Skip to contentSkip to conversation

Introduce threaded trace streaming for basic perf record operation

PATCH Superseded Alexey Bayduraev 2 participants 35 messages Aug 17, 2021 – Sep 20, 2021 perf events

v11 of 13 Superseded Latest is v13 →

lore
Share & tools
lore
#0 lore raw

v11 cover Introduce threaded trace streaming for basic perf record operation

Changes in v11:
- removed python dependency on zstd (perf test 19)
- captured tags from Riccardo Mancini

v10: https://lore.kernel.org/lkml/[email protected]/

Changes in v10:
- renamed fdarray__clone to fdarray__dup_entry_from
- captured Acked-by: tags by Namhyung Kim for 09/24

v9: https://lore.kernel.org/lkml/[email protected]/

Changes in v9:
- fixes in [v9 01/24]:
- move 'nr_threads' to before 'thread_masks'
- combined decl+assign into one line in record__thread_mask_alloc
- releasing masks inplace in record__alloc_thread_masks
- split patch [v8 02/22] to [v9 02/24] and [v9 03/24]
- fixes in [v9 03/24]:
- renamed 'struct thread_data' to 'struct record_thread'
- moved nr_mmaps after ctlfd_pos
- releasing resources inplace in record__thread_data_init_maps
- initializing pipes by -1 value
- added temporary gettid() wrapper
- split patch [v8 03/22] to [v9 04/24] and [v9 05/24]
- removed upstreamed [v8 09/22]
- split [v8 10/22] to [v9 12/24] and [v9 13/24]
- moved --threads documentation to the related patches
- fixed output of written/compressed stats in [v9 10/24]
- split patch [v8 12/22] to [v9 15/24] and [v9 16/24]
- fixed order of error checking for decompressed events in [v9 16/24]
- merged patch [v8 21/22] with [v9 23/24] and [v9 24/24]
- moved patch [v8 22/22] to [v9 09/24]
- added max reader size constant in [v9 24/24]

v8: https://lore.kernel.org/lkml/[email protected]/

Changes in v8:
- captured Acked-by: tags by Namhyung Kim
- merged with origin/perf/core
- added patch 21/22 introducing READER_NODATA state
- added patch 22/22 fixing --max-size option

v7: https://lore.kernel.org/lkml/[email protected]/

Changes in v7:
- fixed possible crash after out_free_threads label
- added missing pthread_attr_destroy() call
- added check of correctness of user masks
- fixed zsts_data finalization

v6: https://lore.kernel.org/lkml/[email protected]/

Changes in v6:
- fixed leaks and possible double free in record__thread_mask_alloc()
- fixed leaks in record__init_thread_user_masks()
- fixed final mmaps flushing for threads id > 0
- merged with origin/perf/core

v5: https://lore.kernel.org/lkml/[email protected]/

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/[email protected]/

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/[email protected]/

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/[email protected]/

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/[email protected]/

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/[email protected]/

Alexey Bayduraev (24):
perf record: Introduce thread affinity and mmap masks tools lib: Introduce fdarray duplicate function
perf record: Introduce thread specific data array perf record: Introduce function to propagate control commands 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 bytes written stats to support --max-size
option
perf record: Introduce data transferred and compressed stats perf record: Init data file at mmap buffer object perf record: Introduce --threads command line option perf record: Extend --threads command line option perf record: Implement compatibility checks
perf report: Output non-zero offset for decompressed records 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/lib/api/fd/array.c                 |   17 +
 tools/lib/api/fd/array.h                 |    1 +
 tools/perf/Documentation/perf-record.txt |   30 +
 tools/perf/builtin-inject.c              |    3 +-
 tools/perf/builtin-kvm.c                 |    2 +-
 tools/perf/builtin-record.c              | 1197 ++++++++++++++++++++--
 tools/perf/builtin-top.c                 |    2 +-
 tools/perf/builtin-trace.c               |    2 +-
 tools/perf/util/evlist.c                 |   16 +
 tools/perf/util/evlist.h                 |    1 +
 tools/perf/util/mmap.c                   |   10 +
 tools/perf/util/mmap.h                   |    6 +
 tools/perf/util/ordered-events.c         |    3 +-
 tools/perf/util/ordered-events.h         |    3 +-
 tools/perf/util/record.h                 |    2 +
 tools/perf/util/session.c                |  506 ++++++---
 tools/perf/util/session.h                |    7 +-
 tools/perf/util/tool.h                   |    3 +-
 18 files changed, 1597 insertions(+), 214 deletions(-)
Signature · 1 line
2.19.0
#1 lore raw

PATCH v11 1/24 perf record: Introduce thread affinity and mmap masks

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.

---
 tools/perf/builtin-record.c | 135 ++++++++++++++++++++++++++++++++++++
 1 file changed, 135 insertions(+)
tools/perf/builtin-record.c +135 -0
index 548c1dbde6c5..021a607510b5 100644
@@ -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 */
+	int			nr_threads;
+	struct thread_mask	*thread_masks;
 };
 
 static volatile int done;
@@ -2187,6 +2194,54 @@ 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)
+{
+	int ret = record__mmap_cpu_mask_alloc(&mask->maps, nr_bits);
+
+	if (ret) {
+		mask->affinity.bits = NULL;
+		return ret;
+	}
+
+	ret = record__mmap_cpu_mask_alloc(&mask->affinity, nr_bits);
+	if (ret) {
+		record__mmap_cpu_mask_free(&mask->maps);
+		mask->maps.bits = NULL;
+	}
+
+	return ret;
+}
+
+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)
 {
@@ -2635,6 +2690,79 @@ 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 void record__free_thread_masks(struct record *rec, int nr_threads)
+{
+	int t;
+
+	if (rec->thread_masks)
+		for (t = 0; t < nr_threads; t++)
+			record__thread_mask_free(&rec->thread_masks[t]);
+
+	zfree(&rec->thread_masks);
+}
+
+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)
+			goto out_free;
+		record__thread_mask_clear(&rec->thread_masks[t]);
+	}
+
+	return 0;
+
+out_free:
+	record__free_thread_masks(rec, nr_threads);
+
+	return ret;
+}
+
+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 void record__fini_thread_masks(struct record *rec)
+{
+	record__free_thread_masks(rec, rec->nr_threads);
+
+	rec->nr_threads = 0;
+}
+
 int cmd_record(int argc, const char **argv)
 {
 	int err;
@@ -2883,6 +3011,12 @@ 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);
@@ -2901,6 +3035,7 @@ 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;
 }
Signature · 1 line
2.19.0
#2 lore raw

Re: [PATCH v11 01/24] perf record: Introduce thread affinity and mmap masks

On Tue, Aug 17, 2021 at 11:23:04AM +0300, Alexey Bayduraev wrote:

SNIP

Quoted 17 lines
tools/perf/builtin-record.c +17 -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)
+{

tools/perf/builtin-record.c static struct option __record_options[] = { · line 2763

Quoted 10 lines
tools/perf/builtin-record.c +10 -0
+	struct perf_cpu_map *cpus = rec->evlist->core.cpus;
+
+	return record__init_thread_default_masks(rec, cpus);
+}
+
+static void record__fini_thread_masks(struct record *rec)
+{
+	record__free_thread_masks(rec, rec->nr_threads);
+
+	rec->nr_threads = 0;

hum, is this needed?

jirka

Quoted 29 lines
tools/perf/builtin-record.c +9 -0
+}
+
 int cmd_record(int argc, const char **argv)
 {
 	int err;
@@ -2883,6 +3011,12 @@ 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);
@@ -2901,6 +3035,7 @@ 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;
 }
Signature · 1 line
2.19.0
#3 lore raw

PATCH v11 2/24 tools lib: Introduce fdarray duplicate function

In reply to Alexey Bayduraev

Introduce a function to duplicate an existing file descriptor in the fdarray structure. The function returns the position of the duplicated file descriptor.

