted by kernel.
+ */
+static int __perf_session__process_dir_events(struct perf_session *session)
+{
+	struct perf_data *data = session->data;
+	struct perf_tool *tool = session->tool;
+	int i, ret = 0, readers = 1;
+	struct ui_progress prog;
+	u64 total_size = perf_data__size(session->data);
+	struct reader *rd;
+
+	perf_tool__fill_defaults(tool);
+
+	ui_progress__init_size(&prog, total_size, "Sorting events...");
+
+	for (i = 0; i < data->dir.nr; i++) {
+		if (data->dir.files[i].size)
+			readers++;
+	}
+
+	rd = session->readers = zalloc(readers * sizeof(struct reader));
+	if (!rd)
+		return -ENOMEM;
+	session->nr_readers = readers;
+	readers = 0;
+
+	rd[readers] = (struct reader) {
+		.fd		= perf_data__fd(session->data),
+		.path		= session->data->file.path,
+		.data_size	= session->header.data_size,
+		.data_offset	= session->header.data_offset,
+	};
+	ret = reader__init(&rd[readers], NULL);
+	if (ret)
+		goto out_err;
+	ret = reader__mmap(&rd[readers], session);
+	if (ret != READER_OK) {
+		if (ret == READER_EOF)
+			ret = -EINVAL;
+		goto out_err;
+	}
+	readers++;
+
+	for (i = 0; i < data->dir.nr; i++) {
+		if (data->dir.files[i].size) {
+			rd[readers] = (struct reader) {
+				.fd		= data->dir.files[i].fd,
+				.path		= data->dir.files[i].path,
+				.data_size	= data->dir.files[i].size,
+				.data_offset	= 0,
+			};
+			ret = reader__init(&rd[readers], NULL);
+			if (ret)
+				goto out_err;
+			ret = reader__mmap(&rd[readers], session);
+			if (ret != READER_OK) {
+				if (ret == READER_EOF)
+					ret = -EINVAL;
+				goto out_err;
+			}
+			readers++;
+		}
+	}
+
+	i = 0;
+
+	while ((ret >= 0) && readers) {
+		if (session_done())
+			return 0;
+
+		if (rd[i].state.eof) {
+			i = (i + 1) % session->nr_readers;
+			continue;
+		}
+
+		ret = reader__read_event(&rd[i], session, &prog);
+		if (ret < 0)
+			break;
+		if (ret == READER_EOF) {
+			ret = reader__mmap(&rd[i], session);
+			if (ret < 0)
+				goto out_err;
+			if (ret == READER_EOF)
+				readers--;
+		}
+
+		/*
+		 * Processing 10MBs of data from each reader in sequence,
+		 * because that's the way the ordered events sorting works
+		 * most efficiently.
+		 */
+		if (rd[i].state.size >= 10*1024*1024) {
+			rd[i].state.size = 0;
+			i = (i + 1) % session->nr_readers;
+		}
+	}
+
+	ret = ordered_events__flush(&session->ordered_events, OE_FLUSH__FINAL);
+	if (ret)
+		goto out_err;
+
+	ret = perf_session__flush_thread_stacks(session);
+out_err:
+	ui_progress__finish();
+
+	if (!tool->no_warn)
+		perf_session__warn_about_errors(session);
+
+	/*
+	 * We may switching perf.data output, make ordered_events
+	 * reusable.
+	 */
+	ordered_events__reinit(&session->ordered_events);
+
+	session->one_mmap = false;
+
+	return ret;
+}
+
 int perf_session__process_events(struct perf_session *session)
 {
 	if (perf_session__register_idle_thread(session) < 0)
@@ -2422,6 +2546,9 @@ int perf_session__process_events(struct perf_session *session)
 	if (perf_data__is_pipe(session->data))
 		return __perf_session__process_pipe_events(session);
 
+	if (perf_data__is_dir(session->data))
+		return __perf_session__process_dir_events(session);
+
 	return __perf_session__process_events(session);
 }
 
-- 
2.19.0

Re: [PATCH v5 00/20] Introduce threaded trace streaming for basic perf record operation

[Namhyung Kim]

