mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Wang Nan <wangnan0@huawei.com>
To: Alexei Starovoitov <ast@kernel.org>,
	Arnaldo Carvalho de Melo <acme@redhat.com>,
	Arnaldo Carvalho de Melo <acme@kernel.org>,
	Brendan Gregg <brendan.d.gregg@gmail.com>
Cc: "Adrian Hunter" <adrian.hunter@intel.com>,
	"Cody P Schafer" <dev@codyps.com>,
	"David S. Miller" <davem@davemloft.net>,
	"He Kuang" <hekuang@huawei.com>,
	"Jérémie Galarneau" <jeremie.galarneau@efficios.com>,
	"Jiri Olsa" <jolsa@kernel.org>,
	"Kirill Smelkov" <kirr@nexedi.com>,
	"Li Zefan" <lizefan@huawei.com>,
	"Masami Hiramatsu" <masami.hiramatsu.pt@hitachi.com>,
	"Namhyung Kim" <namhyung@kernel.org>,
	"Peter Zijlstra" <peterz@infradead.org>,
	pi3orama@163.com, "Wang Nan" <wangnan0@huawei.com>,
	linux-kernel@vger.kernel.org
Subject: [PATCH 33/48] perf tools: Add evlist channel helpers
Date: Mon, 22 Feb 2016 09:11:00 +0000	[thread overview]
Message-ID: <1456132275-98875-34-git-send-email-wangnan0@huawei.com> (raw)
In-Reply-To: <1456132275-98875-1-git-send-email-wangnan0@huawei.com>

In this commit sereval helpers are introduced to support the principle
of channel. Channels hold different groups of evsels which configured
differently. It will be used for overwritable evsels, which allows perf
record some events continuously while capture snapshot for other events
when something happen. Tracking events (mmap, mmap2, fork, exit ...)
are another possible events worth to be put into a separated channel.

Channels are represented by an array with channel flags. Each channel
contains evlist->nr_mmaps mmaps. Channels are configured before
perf_evlist__mmap_ex(). During that function nr_mmaps mmaps for each
channel are allocated together as a big array.
perf_evlist__channel_idx() converts index in the big array and the
channel number. For API functions which accept idx, _ex() versions are
introduced to accept selecting an mmap from a channel.

Signed-off-by: Wang Nan <wangnan0@huawei.com>
Signed-off-by: He Kuang <hekuang@huawei.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Zefan Li <lizefan@huawei.com>
Cc: pi3orama@163.com
---
 tools/perf/builtin-record.c |   6 ++
 tools/perf/util/evlist.c    | 132 ++++++++++++++++++++++++++++++++++++++++++--
 tools/perf/util/evlist.h    |  58 +++++++++++++++++++
 3 files changed, 190 insertions(+), 6 deletions(-)

diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c
index 3a7de24..24c776c 100644
--- a/tools/perf/builtin-record.c
+++ b/tools/perf/builtin-record.c
@@ -356,6 +356,12 @@ try_again:
 		goto out;
 	}
 
+	perf_evlist__channel_reset(evlist);
+	rc = perf_evlist__channel_add(evlist, 0, true);
+	if (rc < 0)
+		goto out;
+	rc = 0;
+
 	if (perf_evlist__mmap_ex(evlist, opts->mmap_pages, false,
 				 opts->auxtrace_mmap_pages,
 				 opts->auxtrace_snapshot_mode) < 0) {
diff --git a/tools/perf/util/evlist.c b/tools/perf/util/evlist.c
index fef465a..a6b52fc 100644
--- a/tools/perf/util/evlist.c
+++ b/tools/perf/util/evlist.c
@@ -679,14 +679,51 @@ static struct perf_evsel *perf_evlist__event2evsel(struct perf_evlist *evlist,
 	return NULL;
 }
 