---
 tools/lib/api/fd/array.c | 17 +++++++++++++++++
 tools/lib/api/fd/array.h |  1 +
 2 files changed, 18 insertions(+)
tools/lib/api/fd/array.c +17 -0
index 5e6cb9debe37..f0f195207fca 100644
@@ -88,6 +88,23 @@ int fdarray__add(struct fdarray *fda, int fd, short revents, enum fdarray_flags
 	return pos;
 }
 
+int fdarray__dup_entry_from(struct fdarray *fda, int pos, struct fdarray *from)
+{
+	struct pollfd *entry;
+	int npos;
+
+	if (pos >= from->nr)
+		return -EINVAL;
+
+	entry = &from->entries[pos];
+
+	npos = fdarray__add(fda, entry->fd, entry->events, from->priv[pos].flags);
+	if (npos >= 0)
+		fda->priv[npos] = from->priv[pos];
+
+	return npos;
+}
+
 int fdarray__filter(struct fdarray *fda, short revents,
 		    void (*entry_destructor)(struct fdarray *fda, int fd, void *arg),
 		    void *arg)
tools/lib/api/fd/array.h +1 -0
index 7fcf21a33c0c..60ad197c8ee9 100644
@@ -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__dup_entry_from(struct fdarray *fda, int pos, struct fdarray *from);
 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),
Signature · 1 line
2.19.0
#4 lore raw

PATCH v11 3/24 perf record: Introduce thread specific data array

In reply to 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.

---
 tools/perf/builtin-record.c | 254 +++++++++++++++++++++++++++++++++++-
 1 file changed, 251 insertions(+), 3 deletions(-)
tools/perf/builtin-record.c +251 -3
index 021a607510b5..50bdf544eeeb 100644
@@ -58,6 +58,7 @@
 #include <poll.h>
 #include <pthread.h>
 #include <unistd.h>
+#include <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 record_thread {
+	pid_t			tid;
+	struct thread_mask	*mask;
+	struct {
+		int		msg[2];
+		int		ack[2];
+	} pipes;
+	struct fdarray		pollfd;
+	int			ctlfd_pos;
+	int			nr_mmaps;
+	struct mmap		**maps;
+	struct mmap		**overwrite_maps;
+	struct record		*rec;
+	unsigned long long	samples;
+	unsigned long		waking;
+};
+
 struct record {
 	struct perf_tool	tool;
 	struct record_opts	opts;
@@ -118,6 +136,7 @@ struct record {
 	unsigned long		output_max_size;	/* = 0: unlimited */
 	int			nr_threads;
 	struct thread_mask	*thread_masks;
+	struct record_thread	*thread_data;
 };
 
 static volatile int done;
@@ -130,6 +149,13 @@ static const char *affinity_tags[PERF_AFFINITY_MAX] = {
 	"SYS", "NODE", "CPU"
 };
 
+#ifndef HAVE_GETTID
+static inline pid_t gettid(void)
+{
+	return (pid_t)syscall(__NR_gettid);
+}
+#endif
+
 static bool switch_output_signal(struct record *rec)
 {
 	return rec->switch_output.signal &&
@@ -847,9 +873,225 @@ static int record__kcore_copy(struct machine *machine, struct perf_data *data)
 	return kcore_copy(from_dir, kcore_dir);
 }
 
+static void record__thread_data_init_pipes(struct record_thread *thread_data)
+{
+	thread_data->pipes.msg[0] = -1;
+	thread_data->pipes.msg[1] = -1;
+	thread_data->pipes.ack[0] = -1;
+	thread_data->pipes.ack[1] = -1;
+}
+
+static int record__thread_data_open_pipes(struct record_thread *thread_data)
+{
+	if (pipe(thread_data->pipes.msg)) {
+		pr_err("Failed to create thread msg communication pipe: %s\n", strerror(errno));
+		return -ENOMEM;
+	}
+
+	if (pipe(thread_data->pipes.ack)) {
+		pr_err("Failed to create thread ack communication pipe: %s\n", strerror(errno));
+		close(thread_data->pipes.msg[0]);
+		thread_data->pipes.msg[0] = -1;
+		close(thread_data->pipes.msg[1]);
+		thread_data->pipes.msg[1] = -1;
+		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 void record__thread_data_close_pipes(struct record_thread *thread_data)
+{
+	if (thread_data->pipes.msg[0] != -1) {
+		close(thread_data->pipes.msg[0]);
+		thread_data->pipes.msg[0] = -1;
+	}
+	if (thread_data->pipes.msg[1] != -1) {
+		close(thread_data->pipes.msg[1]);
+		thread_data->pipes.msg[1] = -1;
+	}
+	if (thread_data->pipes.ack[0] != -1) {
+		close(thread_data->pipes.ack[0]);
+		thread_data->pipes.ack[0] = -1;
+	}
+	if (thread_data->pipes.ack[1] != -1) {
+		close(thread_data->pipes.ack[1]);
+		thread_data->pipes.ack[1] = -1;
+	}
+}
+
+static int record__thread_data_init_maps(struct record_thread *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 thread maps\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 thread overwrite maps\n");
+			zfree(&thread_data->maps);
+			return -ENOMEM;
+		}
+	}
+	pr_debug2("thread_data[%p]: nr_mmaps=%d, maps=%p, ow_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]: ow_maps[%d] -> ow_mmap[%d], cpus[%d]\n",
+					  thread_data, tm, m, cpus->map[m]);
+			}
+			tm++;
+		}
+	}
+
+	return 0;
+}
+
+static int record__thread_data_init_pollfd(struct record_thread *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__dup_entry_from(&thread_data->pollfd, f,
+							      &evlist->core.pollfd);
+				if (pos < 0) {
+					pr_err("Failed to duplicate descriptor in thread pollfd\n");
+					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 void record__free_thread_data(struct record *rec)
+{
+	int t;
+	struct record_thread *thread_data = rec->thread_data;
+
+	if (thread_data == NULL)
+		return;
+
+	for (t = 0; t < rec->nr_threads; t++) {
+		record__thread_data_close_pipes(&thread_data[t]);
+		zfree(&thread_data[t].maps);
+		zfree(&thread_data[t].overwrite_maps);
+		fdarray__exit(&thread_data[t].pollfd);
+	}
+
+	zfree(&rec->thread_data);
+}
+
+static int record__alloc_thread_data(struct record *rec, struct evlist *evlist)
+{
+	int t, ret;
+	struct record_thread *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++)
+		record__thread_data_init_pipes(&thread_data[t]);
+
+	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)
+			goto out_free;
+		ret = record__thread_data_init_pollfd(&thread_data[t], evlist);
+		if (ret) {
+			pr_err("Failed to initialize thread pollfd\n");
+			goto out_free;
+		}
+		if (t) {
+			thread_data[t].tid = -1;
+			ret = record__thread_data_open_pipes(&thread_data[t]);
+			if (ret)
+				goto out_free;
+			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) {
+				pr_err("Failed to add descriptor to thread pollfd\n");
+				ret = -ENOMEM;
+				goto out_free;
+			}
+			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 = gettid();
+			if (evlist->ctl_fd.pos == -1)
+				continue;
+			thread_data[t].ctlfd_pos = fdarray__dup_entry_from(&thread_data[t].pollfd,
+									   evlist->ctl_fd.pos,
+									   &evlist->core.pollfd);
+			if (thread_data[t].ctlfd_pos < 0) {
+				pr_err("Failed to duplicate descriptor in thread pollfd\n");
+				ret = -ENOMEM;
+				goto out_free;
+			}
+			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;
+
+out_free:
+	record__free_thread_data(rec);
+
+	return ret;
+}
+
 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 +1122,14 @@ 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;
 }
 
@@ -1851,9 +2101,6 @@ 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) {
@@ -2011,6 +2258,7 @@ static int __cmd_record(struct record *rec, int argc, const char **argv)
 out_child:
 	evlist__finalize_ctlfd(rec->evlist);
 	record__mmap_read_all(rec, true);
+	record__free_thread_data(rec);
 	record__aio_mmap_read_sync(rec);
 
 	if (rec->session->bytes_transferred && rec->session->bytes_compressed) {
Signature · 1 line
2.19.0
#5 lore raw

PATCH v11 4/24 perf record: Introduce function to propagate control commands

In reply to Alexey Bayduraev

Introduce evlist__ctlfd_update() function to propagate external control commands to global evlist object.

---
 tools/perf/util/evlist.c | 16 ++++++++++++++++
 tools/perf/util/evlist.h |  1 +
 2 files changed, 17 insertions(+)
tools/perf/util/evlist.c +16 -0
index 47581a237c7a..088f57593ebc 100644
@@ -2114,6 +2114,22 @@ 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;
tools/perf/util/evlist.h +1 -0
index 5c22383489ae..2429c7bcb4b6 100644
@@ -356,6 +356,7 @@ 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);
 
Signature · 1 line
2.19.0
#6 lore raw

PATCH v11 5/24 perf record: Introduce thread local variable

In reply to 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.