Hello,

On Tue, May 4, 2021 at 12:05 AM Alexey Bayduraev
<alexey.v.bayduraev@linux.intel.com> wrote:
>
> Changes in v5:
> - fixed leaks in record__init_thread_masks_spec()
> - fixed leaks after failed realloc
> - replaced "%m" to strerror()
> - added masks examples to the documentation
> - captured Acked-by: tags by Andi Kleen
> - do not allow --thread option for full_auxtrace mode
> - split patch 06/12 to 06/20 and 07/20
> - split patch 08/12 to 09/20 and 10/20
> - split patches 11/12 and 11/12 to 13/20-20/20
>
> v4: https://lore.kernel.org/lkml/6c15adcb-6a9d-320e-70b5-957c4c8b6ff2@linux.intel.com/
>
> Changes in v4:
> - renamed 'comm' structure to 'pipes'
> - moved thread fd/maps messages to verbose=2
> - fixed leaks during allocation of thread_data structures
> - fixed leaks during allocation of thread masks
> - fixed possible fails when releasing thread masks
>
> v3: https://lore.kernel.org/lkml/7d197a2d-56e2-896d-bf96-6de0a4db1fb8@linux.intel.com/
>
> Changes in v3:
> - avoided skipped redundant patch 3/15
> - applied "data file" and "data directory" terms allover the patch set
> - captured Acked-by: tags by Namhyung Kim
> - avoided braces where don't needed
> - employed thread local variable for serial trace streaming
> - added specs for --thread option - core, socket, numa and user defined
> - added parallel loading of data directory files similar to the prototype [1]
>
> v2: https://lore.kernel.org/lkml/1ec29ed6-0047-d22f-630b-a7f5ccee96b4@linux.intel.com/
>
> Changes in v2:
> - explicitly added credit tags to patches 6/15 and 15/15,
>   additionally to cites [1], [2]
> - updated description of 3/15 to explicitly mention the reason
>   to open data directories in read access mode (e.g. for perf report)
> - implemented fix for compilation error of 2/15
> - explicitly elaborated on found issues to be resolved for
>   threaded AUX trace capture
>
> v1: https://lore.kernel.org/lkml/810f3a69-0004-9dff-a911-b7ff97220ae0@linux.intel.com/
>
> Patch set provides parallel threaded trace streaming mode for basic
> perf record operation. Provided mode mitigates profiling data losses
> and resolves scalability issues of serial and asynchronous (--aio)
> trace streaming modes on multicore server systems. The design and
> implementation are based on the prototype [1], [2].
>
> Parallel threaded mode executes trace streaming threads that read kernel
> data buffers and write captured data into several data files located at
> data directory. Layout of trace streaming threads and their mapping to data
> buffers to read can be configured using a value of --thread command line
> option. Specification value provides masks separated by colon so the masks
> define cpus to be monitored by one thread and thread affinity mask is
> separated by slash. <cpus mask 1>/<affinity mask 1>:<cpu mask 2>/<affinity mask 2>
> specifies parallel threads layout that consists of two threads with
> corresponding assigned cpus to be monitored. Specification value can be
> a string e.g. "cpu", "core" or "socket" meaning creation of data streaming
> thread for monitoring every cpu, whole core or socket. The option provided
> with no or empty value defaults to "cpu" layout creating data streaming
> thread for every cpu being monitored. Specification masks are filtered
> by the mask provided via -C option.
>
> Parallel streaming mode is compatible with Zstd compression/decompression
> (--compression-level) and external control commands (--control). The mode
> is not enabled for pipe mode. The mode is not enabled for AUX area tracing,
> related and derived modes like --snapshot or --aux-sample. --switch-output-*
> and --timestamp-filename options are not enabled for parallel streaming.
> Initial intent to enable AUX area tracing faced the need to define some
> optimal way to store index data in data directory. --switch-output-* and
> --timestamp-filename use cases are not clear for data directories.
> Asynchronous(--aio) trace streaming and affinity (--affinity) modes are
> mutually exclusive to parallel streaming mode.
>
> Basic analysis of data directories is provided in perf report mode.
> Raw dump and aggregated reports are available for data directories,
> still with no memory consumption optimizations.

Do you have an idea how to improve it?

I have to say again that I don't like merely adding more threads to
record.  Yeah, parallelizing the perf record is good, but we have to
think about the perf report (and others) too.

