perf record: Add --tail-synthesize option

When working with overwritable ring buffer there's a inconvenience
problem: if perf dumps data after a long period after it starts,
non-sample events may lost, which makes following 'perf report' unable
to identify proc name and mmap layout. For example:

 # perf record -m 4 -e raw_syscalls:* -g --overwrite --switch-output \
        dd if=/dev/zero of=/dev/null

send SIGUSR2 after dd runs long enough. The resuling perf.data lost
correct comm and mmap events:

 # perf script -i perf.data.2016061522374354
 perf 24478 [004] 2581325.601789:  raw_syscalls:sys_exit: NR 0 = 512
 ^^^^
 Should be 'dd'
                   27b2e8 syscall_slow_exit_work+0xfe2000e3 (/lib/modules/4.6.0-rc3+/build/vmlinux)
                   203cc7 do_syscall_64+0xfe200117 (/lib/modules/4.6.0-rc3+/build/vmlinux)
                   b18d83 return_from_SYSCALL_64+0xfe200000 (/lib/modules/4.6.0-rc3+/build/vmlinux)
             7f47c417edf0 [unknown] ([unknown])
             ^^^^^^^^^^^^
             Fail to unwind

This patch provides a '--tail-synthesize' option, allows perf to collect
system status when finalizing output file. In resuling output file, the
non-sample events reflect system status when dumping data.

After this patch:
 # perf record -m 4 -e raw_syscalls:* -g --overwrite --switch-output --tail-synthesize \
        dd if=/dev/zero of=/dev/null

 # perf script -i perf.data.2016061600544998
 dd 27364 [004] 2583244.994464: raw_syscalls:sys_enter: NR 1 (1, ...
 ^^
 Correct comm
                   203a18 syscall_trace_enter_phase2+0xfe2001a8 ([kernel.kallsyms])
                   203aa5 syscall_trace_enter+0xfe200055 ([kernel.kallsyms])
                   203caa do_syscall_64+0xfe2000fa ([kernel.kallsyms])
                   b18d83 return_from_SYSCALL_64+0xfe200000 ([kernel.kallsyms])
                    d8e50 __GI___libc_write+0xffff01d9639f4010 (/tmp/oxygen_root-w00229757/lib64/libc-2.18.so)
                    ^^^^^
                    Correct unwind

This option doesn't aim to solve this problem completely. If a process
terminates before SIGUSR2, we still lost its COMM and MMAP events. For
example, we can't unwind correctly from the final perf.data we get from
the previous example, because when perf collects the final output file
(when we press C-c), 'dd' has been terminated so its '/proc/<pid>/mmap'
becomes empty.

However, this is a cheaper choice. To completely solve this problem we
need to continously output non-sample events. To satisify the
requirement of daemonization, we need to merge them periodically. It is
possible but requires much more code and cycles.

Automatically select --tail-synthesize when --overwrite is provided.

Signed-off-by: Wang Nan <wangnan0@huawei.com>
Cc: He Kuang <hekuang@huawei.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Nilay Vaish <nilayvaish@gmail.com>
Cc: Zefan Li <lizefan@huawei.com>
Cc: pi3orama@163.com
Link: http://lkml.kernel.org/r/1468485287-33422-16-git-send-email-wangnan0@huawei.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>

3 files changed, 34 insertions, 6 deletions

diff --git a/tools/perf/Documentation/perf-record.txt b/tools/perf/Documentation/perf-record.txt
index 384c630436f8..69966abf65d1 100644
--- a/tools/perf/Documentation/perf-record.txt
+++ b/tools/perf/Documentation/perf-record.txt
@@ -367,6 +367,12 @@ options.
 'perf record --dry-run -e' can act as a BPF script compiler if llvm.dump-obj
 in config file is set to true.
 
+--tail-synthesize::
+Instead of collecting non-sample events (for example, fork, comm, mmap) at
+the beginning of record, collect them during finalizing an output file.
+The collected non-sample events reflects the status of the system when
+record is finished.
+
 --overwrite::
 Makes all events use an overwritable ring buffer. An overwritable ring
 buffer works like a flight recorder: when it gets full, the kernel will
@@ -381,6 +387,8 @@ those fitting in the ring buffer at that moment.
 'overwrite' attribute can also be set or canceled for an event using
 config terms. For example: 'cycles/overwrite/' and 'instructions/no-overwrite/'.
 