---
 tools/perf/builtin-record.c | 139 ++++++++++++++++++++++++------------
 1 file changed, 93 insertions(+), 46 deletions(-)
tools/perf/builtin-record.c +93 -46
index 50bdf544eeeb..9d6f6f8d788a 100644
@@ -110,6 +110,8 @@ struct record_thread {
 	unsigned long		waking;
 };
 
+static __thread struct record_thread *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 */
 	int			nr_threads;
 	struct thread_mask	*thread_masks;
@@ -574,7 +575,7 @@ 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);
 }
 
@@ -1317,19 +1318,29 @@ static struct perf_event_header finished_round_event = {
 	.type = PERF_RECORD_FINISHED_ROUND,
 };
 
-static void record__adjust_affinity(struct record *rec, struct mmap *map)
+static int 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());
+		}
 	}
+
+	return ret;
 }
 
 static size_t process_comp_header(void *record, size_t increment)
@@ -1369,14 +1380,17 @@ 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;
 
@@ -1386,9 +1400,9 @@ 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);
@@ -1451,6 +1465,15 @@ 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;
@@ -1865,6 +1888,33 @@ static void record__uniquify_name(struct record *rec)
 	}
 }
 
+static int record__start_threads(struct record *rec)
+{
+	struct record_thread *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 record_thread *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;
@@ -1973,7 +2023,7 @@ 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;
 
@@ -1981,7 +2031,7 @@ 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;
 		}
 	}
 
@@ -1992,7 +2042,7 @@ 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;
 	}
 
 	/*
@@ -2010,11 +2060,11 @@ 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;
@@ -2022,16 +2072,16 @@ 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;
@@ -2040,10 +2090,13 @@ 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,
@@ -2114,7 +2167,7 @@ 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
@@ -2183,20 +2236,24 @@ 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) {
@@ -2249,18 +2306,20 @@ 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:
-	evlist__finalize_ctlfd(rec->evlist);
+	record__stop_threads(rec, &waking);
 	record__mmap_read_all(rec, true);
+out_free_threads:
 	record__free_thread_data(rec);
+	evlist__finalize_ctlfd(rec->evlist);
 	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;
@@ -3130,17 +3189,6 @@ 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;
@@ -3278,7 +3326,6 @@ 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);
Signature · 1 line
2.19.0
#7 lore raw

PATCH v11 6/24 perf record: Stop threads in the end of trace streaming

In reply to 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.

---
 tools/perf/builtin-record.c | 31 +++++++++++++++++++++++++++++++
 1 file changed, 31 insertions(+)
tools/perf/builtin-record.c +31 -0
index 9d6f6f8d788a..04e77b5487bb 100644
@@ -112,6 +112,16 @@ struct record_thread {
 
 static __thread struct record_thread *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;
@@ -1888,6 +1898,24 @@ static void record__uniquify_name(struct record *rec)
 	}
 }
 
+static int record__terminate_thread(struct record_thread *thread_data)
+{
+	int res;
+	enum thread_msg ack = THREAD_MSG__UNDEFINED;
+	pid_t tid = thread_data->tid;
+
+	close(thread_data->pipes.msg[1]);
+	thread_data->pipes.msg[1] = -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 record_thread *thread_data = rec->thread_data;
@@ -1904,6 +1932,9 @@ static int record__stop_threads(struct record *rec, unsigned long *waking)
 	int t;
 	struct record_thread *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;
Signature · 1 line
2.19.0
#8 lore raw

PATCH v11 7/24 perf record: Start threads in the beginning of trace streaming

In reply to 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.

---
 tools/perf/builtin-record.c | 109 +++++++++++++++++++++++++++++++++++-
 1 file changed, 108 insertions(+), 1 deletion(-)
tools/perf/builtin-record.c +108 -1
index 04e77b5487bb..fb1c1b46dd04 100644
@@ -1484,6 +1484,67 @@ 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 = 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]);
+			thread->pipes.msg[0] = -1;
+			pollfd->entries[ctlfd_pos].fd = -1;
+			pollfd->entries[ctlfd_pos].events = 0;
+		}
+
+		pollfd->entries[ctlfd_pos].revents = 0;
+	}
+	record__mmap_read_all(thread->rec, true);
+
+	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;
@@ -1918,13 +1979,59 @@ static int record__terminate_thread(struct record_thread *thread_data)
 
 static int record__start_threads(struct record *rec)
 {
+	int t, tt, ret = 0, nr_threads = rec->nr_threads;
 	struct record_thread *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:
+	pthread_attr_destroy(&attrs);
+
+	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)
Signature · 1 line
2.19.0
#9 lore raw

Re: [PATCH v11 07/24] perf record: Start threads in the beginning of trace streaming

On Tue, Aug 17, 2021 at 11:23:10AM +0300, Alexey Bayduraev wrote:

SNIP

Quoted 19 lines
tools/perf/builtin-record.c +15 -0
+			pollfd->entries[ctlfd_pos].events = 0;
+		}
+
+		pollfd->entries[ctlfd_pos].revents = 0;
+	}
+	record__mmap_read_all(thread->rec, true);
+
+	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;

tools/perf/builtin-record.c static int record__terminate_thread(struct record_thread *thread_data) · line 1986

Quoted 8 lines
tools/perf/builtin-record.c +4 -0
@@ -1918,13 +1979,59 @@ static int record__terminate_thread(struct record_thread *thread_data)
 
 static int record__start_threads(struct record *rec)
 {
+	int t, tt, ret = 0, nr_threads = rec->nr_threads;
 	struct record_thread *thread_data = rec->thread_data;
+	sigset_t full, mask;
+	pthread_t handle;
+	pthread_attr_t attrs;

should we return here in case nr_threads == 1?

jirka

Quoted 55 lines
tools/perf/builtin-record.c +43 -1
+
+	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:
+	pthread_attr_destroy(&attrs);
+
+	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)
Signature · 1 line
2.19.0
#10 lore raw

PATCH v11 8/24 perf record: Introduce data file at mmap buffer object

In reply to 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.

---
 tools/perf/builtin-record.c |  3 +++
 tools/perf/util/mmap.c      | 10 ++++++++++
 tools/perf/util/mmap.h      |  3 +++
 3 files changed, 16 insertions(+)
tools/perf/builtin-record.c +3 -0
index fb1c1b46dd04..eff6f8db60b2 100644
@@ -197,6 +197,9 @@ 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;
tools/perf/util/mmap.c +10 -0
index ab7108d22428..b693b828a840 100644
@@ -230,6 +230,10 @@ void mmap__munmap(struct mmap *map)
 {
 	bitmap_free(map->affinity_mask.bits);
 
+#ifndef PYTHON_PERF
+	zstd_fini(&map->zstd_data);
+#endif
+
 	perf_mmap__aio_munmap(map);
 	if (map->data != NULL) {
 		munmap(map->data, mmap__mmap_len(map));
@@ -291,6 +295,12 @@ 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;
+#ifndef PYTHON_PERF
+	if (zstd_init(&map->zstd_data, map->comp_level)) {
+		pr_debug2("failed to init mmap commpressor, error %d\n", errno);
+		return -1;
+	}
+#endif
 
 	if (map->comp_level && !perf_mmap__aio_enabled(map)) {
 		map->data = mmap(NULL, mmap__mmap_len(map), PROT_READ|PROT_WRITE,
tools/perf/util/mmap.h +3 -0
index 9d5f589f02ae..c4aed6e89549 100644
@@ -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 {
Signature · 1 line
2.19.0
#11 lore raw

Re: [PATCH v11 08/24] perf record: Introduce data file at mmap buffer object

Quoted 20 lines · 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.