Thanks,
Namhyung


>
> Tested:
>
> tools/perf/perf record -o prof.data --threads -- matrix.gcc.g.O3
> tools/perf/perf record -o prof.data --threads= -- matrix.gcc.g.O3
> tools/perf/perf record -o prof.data --threads=cpu -- matrix.gcc.g.O3
> tools/perf/perf record -o prof.data --threads=core -- matrix.gcc.g.O3
> tools/perf/perf record -o prof.data --threads=socket -- matrix.gcc.g.O3
> tools/perf/perf record -o prof.data --threads=numa -- matrix.gcc.g.O3
> tools/perf/perf record -o prof.data --threads=0-3/3:4-7/4 -- matrix.gcc.g.O3
> tools/perf/perf record -o prof.data -C 2,5 --threads=0-3/3:4-7/4 -- matrix.gcc.g.O3
> tools/perf/perf record -o prof.data -C 3,4 --threads=0-3/3:4-7/4 -- matrix.gcc.g.O3
> tools/perf/perf record -o prof.data -C 0,4,2,6 --threads=core -- matrix.gcc.g.O3
> tools/perf/perf record -o prof.data -C 0,4,2,6 --threads=numa -- matrix.gcc.g.O3
> tools/perf/perf record -o prof.data --threads -g --call-graph dwarf,4096 -- matrix.gcc.g.O3
> tools/perf/perf record -o prof.data --threads -g --call-graph dwarf,4096 --compression-level=3 -- matrix.gcc.g.O3
> tools/perf/perf record -o prof.data --threads -a
> tools/perf/perf record -D -1 -e cpu-cycles -a --control fd:10,11 -- sleep 30
> tools/perf/perf record --threads -D -1 -e cpu-cycles -a --control fd:10,11 -- sleep 30
>
> tools/perf/perf report -i prof.data
> tools/perf/perf report -i prof.data --call-graph=callee
> tools/perf/perf report -i prof.data --stdio --header
> tools/perf/perf report -i prof.data -D --header
>
> [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-jolsa@kernel.org/
>
> ---
>
> Alexey Bayduraev (20):
>   perf record: introduce thread affinity and mmap masks
>   perf record: introduce thread specific data array
>   perf record: introduce thread local variable
>   perf record: stop threads in the end of trace streaming
>   perf record: start threads in the beginning of trace streaming
>   perf record: introduce data file at mmap buffer object
>   perf record: introduce data transferred and compressed stats
>   perf record: init data file at mmap buffer object
>   tools lib: introduce bitmap_intersects() operation
>   perf record: introduce --threads=<spec> command line option
>   perf record: document parallel data streaming mode
>   perf report: output data file name in raw trace dump
>   perf session: move reader structure to the top
>   perf session: introduce reader_state in reader object
>   perf session: introduce reader objects in session object
>   perf session: introduce decompressor into trace reader object
>   perf session: move init into reader__init function
>   perf session: move map/unmap into reader__mmap function
>   perf session: load single file for analysis
>   perf session: load data directory files for analysis
>
>  tools/include/linux/bitmap.h             |   11 +
>  tools/lib/api/fd/array.c                 |   17 +
>  tools/lib/api/fd/array.h                 |    1 +
>  tools/lib/bitmap.c                       |   14 +
>  tools/perf/Documentation/perf-record.txt |   30 +
>  tools/perf/builtin-inject.c              |    3 +-
>  tools/perf/builtin-record.c              | 1066 ++++++++++++++++++++--
>  tools/perf/util/evlist.c                 |   16 +
>  tools/perf/util/evlist.h                 |    1 +
>  tools/perf/util/mmap.c                   |    6 +
>  tools/perf/util/mmap.h                   |    6 +
>  tools/perf/util/ordered-events.h         |    1 +
>  tools/perf/util/record.h                 |    2 +
>  tools/perf/util/session.c                |  491 +++++++---
>  tools/perf/util/session.h                |    5 +
>  tools/perf/util/tool.h                   |    3 +-
>  16 files changed, 1474 insertions(+), 199 deletions(-)
>
> --
> 2.19.0
>

Re: [PATCH v5 00/20] Introduce threaded trace streaming for basic perf record operation

[Bayduraev, Alexey V]

Hi,

On 06.05.2021 9:20, Namhyung Kim wrote:
> Hello,
> 
> On Tue, May 4, 2021 at 12:05 AM Alexey Bayduraev
> <alexey.v.bayduraev@linux.intel.com> wrote:
>>
<SNIP>>>
>> Basic analysis of data directories is provided in perf report mode.
>> Raw dump and aggregated reports are available for data directories,
>> still with no memory consumption optimizations.
> 
> Do you have an idea how to improve it?
> 
> I have to say again that I don't like merely adding more threads to
> record.  Yeah, parallelizing the perf record is good, but we have to
> think about the perf report (and others) too.

There is your idea about separating tracking records and process them 
first, but these changes can be much larger than my patch and I think 
they looks like independent patch and could be introduced as extension 
of parallel data loading.

I also thought and experimented with the intermediate flushing of 
the ordered queue. This is simple for per-cpu data files (sorted 
by time), but not clear for arbitrary CPU masks.

I think my patch can be the first step to introduce parallel mode 
to the perf tool. It just extends perf-record (already used in our 
vtune tool) and allows to load parallel data in experimental mode. 
Next patches could optimize and extend parallel data loading.