-union perf_event *perf_evlist__mmap_read(struct perf_evlist *evlist, int idx)
+int perf_evlist__channel_idx(struct perf_evlist *evlist,
+			     int *p_channel, int *p_idx)
+{
+	int channel = *p_channel;
+	int _idx = *p_idx;
+
+	if (_idx < 0)
+		return -EINVAL;
+	/*
+	 * Negative channel means caller explicitly use real index.
+	 */
+	if (channel < 0) {
+		channel = perf_evlist__idx_channel(evlist, _idx);
+		_idx = _idx % evlist->nr_mmaps;
+	}
+	if (channel < 0)
+		return channel;
+	if (channel >= PERF_EVLIST__NR_CHANNELS)
+		return -E2BIG;
+	if (_idx >= evlist->nr_mmaps)
+		return -E2BIG;
+
+	*p_channel = channel;
+	*p_idx = evlist->nr_mmaps * channel + _idx;
+	return 0;
+}
+
+union perf_event *perf_evlist__mmap_read_ex(struct perf_evlist *evlist,
+					    int channel, int idx)
 {
+	int err = perf_evlist__channel_idx(evlist, &channel, &idx);
 	struct perf_mmap *md = &evlist->mmap[idx];
 	u64 head;
-	u64 old = md->prev;
-	unsigned char *data = md->base + page_size;
+	u64 old;
+	unsigned char *data;
 	union perf_event *event = NULL;
 
+	if (err || !perf_evlist__channel_is_enabled(evlist, channel)) {
+		pr_err("ERROR: invalid mmap index: channel %d, idx: %d\n",
+		       channel, idx);
+		return NULL;
+	}
+	old = md->prev;
+	data = md->base + page_size;
+
 	/*
 	 * Check if event was unmapped due to a POLLHUP/POLLERR.
 	 */
@@ -748,6 +785,11 @@ union perf_event *perf_evlist__mmap_read(struct perf_evlist *evlist, int idx)
 	return event;
 }
 
+union perf_event *perf_evlist__mmap_read(struct perf_evlist *evlist, int idx)
+{
+	return perf_evlist__mmap_read_ex(evlist, -1, idx);
+}
+
 static bool perf_mmap__empty(struct perf_mmap *md)
 {
 	return perf_mmap__read_head(md) == md->prev && !md->auxtrace_mmap.base;
@@ -766,10 +808,18 @@ static void perf_evlist__mmap_put(struct perf_evlist *evlist, int idx)
 		__perf_evlist__munmap(evlist, idx);
 }
 
-void perf_evlist__mmap_consume(struct perf_evlist *evlist, int idx)
+void perf_evlist__mmap_consume_ex(struct perf_evlist *evlist,
+				  int channel, int idx)
 {
+	int err = perf_evlist__channel_idx(evlist, &channel, &idx);
 	struct perf_mmap *md = &evlist->mmap[idx];
 
+	if (err || !perf_evlist__channel_is_enabled(evlist, channel)) {
+		pr_err("ERROR: invalid mmap index: channel %d, idx: %d\n",
+		       channel, idx);
+		return;
+	}
+
 	if (!evlist->overwrite) {
 		u64 old = md->prev;
 
@@ -780,6 +830,11 @@ void perf_evlist__mmap_consume(struct perf_evlist *evlist, int idx)
 		perf_evlist__mmap_put(evlist, idx);
 }
 
+void perf_evlist__mmap_consume(struct perf_evlist *evlist, int idx)
+{
+	perf_evlist__mmap_consume_ex(evlist, -1, idx);
+}
+
 int __weak auxtrace_mmap__mmap(struct auxtrace_mmap *mm __maybe_unused,
 			       struct auxtrace_mmap_params *mp __maybe_unused,
 			       void *userpg __maybe_unused,
@@ -825,7 +880,7 @@ void perf_evlist__munmap(struct perf_evlist *evlist)
 	if (evlist->mmap == NULL)
 		return;
 