---
 tools/perf/builtin-record.c |  3 +++
 tools/perf/util/mmap.c      | 10 ++++++++++
 tools/perf/util/mmap.h      |  3 +++
 3 files changed, 16 insertions(+)
tools/perf/builtin-record.c +0 -0
index fb1c1b46dd04..eff6f8db60b2 100644

tools/perf/builtin-record.c static int record__write(struct record *rec, struct mmap *map __maybe_unused, · line 201

Quoted 5 lines
tools/perf/builtin-record.c +2 -0
@@ -197,6 +197,9 @@ 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;

I understand this change, but not sure how it's related to the changes below, should this be separated? also should this change be part of the patch where we actually set map->file?

jirka

Quoted 56 lines
tools/perf/builtin-record.c +1 -0
+
 	if (perf_data_file__write(file, bf, size) < 0) {
 		pr_err("failed to write perf data, error: %m\n");
 		return -1;
tools/perf/util/mmap.c +10 -0
index ab7108d22428..b693b828a840 100644
@@ -230,6 +230,10 @@ void mmap__munmap(struct mmap *map)
 {
 	bitmap_free(map->affinity_mask.bits);
 
+#ifndef PYTHON_PERF
+	zstd_fini(&map->zstd_data);
+#endif
+
 	perf_mmap__aio_munmap(map);
 	if (map->data != NULL) {
 		munmap(map->data, mmap__mmap_len(map));
@@ -291,6 +295,12 @@ 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;
+#ifndef PYTHON_PERF
+	if (zstd_init(&map->zstd_data, map->comp_level)) {
+		pr_debug2("failed to init mmap commpressor, error %d\n", errno);
+		return -1;
+	}
+#endif
 
 	if (map->comp_level && !perf_mmap__aio_enabled(map)) {
 		map->data = mmap(NULL, mmap__mmap_len(map), PROT_READ|PROT_WRITE,
tools/perf/util/mmap.h +3 -0
index 9d5f589f02ae..c4aed6e89549 100644
@@ -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 {
Signature · 1 line
2.19.0
#12 lore raw

PATCH v11 9/24 perf record: Introduce bytes written stats to support --max-size option

In reply to Alexey Bayduraev

Adding a function to calculate the total amount of data written and using it to support the --max-size option.

---
 tools/perf/builtin-record.c | 30 +++++++++++++++++++++++++++---
 tools/perf/util/mmap.h      |  1 +
 2 files changed, 28 insertions(+), 3 deletions(-)
tools/perf/builtin-record.c +27 -3
index eff6f8db60b2..cb155f1ba979 100644
@@ -186,10 +186,28 @@ static bool switch_output_time(struct record *rec)
 	       trigger_is_ready(&switch_output_trigger);
 }
 
+static u64 record__bytes_written(struct record *rec)
+{
+	int t, tm;
+	struct record_thread *thread_data = rec->thread_data;
+	u64 bytes_written = rec->bytes_written;
+
+	for (t = 0; t < rec->nr_threads; t++) {
+		for (tm = 0; tm < thread_data[t].nr_mmaps; tm++) {
+			if (thread_data[t].maps)
+				bytes_written += thread_data[t].maps[tm]->bytes_written;
+			if (thread_data[t].overwrite_maps)
+				bytes_written += thread_data[t].overwrite_maps[tm]->bytes_written;
+		}
+	}
+
+	return bytes_written;
+}
+
 static bool record__output_max_size_exceeded(struct record *rec)
 {
 	return rec->output_max_size &&
-	       (rec->bytes_written >= rec->output_max_size);
+	       (record__bytes_written(rec) >= rec->output_max_size);
 }
 
 static int record__write(struct record *rec, struct mmap *map __maybe_unused,
@@ -205,15 +223,21 @@ static int record__write(struct record *rec, struct mmap *map __maybe_unused,
 		return -1;
 	}
 
-	rec->bytes_written += size;
+	if (map && map->file)
+		map->bytes_written += size;
+	else
+		rec->bytes_written += size;
 
 	if (record__output_max_size_exceeded(rec) && !done) {
 		fprintf(stderr, "[ perf record: perf size limit reached (%" PRIu64 " KB),"
 				" stopping session ]\n",
-				rec->bytes_written >> 10);
+				record__bytes_written(rec) >> 10);
 		done = 1;
 	}
 
+	if (map && map->file)
+		return 0;
+
 	if (switch_output_size(rec))
 		trigger_hit(&switch_output_trigger);
 
tools/perf/util/mmap.h +1 -0
index c4aed6e89549..67d41003d82e 100644
@@ -46,6 +46,7 @@ struct mmap {
 	int		comp_level;
 	struct perf_data_file *file;
 	struct zstd_data      zstd_data;
+	u64		      bytes_written;
 };
 
 struct mmap_params {
Signature · 1 line
2.19.0
#13 lore raw

Re: [PATCH v11 09/24] perf record: Introduce bytes written stats to support --max-size option

Quoted 47 lines · Alexey Bayduraev

Adding a function to calculate the total amount of data written and using it to support the --max-size option.

---
 tools/perf/builtin-record.c | 30 +++++++++++++++++++++++++++---
 tools/perf/util/mmap.h      |  1 +
 2 files changed, 28 insertions(+), 3 deletions(-)
tools/perf/builtin-record.c +19 -1
index eff6f8db60b2..cb155f1ba979 100644
@@ -186,10 +186,28 @@ static bool switch_output_time(struct record *rec)
 	       trigger_is_ready(&switch_output_trigger);
 }
 
+static u64 record__bytes_written(struct record *rec)
+{
+	int t, tm;
+	struct record_thread *thread_data = rec->thread_data;
+	u64 bytes_written = rec->bytes_written;
+
+	for (t = 0; t < rec->nr_threads; t++) {
+		for (tm = 0; tm < thread_data[t].nr_mmaps; tm++) {
+			if (thread_data[t].maps)
+				bytes_written += thread_data[t].maps[tm]->bytes_written;
+			if (thread_data[t].overwrite_maps)
+				bytes_written += thread_data[t].overwrite_maps[tm]->bytes_written;
+		}
+	}
+
+	return bytes_written;
+}
+
 static bool record__output_max_size_exceeded(struct record *rec)
 {
 	return rec->output_max_size &&
-	       (rec->bytes_written >= rec->output_max_size);
+	       (record__bytes_written(rec) >= rec->output_max_size);
 }
 
 static int record__write(struct record *rec, struct mmap *map __maybe_unused,

tools/perf/builtin-record.c static int record__write(struct record *rec, struct mmap *map __maybe_unused, · line 227

Quoted 6 lines
tools/perf/builtin-record.c +2 -1
@@ -205,15 +223,21 @@ static int record__write(struct record *rec, struct mmap *map __maybe_unused,
 		return -1;
 	}
 
-	rec->bytes_written += size;
+	if (map && map->file)
+		map->bytes_written += size;

could we instead have bytes_written in thread data? so we don't need to iterate all the maps?

tools/perf/builtin-record.c static int record__write(struct record *rec, struct mmap *map __maybe_unused, · line 234

Quoted 8 lines
tools/perf/builtin-record.c +3 -1
+	else
+		rec->bytes_written += size;
 
 	if (record__output_max_size_exceeded(rec) && !done) {
 		fprintf(stderr, "[ perf record: perf size limit reached (%" PRIu64 " KB),"
 				" stopping session ]\n",
-				rec->bytes_written >> 10);
+				record__bytes_written(rec) >> 10);

you're calling record__bytes_written twice.. could we just save the bytes_written from the first call and use it in the printf?

tools/perf/builtin-record.c static int record__write(struct record *rec, struct mmap *map __maybe_unused, · line 239

Quoted 5 lines
tools/perf/builtin-record.c +2 -0
 		done = 1;
 	}
 
+	if (map && map->file)
+		return 0;

please make comment why quit in here, we don't support switch-output for threads?

jirka

Quoted 19 lines
tools/perf/builtin-record.c +1 -0
+
 	if (switch_output_size(rec))
 		trigger_hit(&switch_output_trigger);
 
tools/perf/util/mmap.h +1 -0
index c4aed6e89549..67d41003d82e 100644
@@ -46,6 +46,7 @@ struct mmap {
 	int		comp_level;
 	struct perf_data_file *file;
 	struct zstd_data      zstd_data;
+	u64		      bytes_written;
 };
 
 struct mmap_params {
Signature · 1 line
2.19.0
#14 lore raw

Re: [PATCH v11 09/24] perf record: Introduce bytes written stats to support --max-size option

Quoted 1 line · Jiri Olsa

On Tue, Aug 17, 2021 at 11:23:12AM +0300, Alexey Bayduraev wrote:

<SNIP>

Quoted 12 lines
Quoted 12 lines

static bool record__output_max_size_exceeded(struct record *rec) {

tools/perf/builtin-record.c +1 -1
 	return rec->output_max_size &&
-	       (rec->bytes_written >= rec->output_max_size);
+	       (record__bytes_written(rec) >= rec->output_max_size);
 }
 
 static int record__write(struct record *rec, struct mmap *map __maybe_unused,
@@ -205,15 +223,21 @@ static int record__write(struct record *rec, struct mmap *map __maybe_unused,
 		return -1;
 	}
 
Quoted 6 lines

tools/perf/builtin-record.c static int record__write(struct record *rec, struct mmap *map __maybe_unused, · line 227

Quoted 3 lines
tools/perf/builtin-record.c +2 -1
-	rec->bytes_written += size;
+	if (map && map->file)
+		map->bytes_written += size;

could we instead have bytes_written in thread data? so we don't need to iterate all the maps?