Regards,
Alexey

> 
> Thanks,
> Namhyung
> 
> 
>>
>> Tested:
>>
>> tools/perf/perf record -o prof.data --threads -- matrix.gcc.g.O3
>> tools/perf/perf record -o prof.data --threads= -- matrix.gcc.g.O3
>> tools/perf/perf record -o prof.data --threads=cpu -- matrix.gcc.g.O3
>> tools/perf/perf record -o prof.data --threads=core -- matrix.gcc.g.O3
>> tools/perf/perf record -o prof.data --threads=socket -- matrix.gcc.g.O3
>> tools/perf/perf record -o prof.data --threads=numa -- matrix.gcc.g.O3
>> tools/perf/perf record -o prof.data --threads=0-3/3:4-7/4 -- matrix.gcc.g.O3
>> tools/perf/perf record -o prof.data -C 2,5 --threads=0-3/3:4-7/4 -- matrix.gcc.g.O3
>> tools/perf/perf record -o prof.data -C 3,4 --threads=0-3/3:4-7/4 -- matrix.gcc.g.O3
>> tools/perf/perf record -o prof.data -C 0,4,2,6 --threads=core -- matrix.gcc.g.O3
>> tools/perf/perf record -o prof.data -C 0,4,2,6 --threads=numa -- matrix.gcc.g.O3
>> tools/perf/perf record -o prof.data --threads -g --call-graph dwarf,4096 -- matrix.gcc.g.O3
>> tools/perf/perf record -o prof.data --threads -g --call-graph dwarf,4096 --compression-level=3 -- matrix.gcc.g.O3
>> tools/perf/perf record -o prof.data --threads -a
>> tools/perf/perf record -D -1 -e cpu-cycles -a --control fd:10,11 -- sleep 30
>> tools/perf/perf record --threads -D -1 -e cpu-cycles -a --control fd:10,11 -- sleep 30
>>
>> tools/perf/perf report -i prof.data
>> tools/perf/perf report -i prof.data --call-graph=callee
>> tools/perf/perf report -i prof.data --stdio --header
>> tools/perf/perf report -i prof.data -D --header
>>
>> [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-jolsa@kernel.org/
>>
>> ---
>>
>> Alexey Bayduraev (20):
>>   perf record: introduce thread affinity and mmap masks
>>   perf record: introduce thread specific data array
>>   perf record: introduce thread local variable
>>   perf record: stop threads in the end of trace streaming
>>   perf record: start threads in the beginning of trace streaming
>>   perf record: introduce data file at mmap buffer object
>>   perf record: introduce data transferred and compressed stats
>>   perf record: init data file at mmap buffer object
>>   tools lib: introduce bitmap_intersects() operation
>>   perf record: introduce --threads=<spec> command line option
>>   perf record: document parallel data streaming mode
>>   perf report: output data file name in raw trace dump
>>   perf session: move reader structure to the top
>>   perf session: introduce reader_state in reader object
>>   perf session: introduce reader objects in session object
>>   perf session: introduce decompressor into trace reader object
>>   perf session: move init into reader__init function
>>   perf session: move map/unmap into reader__mmap function
>>   perf session: load single file for analysis
>>   perf session: load data directory files for analysis
>>
>>  tools/include/linux/bitmap.h             |   11 +
>>  tools/lib/api/fd/array.c                 |   17 +
>>  tools/lib/api/fd/array.h                 |    1 +
>>  tools/lib/bitmap.c                       |   14 +
>>  tools/perf/Documentation/perf-record.txt |   30 +
>>  tools/perf/builtin-inject.c              |    3 +-
>>  tools/perf/builtin-record.c              | 1066 ++++++++++++++++++++--
>>  tools/perf/util/evlist.c                 |   16 +
>>  tools/perf/util/evlist.h                 |    1 +
>>  tools/perf/util/mmap.c                   |    6 +
>>  tools/perf/util/mmap.h                   |    6 +
>>  tools/perf/util/ordered-events.h         |    1 +
>>  tools/perf/util/record.h                 |    2 +
>>  tools/perf/util/session.c                |  491 +++++++---
>>  tools/perf/util/session.h                |    5 +
>>  tools/perf/util/tool.h                   |    3 +-
>>  16 files changed, 1474 insertions(+), 199 deletions(-)
>>
>> --
>> 2.19.0
>>

Re: [PATCH v5 00/20] Introduce threaded trace streaming for basic perf record operation

[Andi Kleen]

On 5/5/2021 11:20 PM, Namhyung Kim wrote:
>
> Do you have an idea how to improve it?
>
> I have to say again that I don't like merely adding more threads to
> record.  Yeah, parallelizing the perf record is good, but we have to
> think about the perf report (and others) too.

perf report/script can be already parallelized with --time xx/x% and a 
simple shell script that runs multiple processes. While that's a bit 
awkward for interactive use it works fine for scripting. I use it all 
the time for PT batch processing for example. The real bottleneck we 
have is really record on systems with many CPUs (which are more and more 
common), and that can only be fixed with some variant of this patch kit.

-Andi

Re: [PATCH v5 00/20] Introduce threaded trace streaming for basic perf record operation

[Namhyung Kim]

On Thu, May 6, 2021 at 5:44 AM Bayduraev, Alexey V
<alexey.v.bayduraev@linux.intel.com> wrote:
>
> Hi,
>
> On 06.05.2021 9:20, Namhyung Kim wrote:
> > Hello,
> >
> > On Tue, May 4, 2021 at 12:05 AM Alexey Bayduraev
> > <alexey.v.bayduraev@linux.intel.com> wrote:
> >>
> <SNIP>>>
> >> Basic analysis of data directories is provided in perf report mode.
> >> Raw dump and aggregated reports are available for data directories,
> >> still with no memory consumption optimizations.
> >
> > Do you have an idea how to improve it?
> >
> > I have to say again that I don't like merely adding more threads to
> > record.  Yeah, parallelizing the perf record is good, but we have to
> > think about the perf report (and others) too.
>
> There is your idea about separating tracking records and process them
> first, but these changes can be much larger than my patch and I think
> they looks like independent patch and could be introduced as extension
> of parallel data loading.
>
> I also thought and experimented with the intermediate flushing of
> the ordered queue. This is simple for per-cpu data files (sorted
> by time), but not clear for arbitrary CPU masks.
>
> I think my patch can be the first step to introduce parallel mode
> to the perf tool. It just extends perf-record (already used in our
> vtune tool) and allows to load parallel data in experimental mode.
> Next patches could optimize and extend parallel data loading.