-	for (i = 0; i < evlist->nr_mmaps; i++)
+	for (i = 0; i < perf_evlist__mmap_nr(evlist); i++)
 		__perf_evlist__munmap(evlist, i);
 
 	zfree(&evlist->mmap);
@@ -833,10 +888,17 @@ void perf_evlist__munmap(struct perf_evlist *evlist)
 
 static int perf_evlist__alloc_mmap(struct perf_evlist *evlist)
 {
+	int total_mmaps;
+
 	evlist->nr_mmaps = cpu_map__nr(evlist->cpus);
 	if (cpu_map__empty(evlist->cpus))
 		evlist->nr_mmaps = thread_map__nr(evlist->threads);
-	evlist->mmap = zalloc(evlist->nr_mmaps * sizeof(struct perf_mmap));
+
+	total_mmaps = perf_evlist__mmap_nr(evlist);
+	if (!total_mmaps)
+		return -EINVAL;
+
+	evlist->mmap = zalloc(total_mmaps * sizeof(struct perf_mmap));
 	return evlist->mmap != NULL ? 0 : -ENOMEM;
 }
 
@@ -1137,6 +1199,12 @@ int perf_evlist__mmap_ex(struct perf_evlist *evlist, unsigned int pages,
 int perf_evlist__mmap(struct perf_evlist *evlist, unsigned int pages,
 		      bool overwrite)
 {
+	int err;
+
+	perf_evlist__channel_reset(evlist);
+	err = perf_evlist__channel_add(evlist, 0, true);
+	if (err < 0)
+		return err;
 	return perf_evlist__mmap_ex(evlist, pages, overwrite, 0, false);
 }
 
@@ -1764,3 +1832,55 @@ perf_evlist__find_evsel_by_str(struct perf_evlist *evlist,
 
 	return NULL;
 }
+
+int perf_evlist__channel_nr(struct perf_evlist *evlist)
+{
+	int i;
+
+	for (i = PERF_EVLIST__NR_CHANNELS - 1; i >= 0; i--) {
+		unsigned long flags = evlist->channel_flags[i];
+
+		if (flags & PERF_EVLIST__CHANNEL_ENABLED)
+			return i + 1;
+	}
+	return 0;
+}
+
+int perf_evlist__mmap_nr(struct perf_evlist *evlist)
+{
+	return evlist->nr_mmaps * perf_evlist__channel_nr(evlist);
+}
+
+void perf_evlist__channel_reset(struct perf_evlist *evlist)
+{
+	int i;
+
+	BUG_ON(evlist->mmap);
+
+	for (i = 0; i < PERF_EVLIST__NR_CHANNELS; i++)
+		evlist->channel_flags[i] = 0;
+}
+
+int perf_evlist__channel_add(struct perf_evlist *evlist,
+			     unsigned long flag,
+			     bool is_default)
+{
+	int n = perf_evlist__channel_nr(evlist);
+	unsigned long *flags = evlist->channel_flags;
+
+	BUG_ON(evlist->mmap);
+
+	if (n >= PERF_EVLIST__NR_CHANNELS) {
+		pr_debug("ERROR: too many channels. Increase PERF_EVLIST__NR_CHANNELS\n");
+		return -ENOSPC;
+	}
+
+	if (is_default) {
+		memmove(&flags[1], &flags[0],
+			sizeof(evlist->channel_flags) -
+			sizeof(evlist->channel_flags[0]));
+		n = 0;
+	}
+	flags[n] = flag | PERF_EVLIST__CHANNEL_ENABLED;
+	return n;
+}
diff --git a/tools/perf/util/evlist.h b/tools/perf/util/evlist.h
index a0d1522..1812652 100644
--- a/tools/perf/util/evlist.h
+++ b/tools/perf/util/evlist.h
@@ -20,6 +20,11 @@ struct record_opts;
 #define PERF_EVLIST__HLIST_BITS 8
 #define PERF_EVLIST__HLIST_SIZE (1 << PERF_EVLIST__HLIST_BITS)
 
+#define PERF_EVLIST__NR_CHANNELS	1
+enum perf_evlist_mmap_flag {
+	PERF_EVLIST__CHANNEL_ENABLED	= 1,
+};
+
 /**
  * struct perf_mmap - perf's ring buffer mmap details
  *
@@ -52,6 +57,7 @@ struct perf_evlist {
 		pid_t	pid;
 	} workload;
 	struct fdarray	 pollfd;
+	unsigned long channel_flags[PERF_EVLIST__NR_CHANNELS];
 	struct perf_mmap *mmap;
 	struct thread_map *threads;
 	struct cpu_map	  *cpus;
@@ -116,9 +122,61 @@ struct perf_evsel *perf_evlist__id2evsel_strict(struct perf_evlist *evlist,
 
 struct perf_sample_id *perf_evlist__id2sid(struct perf_evlist *evlist, u64 id);
 