Hi,

As I remember the main issue is that bytes_written should be atomic64_t. Unfortunately we don't have atomic64 framework in tools/lib (even atomic32_add is missing). Thus I decided to calculate total size on each iteration. But I think your suggestion to move record__output_max_size_exceeded to trigger framework is better.

Quoted 13 lines

tools/perf/builtin-record.c static int record__write(struct record *rec, struct mmap *map __maybe_unused, · line 234

Quoted 9 lines
tools/perf/builtin-record.c +3 -1
+	else
+		rec->bytes_written += size;
 
 	if (record__output_max_size_exceeded(rec) && !done) {
 		fprintf(stderr, "[ perf record: perf size limit reached (%" PRIu64 " KB),"
 				" stopping session ]\n",
-				rec->bytes_written >> 10);
+				record__bytes_written(rec) >> 10);

you're calling record__bytes_written twice.. could we just save the bytes_written from the first call and use it in the printf?

Quoted 8 lines

tools/perf/builtin-record.c static int record__write(struct record *rec, struct mmap *map __maybe_unused, · line 239

Quoted 5 lines
tools/perf/builtin-record.c +2 -0
 		done = 1;
 	}
 
+	if (map && map->file)
+		return 0;

please make comment why quit in here, we don't support switch-output for threads?

Yes, parallel streaming mode doesn't support switch-output and there is a special warning in [PATCH v11 14/24]

Thanks,
Alexey

Quoted 23 lines

jirka

Quoted 19 lines
tools/perf/builtin-record.c +1 -0
+
 	if (switch_output_size(rec))
 		trigger_hit(&switch_output_trigger);
 
tools/perf/util/mmap.h +1 -0
index c4aed6e89549..67d41003d82e 100644
@@ -46,6 +46,7 @@ struct mmap {
 	int		comp_level;
 	struct perf_data_file *file;
 	struct zstd_data      zstd_data;
+	u64		      bytes_written;
 };
 
 struct mmap_params {
Signature · 1 line
2.19.0
#15 lore raw

Re: [PATCH v11 09/24] perf record: Introduce bytes written stats to support --max-size option

In reply to Alexey Bayduraev
Quoted 47 lines · Alexey Bayduraev

Adding a function to calculate the total amount of data written and using it to support the --max-size option.

---
 tools/perf/builtin-record.c | 30 +++++++++++++++++++++++++++---
 tools/perf/util/mmap.h      |  1 +
 2 files changed, 28 insertions(+), 3 deletions(-)
tools/perf/builtin-record.c +19 -1
index eff6f8db60b2..cb155f1ba979 100644
@@ -186,10 +186,28 @@ static bool switch_output_time(struct record *rec)
 	       trigger_is_ready(&switch_output_trigger);
 }
 
+static u64 record__bytes_written(struct record *rec)
+{
+	int t, tm;
+	struct record_thread *thread_data = rec->thread_data;
+	u64 bytes_written = rec->bytes_written;
+
+	for (t = 0; t < rec->nr_threads; t++) {
+		for (tm = 0; tm < thread_data[t].nr_mmaps; tm++) {
+			if (thread_data[t].maps)
+				bytes_written += thread_data[t].maps[tm]->bytes_written;
+			if (thread_data[t].overwrite_maps)
+				bytes_written += thread_data[t].overwrite_maps[tm]->bytes_written;
+		}
+	}
+
+	return bytes_written;
+}
+
 static bool record__output_max_size_exceeded(struct record *rec)
 {
 	return rec->output_max_size &&
-	       (rec->bytes_written >= rec->output_max_size);
+	       (record__bytes_written(rec) >= rec->output_max_size);
 }
 
 static int record__write(struct record *rec, struct mmap *map __maybe_unused,

tools/perf/builtin-record.c static int record__write(struct record *rec, struct mmap *map __maybe_unused, · line 231

Quoted 10 lines
tools/perf/builtin-record.c +4 -1
@@ -205,15 +223,21 @@ static int record__write(struct record *rec, struct mmap *map __maybe_unused,
 		return -1;
 	}
 
-	rec->bytes_written += size;
+	if (map && map->file)
+		map->bytes_written += size;
+	else
+		rec->bytes_written += size;
 
 	if (record__output_max_size_exceeded(rec) && !done) {

also instead of slowing down each write with calling record__output_max_size_exceeded and risking the race with done, I think this should be called as part of trigger framework

we could call this check periodically every second or such, we don't mind the size overhead.. IMO it's still better than caling this each time we write to the file

jirka

Quoted 28 lines
tools/perf/builtin-record.c +4 -1
 		fprintf(stderr, "[ perf record: perf size limit reached (%" PRIu64 " KB),"
 				" stopping session ]\n",
-				rec->bytes_written >> 10);
+				record__bytes_written(rec) >> 10);
 		done = 1;
 	}
 
+	if (map && map->file)
+		return 0;
+
 	if (switch_output_size(rec))
 		trigger_hit(&switch_output_trigger);
 
tools/perf/util/mmap.h +1 -0
index c4aed6e89549..67d41003d82e 100644
@@ -46,6 +46,7 @@ struct mmap {
 	int		comp_level;
 	struct perf_data_file *file;
 	struct zstd_data      zstd_data;
+	u64		      bytes_written;
 };
 
 struct mmap_params {
Signature · 1 line
2.19.0
#16 lore raw

PATCH v11 10/24 perf record: Introduce data transferred and compressed stats

In reply to Alexey Bayduraev

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

---
 tools/perf/builtin-record.c | 68 +++++++++++++++++++++++++++++--------
 tools/perf/util/mmap.h      |  2 ++
 2 files changed, 56 insertions(+), 14 deletions(-)
tools/perf/builtin-record.c +54 -14
index cb155f1ba979..d833095e73cb 100644
@@ -246,8 +246,8 @@ 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,
@@ -381,9 +381,13 @@ 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);
 	}
@@ -608,8 +612,22 @@ 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++;
@@ -1396,18 +1414,15 @@ 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;
 }
 
@@ -2063,8 +2078,10 @@ 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 record_thread *thread_data = rec->thread_data;
+	u64 bytes_written, bytes_transferred, bytes_compressed;
 
 	for (t = 1; t < rec->nr_threads; t++)
 		record__terminate_thread(&thread_data[t]);
@@ -2072,9 +2089,32 @@ 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",
-			 thread_data[t].tid, thread_data[t].samples, thread_data[t].waking,
-			 rec->session->bytes_transferred, rec->session->bytes_compressed);
+		bytes_transferred = 0;
+		bytes_compressed = 0;
+		bytes_written = t ? 0 : rec->bytes_written;
+		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, ",
+			 thread_data[t].tid, thread_data[t].samples, thread_data[t].waking);
+		if (bytes_compressed)
+			pr_debug("trasferred=%ld, compressed=%ld\n",
+				 bytes_transferred, bytes_compressed);
+		else
+			pr_debug("written=%ld\n", bytes_written);
 	}
 
 	return 0;
tools/perf/util/mmap.h +2 -0
index 67d41003d82e..f5e8c8a24e68 100644
@@ -47,6 +47,8 @@ struct mmap {
 	struct perf_data_file *file;
 	struct zstd_data      zstd_data;
 	u64		      bytes_written;
+	u64		      bytes_transferred;
+	u64		      bytes_compressed;
 };
 
 struct mmap_params {
Signature · 1 line
2.19.0
#17 lore raw

PATCH v11 11/24 perf record: Init data file at mmap buffer object

In reply to Alexey Bayduraev

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

---
 tools/perf/builtin-record.c | 41 ++++++++++++++++++++++++++++++-------
 tools/perf/util/record.h    |  1 +
 2 files changed, 35 insertions(+), 7 deletions(-)
tools/perf/builtin-record.c +34 -7
index d833095e73cb..06160627f975 100644
@@ -167,6 +167,11 @@ static inline pid_t gettid(void)
 }
 #endif
 