Yeah I agree that we can change it incrementally and good to
know that you are thinking about the next step. :)

Thanks,
Namhyung

Re: [PATCH v5 00/20] Introduce threaded trace streaming for basic perf record operation

[Namhyung Kim]

Hi Andi,

On Thu, May 6, 2021 at 7:17 AM Andi Kleen <ak@linux.intel.com> wrote:
>
>
> On 5/5/2021 11:20 PM, Namhyung Kim wrote:
> >
> > Do you have an idea how to improve it?
> >
> > I have to say again that I don't like merely adding more threads to
> > record.  Yeah, parallelizing the perf record is good, but we have to
> > think about the perf report (and others) too.
>
> perf report/script can be already parallelized with --time xx/x% and a
> simple shell script that runs multiple processes. While that's a bit
> awkward for interactive use it works fine for scripting. I use it all
> the time for PT batch processing for example. The real bottleneck we
> have is really record on systems with many CPUs (which are more and more
> common), and that can only be fixed with some variant of this patch kit.

Right, spreading partial analysis to multiple processes would work
for some use cases.  I also agree that parallelizing perf record is
more important, but if there's a way to improve perf report/script
we should try that too.

Thanks,
Namhyung

Re: [PATCH v5 10/20] perf record: introduce --threads=<spec> command line option

[Namhyung Kim]