+Implies --tail-synthesize.
+
 SEE ALSO
 --------
 linkperf:perf-stat[1], linkperf:perf-list[1]
diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c
index 39c7486f0607..8f2c16d9275f 100644
--- a/tools/perf/builtin-record.c
+++ b/tools/perf/builtin-record.c
@@ -604,13 +604,16 @@ record__finish_output(struct record *rec)
         return;
 }
 
-static int record__synthesize_workload(struct record *rec)
+static int record__synthesize_workload(struct record *rec, bool tail)
 {
         struct {
                 struct thread_map map;
                 struct thread_map_data map_data;
         } thread_map;
 
+        if (rec->opts.tail_synthesize != tail)
+                return 0;
+
         thread_map.map.nr = 1;
         thread_map.map.map[0].pid = rec->evlist->workload.pid;
         thread_map.map.map[0].comm = NULL;
@@ -621,7 +624,7 @@ static int record__synthesize_workload(struct record *rec)
                                                  rec->opts.proc_map_timeout);
 }
 
-static int record__synthesize(struct record *rec);
+static int record__synthesize(struct record *rec, bool tail);
 
 static int
 record__switch_output(struct record *rec, bool at_exit)
@@ -632,6 +635,10 @@ record__switch_output(struct record *rec, bool at_exit)
         /* Same Size:      "2015122520103046"*/
         char timestamp[] = "InvalidTimestamp";
 
+        record__synthesize(rec, true);
+        if (target__none(&rec->opts.target))
+                record__synthesize_workload(rec, true);
+
         rec->samples = 0;
         record__finish_output(rec);
         err = fetch_current_timestamp(timestamp, sizeof(timestamp));
@@ -654,7 +661,7 @@ record__switch_output(struct record *rec, bool at_exit)
 
         /* Output tracking events */
         if (!at_exit) {
-                record__synthesize(rec);
+                record__synthesize(rec, false);
 
                 /*
                  * In 'perf record --switch-output' without -a,
@@ -666,7 +673,7 @@ record__switch_output(struct record *rec, bool at_exit)
                  * perf_event__synthesize_thread_map() for those events.
                  */
                 if (target__none(&rec->opts.target))
-                        record__synthesize_workload(rec);
+                        record__synthesize_workload(rec, false);
         }
         return fd;
 }
@@ -720,7 +727,7 @@ static const struct perf_event_mmap_page *record__pick_pc(struct record *rec)
         return NULL;
 }
 
-static int record__synthesize(struct record *rec)
+static int record__synthesize(struct record *rec, bool tail)
 {
         struct perf_session *session = rec->session;
         struct machine *machine = &session->machines.host;
@@ -730,6 +737,9 @@ static int record__synthesize(struct record *rec)
         int fd = perf_data_file__fd(file);
         int err = 0;
 
+        if (rec->opts.tail_synthesize != tail)
+                return 0;
+
         if (file->is_pipe) {
                 err = perf_event__synthesize_attrs(tool, session,
                                                    process_synthesized_event);
@@ -893,7 +903,7 @@ static int __cmd_record(struct record *rec, int argc, const char **argv)
 
         machine = &session->machines.host;
 
-        err = record__synthesize(rec);
+        err = record__synthesize(rec, false);
         if (err < 0)
                 goto out_child;
 
@@ -1057,6 +1067,9 @@ static int __cmd_record(struct record *rec, int argc, const char **argv)
         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:
         if (forks) {
                 int exit_status;
@@ -1075,6 +1088,7 @@ out_child:
         } else
                 status = err;
 
+        record__synthesize(rec, true);
         /* this will be recalculated during process_buildids() */
         rec->samples = 0;
 
@@ -1399,6 +1413,8 @@ struct option __record_options[] = {
         OPT_BOOLEAN_SET('i', "no-inherit", &record.opts.no_inherit,
                         &record.opts.no_inherit_set,
                         "child tasks do not inherit counters"),
+        OPT_BOOLEAN(0, "tail-synthesize", &record.opts.tail_synthesize,
+                    "synthesize non-sample events at the end of output"),
         OPT_BOOLEAN(0, "overwrite", &record.opts.overwrite, "use overwrite mode"),
         OPT_UINTEGER('F', "freq", &record.opts.user_freq, "profile at this frequency"),
         OPT_CALLBACK('m', "mmap-pages", &record.opts, "pages[,pages]",
@@ -1610,6 +1626,9 @@ int cmd_record(int argc, const char **argv, const char *prefix __maybe_unused)
                 }
         }
 
+        if (record.opts.overwrite)
+                record.opts.tail_synthesize = true;
+
         if (rec->evlist->nr_entries == 0 &&
             perf_evlist__add_default(rec->evlist) < 0) {
                 pr_err("Not enough memory for event selector list\n");
diff --git a/tools/perf/perf.h b/tools/perf/perf.h
index 608b42bdb1b6..a7e0f1497244 100644
--- a/tools/perf/perf.h
+++ b/tools/perf/perf.h
@@ -59,6 +59,7 @@ struct record_opts {
         bool         record_switch_events;
         bool         all_kernel;
         bool         all_user;
+        bool         tail_synthesize;
         bool         overwrite;
         unsigned int freq;
         unsigned int mmap_pages;