+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 &&
@@ -1147,7 +1152,7 @@ static int record__alloc_thread_data(struct record *rec, struct evlist *evlist)
 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;
@@ -1186,6 +1191,18 @@ 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;
 }
 
@@ -1496,8 +1513,12 @@ 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)
@@ -1613,7 +1634,9 @@ 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);
 
@@ -1624,15 +1647,21 @@ 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)
@@ -2548,8 +2577,6 @@ 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) {
@@ -3350,7 +3377,7 @@ int cmd_record(int argc, const char **argv)
 		goto out_opts;
 	}
 
-	if (rec->opts.kcore)
+	if (rec->opts.kcore || record__threads_enabled(rec))
 		rec->data.is_dir = true;
 
 	if (rec->opts.comp_level != 0) {
tools/perf/util/record.h +1 -0
index 68f471d9a88b..4d68b7e27272 100644
@@ -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;
Signature · 1 line
2.19.0
#18 lore raw

PATCH v11 12/24 perf record: Introduce --threads command line option

In reply to Alexey Bayduraev

Provide --threads option in perf record command line interface. The option creates a data streaming thread for each cpu in the system. Document --threads option in Documentation/perf-record.txt.

---
 tools/perf/Documentation/perf-record.txt |  4 +++
 tools/perf/builtin-record.c              | 44 +++++++++++++++++++++++-
 2 files changed, 47 insertions(+), 1 deletion(-)
tools/perf/Documentation/perf-record.txt +4 -0
index f1079ee7f2ec..0408e677c117 100644
@@ -695,6 +695,10 @@ measurements:
  wait -n ${perf_pid}
  exit $?
 
+--threads::
+Write collected trace data into several data files using parallel threads.
+The option creates a data streaming thread for each cpu in the system.
+
 include::intel-hybrid.txt[]
 
 SEE ALSO
tools/perf/builtin-record.c +43 -1

74 lines not on this page · raw message

Signature · 1 line
2.19.0
#19 lore raw

PATCH v11 13/24 perf record: Extend --threads command line option

In reply to Alexey Bayduraev

Extend --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.

Document --threads option syntax and parallel data streaming modes in Documentation/perf-record.txt.

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/[email protected]/

---
 tools/perf/Documentation/perf-record.txt |  30 ++-
 tools/perf/builtin-record.c              | 314 ++++++++++++++++++++++-
 tools/perf/util/record.h                 |   1 +
 3 files changed, 340 insertions(+), 5 deletions(-)
tools/perf/Documentation/perf-record.txt +28 -2
index 0408e677c117..4bbdc813e2e9 100644
@@ -695,9 +695,35 @@ measurements:
  wait -n ${perf_pid}
  exit $?
 
---threads::
+--threads=<spec>::
 Write collected trace data into several data files using parallel threads.
-The option creates a data streaming thread for each cpu in the system.
+<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[]
 
tools/perf/builtin-record.c +311 -3

366 lines not on this page · raw message

tools/perf/util/record.h +1 -0
index 4d68b7e27272..3da156498f47 100644
@@ -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;
Signature · 1 line
2.19.0
#20 lore raw

Re: [PATCH v11 13/24] perf record: Extend --threads command line option

Quoted 54 lines · Alexey Bayduraev

Extend --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.

Document --threads option syntax and parallel data streaming modes in Documentation/perf-record.txt.

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/[email protected]/

---
 tools/perf/Documentation/perf-record.txt |  30 ++-
 tools/perf/builtin-record.c              | 314 ++++++++++++++++++++++-
 tools/perf/util/record.h                 |   1 +
 3 files changed, 340 insertions(+), 5 deletions(-)
tools/perf/Documentation/perf-record.txt +1 -1
index 0408e677c117..4bbdc813e2e9 100644
@@ -695,9 +695,35 @@ measurements:
  wait -n ${perf_pid}
  exit $?
 
---threads::
+--threads=<spec>::
 Write collected trace data into several data files using parallel threads.

tools/perf/Documentation/perf-record.txt measurements: · line 708

Quoted 10 lines
tools/perf/Documentation/perf-record.txt +9 -1
-The option creates a data streaming thread for each cpu in the system.
+<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

is this special case?

[root@krava perf]# ./perf record --threads=1/1

[root@krava perf]# ps -ef | grep perf
root 132071 130979 9 22:59 pts/6 00:00:00 ./perf record --threads=1/1

[root@krava perf]# cat /proc/131798/task/*/status | grep Cpus_allowed
cat: '/proc/131798/task/*/status': No such file or directory [root@krava perf]# cat /proc/132071/task/*/status | grep Cpus_allowed

Cpus_allowed:   ff
Cpus_allowed_list:      0-7
Cpus_allowed:   ff
Cpus_allowed_list:      0-7

looks like the affinity was not set

thanks,
jirka

#21 lore raw

PATCH v11 14/24 perf record: Implement compatibility checks

In reply to Alexey Bayduraev

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.

---
 tools/perf/builtin-record.c | 49 ++++++++++++++++++++++++++++++++++---
 1 file changed, 46 insertions(+), 3 deletions(-)
tools/perf/builtin-record.c +46 -3
index 8d93b4fe52f4..63f50f12011f 100644
@@ -826,6 +826,12 @@ 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)
@@ -2213,6 +2219,17 @@ 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;
 
@@ -2980,12 +2997,22 @@ 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;
@@ -3283,8 +3310,8 @@ 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,
@@ -3725,6 +3752,17 @@ 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;
@@ -3758,6 +3796,11 @@ 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().
Signature · 1 line
2.19.0
#22 lore raw

PATCH v11 15/24 perf report: Output non-zero offset for decompressed records

In reply to Alexey Bayduraev

Print offset of PERF_RECORD_COMPRESSED record instead of zero for decompressed records in raw trace dump (-D option of perf-report):

0x17cf08 [0x28]: event: 9

instead of:

0 [0x28]: event: 9

---
 tools/perf/util/session.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
tools/perf/util/session.c +2 -2
index d2e27ff96030..0f4471c361e4 100644
@@ -2107,7 +2107,7 @@ 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)
@@ -2123,7 +2123,7 @@ static int __perf_session__process_decomp_events(struct perf_session *session)
 		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)) < 0) {
 			pr_err("%#" PRIx64 " [%#x]: failed to process type: %d\n",
 				decomp->file_pos + decomp->head, event->header.size, event->header.type);
 			return -EINVAL;
Signature · 1 line
2.19.0
#23 lore raw

PATCH v11 16/24 perf report: Output data file name in raw trace dump

In reply to Alexey Bayduraev

Print path and name of a data file into raw dump (-D) <file_offset>@<path/file>:

[email protected] [0x30]: event: 9
or
[email protected]/data.7 [0x30]: event: 9

---
 tools/perf/builtin-inject.c      |  3 +-
 tools/perf/builtin-kvm.c         |  2 +-
 tools/perf/builtin-top.c         |  2 +-
 tools/perf/builtin-trace.c       |  2 +-
 tools/perf/util/ordered-events.c |  3 +-
 tools/perf/util/ordered-events.h |  3 +-
 tools/perf/util/session.c        | 78 +++++++++++++++++++-------------
 tools/perf/util/session.h        |  3 +-
 tools/perf/util/tool.h           |  3 +-
 9 files changed, 59 insertions(+), 40 deletions(-)
tools/perf/builtin-inject.c +2 -1
index 6ad191e731fc..6ecb13a95191 100644
@@ -110,7 +110,8 @@ 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);
 }
tools/perf/builtin-kvm.c +1 -1
index aa1b127ffb5b..240184e64188 100644
@@ -771,7 +771,7 @@ static s64 perf_kvm__mmap_read_idx(struct perf_kvm_stat *kvm, int idx,
 			return -1;
 		}
 