Hello,

On Tue, May 4, 2021 at 12:06 AM Alexey Bayduraev
<alexey.v.bayduraev@linux.intel.com> wrote:
>
> Provide --threads option in perf record command line interface.
> The option can have a value in the form of masks that specify
> cpus to be monitored with data streaming threads and its layout
> in system topology. The masks can be filtered using cpu mask
> provided via -C option.
>
> The specification value can be user defined list of masks. Masks
> separated by colon define cpus to be monitored by one thread and
> affinity mask of that thread is separated by slash. For example:
> <cpus mask 1>/<affinity mask 1>:<cpu mask 2>/<affinity mask 2>
> specifies parallel threads layout that consists of two threads
> with corresponding assigned cpus to be monitored.
>
> The specification value can be a string e.g. "cpu", "core" or
> "socket" meaning creation of data streaming thread for every
> cpu or core or socket to monitor distinct cpus or cpus grouped
> by core or socket.
>
> The option provided with no or empty value defaults to per-cpu
> parallel threads layout creating data streaming thread for every
> cpu being monitored.
>
> Feature design and implementation are based on prototypes [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-jolsa@kernel.org/
>
> Suggested-by: Jiri Olsa <jolsa@kernel.org>
> Suggested-by: Namhyung Kim <namhyung@kernel.org>
> Acked-by: Andi Kleen <ak@linux.intel.com>
> Signed-off-by: Alexey Bayduraev <alexey.v.bayduraev@linux.intel.com>
> ---
>  tools/perf/builtin-record.c | 342 +++++++++++++++++++++++++++++++++++-
>  tools/perf/util/record.h    |   1 +
>  2 files changed, 341 insertions(+), 2 deletions(-)
>
> diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c
> index bf730e1220dc..e5ea1334cc7d 100644
> --- a/tools/perf/builtin-record.c
> +++ b/tools/perf/builtin-record.c
> @@ -51,6 +51,7 @@
>  #include "util/evlist-hybrid.h"
>  #include "asm/bug.h"
>  #include "perf.h"
> +#include "cputopo.h"
>
>  #include <errno.h>
>  #include <inttypes.h>
> @@ -122,6 +123,20 @@ static const char *thread_msg_tags[THREAD_MSG__MAX] = {
>         "UNDEFINED", "READY"
>  };
>
> +enum thread_spec {
> +       THREAD_SPEC__UNDEFINED = 0,
> +       THREAD_SPEC__CPU,
> +       THREAD_SPEC__CORE,
> +       THREAD_SPEC__SOCKET,
> +       THREAD_SPEC__NUMA,
> +       THREAD_SPEC__USER,
> +       THREAD_SPEC__MAX,
> +};
> +
> +static const char *thread_spec_tags[THREAD_SPEC__MAX] = {
> +       "undefined", "cpu", "core", "socket", "numa", "user"
> +};
> +
>  struct record {
>         struct perf_tool        tool;
>         struct record_opts      opts;
> @@ -2704,6 +2719,70 @@ static void record__thread_mask_free(struct thread_mask *mask)
>         record__mmap_cpu_mask_free(&mask->affinity);
>  }
>
> +static int record__thread_mask_or(struct thread_mask *dest, struct thread_mask *src1,
> +                                  struct thread_mask *src2)
> +{
> +       if (src1->maps.nbits != src2->maps.nbits ||
> +           dest->maps.nbits != src1->maps.nbits ||
> +           src1->affinity.nbits != src2->affinity.nbits ||
> +           dest->affinity.nbits != src1->affinity.nbits)
> +               return -EINVAL;
> +
> +       bitmap_or(dest->maps.bits, src1->maps.bits,
> +                 src2->maps.bits, src1->maps.nbits);
> +       bitmap_or(dest->affinity.bits, src1->affinity.bits,
> +                 src2->affinity.bits, src1->affinity.nbits);
> +
> +       return 0;
> +}
> +
> +static int record__thread_mask_intersects(struct thread_mask *mask_1, struct thread_mask *mask_2)
> +{
> +       int res1, res2;
> +
> +       if (mask_1->maps.nbits != mask_2->maps.nbits ||
> +           mask_1->affinity.nbits != mask_2->affinity.nbits)
> +               return -EINVAL;
> +
> +       res1 = bitmap_intersects(mask_1->maps.bits, mask_2->maps.bits,
> +                                mask_1->maps.nbits);
> +       res2 = bitmap_intersects(mask_1->affinity.bits, mask_2->affinity.bits,
> +                                mask_1->affinity.nbits);
> +       if (res1 || res2)
> +               return 1;
> +
> +       return 0;
> +}
> +
> +static int record__parse_threads(const struct option *opt, const char *str, int unset)
> +{
> +       int s;
> +       struct record_opts *opts = opt->value;
> +
> +       if (unset || !str || !strlen(str)) {
> +               opts->threads_spec = THREAD_SPEC__CPU;
> +       } else {
> +               for (s = 1; s < THREAD_SPEC__MAX; s++) {
> +                       if (s == THREAD_SPEC__USER) {
> +                               opts->threads_user_spec = strdup(str);
> +                               opts->threads_spec = THREAD_SPEC__USER;
> +                               break;
> +                       }
> +                       if (!strncasecmp(str, thread_spec_tags[s], strlen(thread_spec_tags[s]))) {
> +                               opts->threads_spec = s;
> +                               break;
> +                       }
> +               }
> +       }
> +
> +       pr_debug("threads_spec: %s", thread_spec_tags[opts->threads_spec]);
> +       if (opts->threads_spec == THREAD_SPEC__USER)
> +               pr_debug("=[%s]", opts->threads_user_spec);
> +       pr_debug("\n");
> +
> +       return 0;
> +}
> +
>  static int parse_output_max_size(const struct option *opt,
>                                  const char *str, int unset)
>  {
> @@ -3134,6 +3213,9 @@ static struct option __record_options[] = {
>                      "\t\t\t  Optionally send control command completion ('ack\\n') to ack-fd descriptor.\n"
>                      "\t\t\t  Alternatively, ctl-fifo / ack-fifo will be opened and used as ctl-fd / ack-fd.",
>                       parse_control_option),
> +       OPT_CALLBACK_OPTARG(0, "threads", &record.opts, NULL, "spec",
> +                           "write collected trace data into several data files using parallel threads",
> +                           record__parse_threads),
>         OPT_END()
>  };
>
> @@ -3147,6 +3229,17 @@ static void record__mmap_cpu_mask_init(struct mmap_cpu_mask *mask, struct perf_c
>                 set_bit(cpus->map[c], mask->bits);
>  }
>
> +static void record__mmap_cpu_mask_init_spec(struct mmap_cpu_mask *mask, char *mask_spec)
> +{
> +       struct perf_cpu_map *cpus;
> +
> +       cpus = perf_cpu_map__new(mask_spec);
> +       if (cpus) {
> +               record__mmap_cpu_mask_init(mask, cpus);
> +               perf_cpu_map__put(cpus);
> +       }
> +}
> +
>  static int record__alloc_thread_masks(struct record *rec, int nr_threads, int nr_bits)
>  {
>         int t, ret;
> @@ -3166,6 +3259,224 @@ static int record__alloc_thread_masks(struct record *rec, int nr_threads, int nr
>
>         return 0;
>  }
> +
> +static int record__init_thread_cpu_masks(struct record *rec, struct perf_cpu_map *cpus)
> +{
> +       int t, ret, nr_cpus = perf_cpu_map__nr(cpus);
> +
> +       ret = record__alloc_thread_masks(rec, nr_cpus, cpu__max_cpu());
> +       if (ret)
> +               return ret;
> +
> +       rec->nr_threads = nr_cpus;
> +       pr_debug("threads: nr_threads=%d\n", rec->nr_threads);
> +
> +       for (t = 0; t < rec->nr_threads; t++) {
> +               set_bit(cpus->map[t], rec->thread_masks[t].maps.bits);
> +               pr_debug("thread_masks[%d]: maps mask [%d]\n", t, cpus->map[t]);
> +               set_bit(cpus->map[t], rec->thread_masks[t].affinity.bits);
> +               pr_debug("thread_masks[%d]: affinity mask [%d]\n", t, cpus->map[t]);
> +       }
> +
> +       return 0;
> +}
> +
> +static int record__init_thread_masks_spec(struct record *rec, struct perf_cpu_map *cpus,
> +                                         char **maps_spec, char **affinity_spec, u32 nr_spec)
> +{
> +       u32 s;
> +       int ret = 0, nr_threads = 0;
> +       struct mmap_cpu_mask cpus_mask;
> +       struct thread_mask thread_mask, full_mask, *prev_masks;
> +
> +       ret = record__mmap_cpu_mask_alloc(&cpus_mask, cpu__max_cpu());
> +       if (ret)
> +               goto out;
> +       record__mmap_cpu_mask_init(&cpus_mask, cpus);
> +       ret = record__thread_mask_alloc(&thread_mask, cpu__max_cpu());
> +       if (ret)
> +               goto out_free_cpu_mask;
> +       ret = record__thread_mask_alloc(&full_mask, cpu__max_cpu());
> +       if (ret)
> +               goto out_free_thread_mask;
> +       record__thread_mask_clear(&full_mask);
> +
> +       for (s = 0; s < nr_spec; s++) {
> +               record__thread_mask_clear(&thread_mask);
> +
> +               record__mmap_cpu_mask_init_spec(&thread_mask.maps, maps_spec[s]);
> +               record__mmap_cpu_mask_init_spec(&thread_mask.affinity, affinity_spec[s]);
> +
> +               if (!bitmap_and(thread_mask.maps.bits, thread_mask.maps.bits,
> +                               cpus_mask.bits, thread_mask.maps.nbits) ||
> +                   !bitmap_and(thread_mask.affinity.bits, thread_mask.affinity.bits,
> +                               cpus_mask.bits, thread_mask.affinity.nbits))
> +                       continue;
> +
> +               ret = record__thread_mask_intersects(&thread_mask, &full_mask);
> +               if (ret)
> +                       goto out_free_full_mask;
> +               record__thread_mask_or(&full_mask, &full_mask, &thread_mask);
> +
> +               prev_masks = rec->thread_masks;
> +               rec->thread_masks = realloc(rec->thread_masks,
> +                                           (nr_threads + 1) * sizeof(struct thread_mask));
> +               if (!rec->thread_masks) {
> +                       pr_err("Failed to allocate thread masks\n");
> +                       rec->thread_masks = prev_masks;
> +                       ret = -ENOMEM;
> +                       goto out_free_full_mask;
> +               }
> +               rec->thread_masks[nr_threads] = thread_mask;
> +               pr_debug("thread_masks[%d]: addr=", nr_threads);
> +               mmap_cpu_mask__scnprintf(&rec->thread_masks[nr_threads].maps, "maps");
> +               pr_debug("thread_masks[%d]: addr=", nr_threads);
> +               mmap_cpu_mask__scnprintf(&rec->thread_masks[nr_threads].affinity, "affinity");

It's not clear mmap_cpu_mask__scnprintf() will call pr_debug()
and it also has some more work to prepare the buffer.
Maybe we can add an 'if (verbose)' block to inform it's all for
debug information and skip the unnecessary work.


> +               nr_threads++;
> +               ret = record__thread_mask_alloc(&thread_mask, cpu__max_cpu());
> +               if (ret)
> +                       goto out_free_full_mask;

I wonder if it's possible to fail on the first bitmap and
have a pointer to the old bitmap for the second one.
It might lead to double free then..


> +       }
> +
> +       rec->nr_threads = nr_threads;
> +       pr_debug("threads: nr_threads=%d\n", rec->nr_threads);
> +
> +out_free_full_mask:
> +       record__thread_mask_free(&full_mask);
> +out_free_thread_mask:
> +       record__thread_mask_free(&thread_mask);
> +out_free_cpu_mask:
> +       record__mmap_cpu_mask_free(&cpus_mask);
> +out:
> +       return ret;
> +}
> +
> +static int record__init_thread_core_masks(struct record *rec, struct perf_cpu_map *cpus)
> +{
> +       int ret;
> +       struct cpu_topology *topo;
> +
> +       topo = cpu_topology__new();
> +       if (!topo)
> +               return -EINVAL;
> +
> +       ret = record__init_thread_masks_spec(rec, cpus, topo->thread_siblings,
> +                                            topo->thread_siblings, topo->thread_sib);
> +       cpu_topology__delete(topo);
> +
> +       return ret;
> +}
> +
> +static int record__init_thread_socket_masks(struct record *rec, struct perf_cpu_map *cpus)
> +{
> +       int ret;
> +       struct cpu_topology *topo;
> +
> +       topo = cpu_topology__new();
> +       if (!topo)
> +               return -EINVAL;
> +
> +       ret = record__init_thread_masks_spec(rec, cpus, topo->core_siblings,
> +                                            topo->core_siblings, topo->core_sib);
> +       cpu_topology__delete(topo);
> +
> +       return ret;
> +}
> +
> +static int record__init_thread_numa_masks(struct record *rec, struct perf_cpu_map *cpus)
> +{
> +       u32 s;
> +       int ret;
> +       char **spec;
> +       struct numa_topology *topo;
> +
> +       topo = numa_topology__new();
> +       if (!topo)
> +               return -EINVAL;
> +       spec = zalloc(topo->nr * sizeof(char *));
> +       if (!spec) {
> +               ret = -ENOMEM;
> +               goto out_delete_topo;
> +       }
> +       for (s = 0; s < topo->nr; s++)
> +               spec[s] = topo->nodes[s].cpus;
> +
> +       ret = record__init_thread_masks_spec(rec, cpus, spec, spec, topo->nr);
> +
> +       zfree(&spec);
> +
> +out_delete_topo:
> +       numa_topology__delete(topo);
> +
> +       return ret;
> +}
> +
> +static int record__init_thread_user_masks(struct record *rec, struct perf_cpu_map *cpus)
> +{
> +       int t, ret;
> +       u32 s, nr_spec = 0;
> +       char **maps_spec = NULL, **affinity_spec = NULL, **prev_spec;
> +       char *spec, *spec_ptr, *user_spec, *mask, *mask_ptr;
> +
> +       for (t = 0, user_spec = (char *)rec->opts.threads_user_spec; ; t++, user_spec = NULL) {
> +               spec = strtok_r(user_spec, ":", &spec_ptr);
> +               if (spec == NULL)
> +                       break;
> +               pr_debug(" spec[%d]: %s\n", t, spec);
> +               mask = strtok_r(spec, "/", &mask_ptr);
> +               if (mask == NULL)
> +                       break;
> +               pr_debug("  maps mask: %s\n", mask);
> +               prev_spec = maps_spec;
> +               maps_spec = realloc(maps_spec, (nr_spec + 1) * sizeof(char *));
> +               if (!maps_spec) {
> +                       pr_err("Failed to realloc maps_spec\n");
> +                       maps_spec = prev_spec;
> +                       ret = -ENOMEM;
> +                       goto out_free_all_specs;
> +               }
> +               maps_spec[nr_spec] = strdup(mask);
> +               if (!maps_spec[nr_spec]) {
> +                       pr_err("Failed to alloc maps_spec[%d]\n", nr_spec);
> +                       ret = -ENOMEM;
> +                       goto out_free_all_specs;
> +               }
> +               mask = strtok_r(NULL, "/", &mask_ptr);
> +               if (mask == NULL)
> +                       break;
> +               pr_debug("  affinity mask: %s\n", mask);
> +               prev_spec = affinity_spec;
> +               affinity_spec = realloc(affinity_spec, (nr_spec + 1) * sizeof(char *));
> +               if (!affinity_spec) {
> +                       pr_err("Failed to realloc affinity_spec\n");
> +                       affinity_spec = prev_spec;
> +                       ret = -ENOMEM;
> +                       goto out_free_all_specs;