-		err = perf_session__queue_event(kvm->session, event, timestamp, 0);
+		err = perf_session__queue_event(kvm->session, event, timestamp, 0, NULL);
 		/*
 		 * FIXME: Here we can't consume the event, as perf_session__queue_event will
 		 *        point to it, and it'll get possibly overwritten by the kernel.
tools/perf/builtin-top.c +1 -1
index a3ae9176a83e..1733fee10a61 100644
@@ -888,7 +888,7 @@ static void perf_top__mmap_read_idx(struct perf_top *top, int idx)
 		if (ret && ret != -1)
 			break;
 
-		ret = ordered_events__queue(top->qe.in, event, last_timestamp, 0);
+		ret = ordered_events__queue(top->qe.in, event, last_timestamp, 0, NULL);
 		if (ret)
 			break;
 
tools/perf/builtin-trace.c +1 -1
index 2bf21194c7b3..b8fda446cec3 100644
@@ -3769,7 +3769,7 @@ static int trace__deliver_event(struct trace *trace, union perf_event *event)
 	if (err && err != -1)
 		return err;
 
-	err = ordered_events__queue(&trace->oe.data, event, trace->oe.last, 0);
+	err = ordered_events__queue(&trace->oe.data, event, trace->oe.last, 0, NULL);
 	if (err)
 		return err;
 
tools/perf/util/ordered-events.c +2 -1
index 48c8f609441b..b887dfeea673 100644
@@ -192,7 +192,7 @@ void ordered_events__delete(struct ordered_events *oe, struct ordered_event *eve
 }
 
 int ordered_events__queue(struct ordered_events *oe, union perf_event *event,
-			  u64 timestamp, u64 file_offset)
+			  u64 timestamp, u64 file_offset, const char *file_path)
 {
 	struct ordered_event *oevent;
 
@@ -217,6 +217,7 @@ int ordered_events__queue(struct ordered_events *oe, union perf_event *event,
 		return -ENOMEM;
 
 	oevent->file_offset = file_offset;
+	oevent->file_path = file_path;
 	return 0;
 }
 
tools/perf/util/ordered-events.h +2 -1
index 75345946c4b9..0b05c3c0aeaa 100644
@@ -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;
 };
@@ -53,7 +54,7 @@ struct ordered_events {
 };
 
 int ordered_events__queue(struct ordered_events *oe, union perf_event *event,
-			  u64 timestamp, u64 file_offset);
+			  u64 timestamp, u64 file_offset, const char *file_path);
 void ordered_events__delete(struct ordered_events *oe, struct ordered_event *event);
 int ordered_events__flush(struct ordered_events *oe, enum oe_flush how);
 int ordered_events__flush_time(struct ordered_events *oe, u64 timestamp);
tools/perf/util/session.c +46 -32

236 lines not on this page · raw message

tools/perf/util/session.h +2 -1
index 5d8bd14a0a39..308cf48e0945 100644
@@ -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;
@@ -81,7 +82,7 @@ int perf_session__peek_events(struct perf_session *session, u64 offset,
 int perf_session__process_events(struct perf_session *session);
 
 int perf_session__queue_event(struct perf_session *s, union perf_event *event,
-			      u64 timestamp, u64 file_offset);
+			      u64 timestamp, u64 file_offset, const char *file_path);
 
 void perf_tool__fill_defaults(struct perf_tool *tool);
 
tools/perf/util/tool.h +2 -1
index bbbc0dcd461f..c966531d3eca 100644
@@ -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);
Signature · 1 line
2.19.0
#24 lore raw

PATCH v11 17/24 perf session: Move reader structure to the top

In reply to Alexey Bayduraev

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

---
 tools/perf/util/session.c | 56 +++++++++++++++++++--------------------
 1 file changed, 28 insertions(+), 28 deletions(-)
tools/perf/util/session.c +28 -28
index 207edce176c4..fe7d986cb094 100644
@@ -36,6 +36,34 @@
 #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;
+	bool		 in_place_update;
+};
+
 #ifdef HAVE_ZSTD_SUPPORT
 static int perf_session__process_compressed_event(struct perf_session *session,
 						  union perf_event *event, u64 file_offset,
@@ -2148,34 +2176,6 @@ 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;
-	bool		 in_place_update;
-};
-
 static int
 reader__process_events(struct reader *rd, struct perf_session *session,
 		       struct ui_progress *prog)
Signature · 1 line
2.19.0
#25 lore raw

PATCH v11 18/24 perf session: Introduce reader_state in reader object

In reply to 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/[email protected]/

---
 tools/perf/util/session.c | 74 +++++++++++++++++++++++----------------
 1 file changed, 43 insertions(+), 31 deletions(-)
tools/perf/util/session.c +43 -31
index fe7d986cb094..a82204d4c4d6 100644
@@ -55,6 +55,17 @@ 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;
@@ -62,6 +73,7 @@ struct reader {
 	u64		 data_offset;
 	reader_cb_t	 process;
 	bool		 in_place_update;
+	struct reader_state state;
 };
 
 #ifdef HAVE_ZSTD_SUPPORT
@@ -2180,29 +2192,28 @@ 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;
@@ -2214,35 +2225,36 @@ 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;
 	}
 
@@ -2251,9 +2263,9 @@ 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;
@@ -2262,8 +2274,8 @@ 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)
@@ -2274,7 +2286,7 @@ 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:
Signature · 1 line
2.19.0
#26 lore raw

PATCH v11 19/24 perf session: Introduce reader objects in session object

In reply to 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/[email protected]/

---
 tools/perf/util/session.c | 33 +++++++++++++++++++++------------
 tools/perf/util/session.h |  3 +++
 2 files changed, 24 insertions(+), 12 deletions(-)
tools/perf/util/session.c +21 -12
index a82204d4c4d6..1f92d3cfcb1d 100644
@@ -343,6 +343,10 @@ 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);
@@ -2303,14 +2307,7 @@ 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,
-		.in_place_update = session->data->in_place_update,
-	};
+	struct reader *rd;
 	struct ordered_events *oe = &session->ordered_events;
 	struct perf_tool *tool = session->tool;
 	struct ui_progress prog;
@@ -2318,12 +2315,24 @@ 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,
+		.in_place_update = session->data->in_place_update,
+	};
 
-	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 */
tools/perf/util/session.h +3 -0
index 308cf48e0945..b9b9468d8f17 100644
@@ -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 {
Signature · 1 line
2.19.0
#27 lore raw

Re: [PATCH v11 19/24] perf session: Introduce reader objects in session object

Quoted 67 lines · 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/[email protected]/

---
 tools/perf/util/session.c | 33 +++++++++++++++++++++------------
 tools/perf/util/session.h |  3 +++
 2 files changed, 24 insertions(+), 12 deletions(-)
tools/perf/util/session.c +17 -10
index a82204d4c4d6..1f92d3cfcb1d 100644
@@ -343,6 +343,10 @@ 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);
@@ -2303,14 +2307,7 @@ 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,
-		.in_place_update = session->data->in_place_update,
-	};
+	struct reader *rd;
 	struct ordered_events *oe = &session->ordered_events;
 	struct perf_tool *tool = session->tool;
 	struct ui_progress prog;
@@ -2318,12 +2315,24 @@ 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,

tools/perf/util/session.c static int __perf_session__process_events(struct perf_session *session) · line 2337

Quoted 10 lines
tools/perf/util/session.c +4 -2
+		.in_place_update = session->data->in_place_update,
+	};
 
-	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;

hm, leaking rd in here?

jirka

Quoted 25 lines
tools/perf/util/session.c +0 -0
 	/* do the final flush for ordered samples */
tools/perf/util/session.h +3 -0
index 308cf48e0945..b9b9468d8f17 100644
@@ -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 {
Signature · 1 line
2.19.0
#28 lore raw

PATCH v11 20/24 perf session: Introduce decompressor into trace reader object

In reply to 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.