This will leak the last maps_spec as nr_spec was not increased.


> +               }
> +               affinity_spec[nr_spec] = strdup(mask);
> +               if (!affinity_spec[nr_spec]) {
> +                       pr_err("Failed to alloc affinity_spec[%d]\n", nr_spec);
> +                       ret = -ENOMEM;
> +                       goto out_free_all_specs;

Ditto.

Thanks,
Namhyung


> +               }
> +               nr_spec++;
> +       }
> +
> +       ret = record__init_thread_masks_spec(rec, cpus, maps_spec, affinity_spec, nr_spec);
> +
> +out_free_all_specs:
> +       for (s = 0; s < nr_spec; s++) {
> +               if (maps_spec)
> +                       free(maps_spec[s]);
> +               if (affinity_spec)
> +                       free(affinity_spec[s]);
> +       }
> +       free(affinity_spec);
> +       free(maps_spec);
> +
> +       return ret;
> +}
> +
>  static int record__init_thread_default_masks(struct record *rec, struct perf_cpu_map *cpus)
>  {
>         int ret;
> @@ -3183,9 +3494,33 @@ static int record__init_thread_default_masks(struct record *rec, struct perf_cpu
>
>  static int record__init_thread_masks(struct record *rec)
>  {
> +       int ret = 0;
>         struct perf_cpu_map *cpus = rec->evlist->core.cpus;
>
> -       return record__init_thread_default_masks(rec, cpus);
> +       if (!record__threads_enabled(rec))
> +               return record__init_thread_default_masks(rec, cpus);
> +
> +       switch (rec->opts.threads_spec) {
> +       case THREAD_SPEC__CPU:
> +               ret = record__init_thread_cpu_masks(rec, cpus);
> +               break;
> +       case THREAD_SPEC__CORE:
> +               ret = record__init_thread_core_masks(rec, cpus);
> +               break;
> +       case THREAD_SPEC__SOCKET:
> +               ret = record__init_thread_socket_masks(rec, cpus);
> +               break;
> +       case THREAD_SPEC__NUMA:
> +               ret = record__init_thread_numa_masks(rec, cpus);
> +               break;
> +       case THREAD_SPEC__USER:
> +               ret = record__init_thread_user_masks(rec, cpus);
> +               break;
> +       default:
> +               break;
> +       }
> +
> +       return ret;
>  }
>
>  static int record__fini_thread_masks(struct record *rec)
> @@ -3436,7 +3771,10 @@ int cmd_record(int argc, const char **argv)
>
>         err = record__init_thread_masks(rec);
>         if (err) {
> -               pr_err("record__init_thread_masks failed, error %d\n", err);
> +               if (err > 0)
> +                       pr_err("ERROR: parallel data streaming masks (--threads) intersect.\n");
> +               else
> +                       pr_err("record__init_thread_masks failed, error %d\n", err);
>                 goto out;
>         }
>
> diff --git a/tools/perf/util/record.h b/tools/perf/util/record.h
> index 4d68b7e27272..3da156498f47 100644
> --- a/tools/perf/util/record.h
> +++ b/tools/perf/util/record.h
> @@ -78,6 +78,7 @@ struct record_opts {
>         int           ctl_fd_ack;
>         bool          ctl_fd_close;
>         int           threads_spec;
> +       const char    *threads_user_spec;
>  };
>
>  extern const char * const *record_usage;
> --
> 2.19.0
>