---
 tools/perf/util/session.c | 54 +++++++++++++++++++++++++++++----------
 tools/perf/util/session.h |  1 +
 2 files changed, 42 insertions(+), 13 deletions(-)
tools/perf/util/session.c +41 -13
index 1f92d3cfcb1d..ed41bccb15ff 100644
@@ -73,6 +73,9 @@ struct reader {
 	u64		 data_offset;
 	reader_cb_t	 process;
 	bool		 in_place_update;
+	struct zstd_data zstd_data;
+	struct decomp	 *decomp;
+	struct decomp	 *decomp_last;
 	struct reader_state state;
 };
 
@@ -85,7 +88,10 @@ 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;
@@ -113,7 +119,7 @@ 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);
@@ -123,12 +129,22 @@ 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);
@@ -320,11 +336,11 @@ 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)
@@ -337,6 +353,8 @@ 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);
@@ -344,10 +362,14 @@ 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);
+			zstd_fini(&session->readers[r].zstd_data);
+		}
 		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) {
@@ -2162,7 +2184,8 @@ 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;
@@ -2219,6 +2242,9 @@ 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;
 
@@ -2262,12 +2288,13 @@ 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));
@@ -2294,6 +2321,7 @@ reader__process_events(struct reader *rd, struct perf_session *session,
 		goto more;
 
 out:
+	session->active_reader = NULL;
 	return err;
 }
 
tools/perf/util/session.h +1 -0
index b9b9468d8f17..8073331d0490 100644
@@ -44,6 +44,7 @@ struct perf_session {
 	struct decomp		*decomp_last;
 	struct reader		*readers;
 	int			nr_readers;
+	struct reader		*active_reader;
 };
 
 struct decomp {
Signature · 1 line
2.19.0
#29 lore raw

PATCH v11 21/24 perf session: Move init into reader__init function

In reply to 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/[email protected]/

---
 tools/perf/util/session.c | 34 ++++++++++++++++++++++++----------
 1 file changed, 24 insertions(+), 10 deletions(-)
tools/perf/util/session.c +24 -10
index ed41bccb15ff..3cb1691483f3 100644
@@ -2216,28 +2216,25 @@ 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));
@@ -2245,6 +2242,20 @@ 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;
 
@@ -2360,6 +2371,9 @@ 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;
Signature · 1 line
2.19.0
#30 lore raw

PATCH v11 22/24 perf session: Move map/unmap into reader__mmap function

In reply to 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/[email protected]/

---
 tools/perf/util/session.c | 60 +++++++++++++++++++++++----------------
 1 file changed, 35 insertions(+), 25 deletions(-)
tools/perf/util/session.c +35 -25
index 3cb1691483f3..e9a7cba31194 100644
@@ -2219,14 +2219,11 @@ 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;
 
@@ -2246,15 +2243,12 @@ 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;
@@ -2265,20 +2259,45 @@ 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:
@@ -2287,17 +2306,8 @@ 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;
Signature · 1 line
2.19.0
#31 lore raw

PATCH v11 23/24 perf session: Load single file for analysis

In reply to 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/[email protected]/

---
 tools/perf/util/session.c | 72 ++++++++++++++++++++++++---------------
 1 file changed, 45 insertions(+), 27 deletions(-)
tools/perf/util/session.c +45 -27
index e9a7cba31194..b49b52768681 100644
@@ -64,6 +64,13 @@ struct reader_state {
 	u64	 file_offset;
 	u64	 data_size;
 	u64	 head;
+	bool	 eof;
+};
+
+enum {
+	READER_EOF,
+	READER_NODATA,
+	READER_OK,
 };
 
 struct reader {
@@ -2250,6 +2257,11 @@ 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;
 
@@ -2278,36 +2290,26 @@ 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_NODATA;
 
 	session->active_reader = rd;
 	size = event->header.size;
@@ -2329,18 +2331,12 @@ 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;
@@ -2384,9 +2380,31 @@ 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_NODATA) {
+			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)
Signature · 1 line
2.19.0
#32 lore raw

PATCH v11 24/24 perf session: Load data directory files for analysis

In reply to 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/[email protected]/

---
 tools/perf/util/session.c | 131 ++++++++++++++++++++++++++++++++++++++
 1 file changed, 131 insertions(+)
tools/perf/util/session.c +131 -0
index b49b52768681..f5a82f66e36b 100644
@@ -48,6 +48,13 @@
 #define NUM_MMAPS 128
 #endif
 
+/*
+ * Processing 10MBs of data from each reader in sequence,
+ * because that's the way the ordered events sorting works
+ * most efficiently.
+ */
+#define READER_MAX_SIZE (10 * 1024 * 1024)
+
 struct reader;
 
 typedef s64 (*reader_cb_t)(struct perf_session *session,
@@ -65,6 +72,7 @@ struct reader_state {
 	u64	 data_size;
 	u64	 head;
 	bool	 eof;
+	u64	 size;
 };
 
 enum {
@@ -2328,6 +2336,7 @@ reader__read_event(struct reader *rd, struct perf_session *session,
 	if (skip)
 		size += skip;
 
+	st->size += size;
 	st->head += size;
 	st->file_pos += size;
 
@@ -2427,6 +2436,125 @@ 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,
+		.in_place_update = session->data->in_place_update,
+	};
+	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)
+			continue;
+		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,
+			.in_place_update = session->data->in_place_update,
+		};
+		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_NODATA) {
+			ret = reader__mmap(&rd[i], session);
+			if (ret < 0)
+				goto out_err;
+			if (ret == READER_EOF)
+				readers--;
+		}
+
+		if (rd[i].state.size >= READER_MAX_SIZE) {
+			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)
@@ -2435,6 +2563,9 @@ 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);
 }
 
Signature · 1 line
2.19.0
#33 lore raw

Re: [PATCH v11 24/24] perf session: Load data directory files for analysis

Quoted 23 lines · 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/[email protected]/

---
 tools/perf/util/session.c | 131 ++++++++++++++++++++++++++++++++++++++
 1 file changed, 131 insertions(+)
tools/perf/util/session.c +0 -0
index b49b52768681..f5a82f66e36b 100644

tools/perf/util/session.c line 56

Quoted 9 lines
tools/perf/util/session.c +6 -0
@@ -48,6 +48,13 @@
 #define NUM_MMAPS 128
 #endif
 
+/*
+ * Processing 10MBs of data from each reader in sequence,
+ * because that's the way the ordered events sorting works
+ * most efficiently.
+ */
+#define READER_MAX_SIZE (10 * 1024 * 1024)

why 10? ;-) I think we discussed this in your earlier iteration, I put the 10M there to speedup the processing, because it worked on my current data file for the test at that time

I think this needs some better justification and some setup based on the env setup and data size perhaps?

jirka

Quoted 159 lines
tools/perf/util/session.c +125 -0
+
 struct reader;
 
 typedef s64 (*reader_cb_t)(struct perf_session *session,
@@ -65,6 +72,7 @@ struct reader_state {
 	u64	 data_size;
 	u64	 head;
 	bool	 eof;
+	u64	 size;
 };
 
 enum {
@@ -2328,6 +2336,7 @@ reader__read_event(struct reader *rd, struct perf_session *session,
 	if (skip)
 		size += skip;
 
+	st->size += size;
 	st->head += size;
 	st->file_pos += size;
 
@@ -2427,6 +2436,125 @@ 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,
+		.in_place_update = session->data->in_place_update,
+	};
+	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)
+			continue;
+		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,
+			.in_place_update = session->data->in_place_update,
+		};
+		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_NODATA) {
+			ret = reader__mmap(&rd[i], session);
+			if (ret < 0)
+				goto out_err;
+			if (ret == READER_EOF)
+				readers--;
+		}
+
+		if (rd[i].state.size >= READER_MAX_SIZE) {
+			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)
@@ -2435,6 +2563,9 @@ 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);
 }
 
Signature · 1 line
2.19.0
#34 lore raw
In reply to Alexey Bayduraev

On Tue, Aug 17, 2021 at 11:23:03AM +0300, Alexey Bayduraev wrote:

SNIP

Quoted 11 lines

Alexey Bayduraev (24):
perf record: Introduce thread affinity and mmap masks tools lib: Introduce fdarray duplicate function
perf record: Introduce thread specific data array perf record: Introduce function to propagate control commands 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 bytes written stats to support --max-size

Quoted 8 lines

option
perf record: Introduce data transferred and compressed stats perf record: Init data file at mmap buffer object perf record: Introduce --threads command line option perf record: Extend --threads command line option perf record: Implement compatibility checks
perf report: Output non-zero offset for decompressed records perf report: Output data file name in raw trace dump

following patches:

Quoted 7 lines

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

up to here

are changing the core of processing perf.data files, I think they should be submitted in separate patchset first and some tests for that would be great

jirka