+union perf_event *perf_evlist__mmap_read_ex(struct perf_evlist *evlist,
+					    int channel, int idx);
 union perf_event *perf_evlist__mmap_read(struct perf_evlist *evlist, int idx);
 
+void perf_evlist__mmap_consume_ex(struct perf_evlist *evlist,
+				  int channel, int idx);
 void perf_evlist__mmap_consume(struct perf_evlist *evlist, int idx);
+int perf_evlist__mmap_nr(struct perf_evlist *evlist);
+
+int perf_evlist__channel_nr(struct perf_evlist *evlist);
+void perf_evlist__channel_reset(struct perf_evlist *evlist);
+int perf_evlist__channel_add(struct perf_evlist *evlist,
+			     unsigned long flag,
+			     bool is_default);
+
+static inline bool
+__perf_evlist__channel_check(struct perf_evlist *evlist, int channel,
+			     enum perf_evlist_mmap_flag bits)
+{
+	if (channel >= PERF_EVLIST__NR_CHANNELS)
+		return false;
+
+	return (evlist->channel_flags[channel] & bits) ? true : false;
+}
+#define perf_evlist__channel_check(e, c, b) \
+		__perf_evlist__channel_check(e, c, PERF_EVLIST__CHANNEL_##b)
+
+static inline bool
+perf_evlist__channel_is_enabled(struct perf_evlist *evlist, int channel)
+{
+	return perf_evlist__channel_check(evlist, channel, ENABLED);
+}
+
+static inline int
+perf_evlist__idx_channel(struct perf_evlist *evlist, int idx)
+{
+	int channel = idx / evlist->nr_mmaps;
+
+	if (channel >= PERF_EVLIST__NR_CHANNELS)
+		return -E2BIG;
+	return channel;
+}
+
+int perf_evlist__channel_idx(struct perf_evlist *evlist,
+			     int *p_channel, int *p_idx);
+
+static inline struct perf_mmap *
+perf_evlist__get_mmap(struct perf_evlist *evlist,
+		      int channel, int idx)
+{
+	if (perf_evlist__channel_idx(evlist, &channel, &idx))
+		return NULL;
+
+	return &evlist->mmap[idx];
+}
 
 int perf_evlist__open(struct perf_evlist *evlist);
 void perf_evlist__close(struct perf_evlist *evlist);
-- 
1.8.3.4

  parent reply	other threads:[~2016-02-22  9:18 UTC|newest]

Thread overview: 76+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-02-22  9:10 [PATCH 00/48] perf tools: Bugfix, BPF improvements and overwrite ring buffer support Wang Nan
2016-02-22  9:10 ` [PATCH 01/48] perf tools: Record text offset in dso to calculate objdump address Wang Nan
2016-02-22  9:10 ` [PATCH 02/48] perf tools: Adjust symbol for shared objects Wang Nan
2016-02-22  9:10 ` [PATCH 03/48] perf bpf: Add API to set values to map entries in a bpf object Wang Nan
2016-02-25  5:39   ` [tip:perf/core] " tip-bot for Wang Nan
2016-02-22  9:10 ` [PATCH 04/48] perf tools: Enable BPF object configure syntax Wang Nan
2016-02-25  5:39   ` [tip:perf/core] " tip-bot for Wang Nan
2016-02-22  9:10 ` [PATCH 05/48] perf record: Apply config to BPF objects before recording Wang Nan
2016-02-25  5:39   ` [tip:perf/core] " tip-bot for Wang Nan
2016-02-22  9:10 ` [PATCH 06/48] perf tools: Enable passing event to BPF object Wang Nan
2016-02-25  5:40   ` [tip:perf/core] " tip-bot for Wang Nan
2016-02-22  9:10 ` [PATCH 07/48] perf tools: Support setting different slots in a BPF map separately Wang Nan
2016-02-25  5:40   ` [tip:perf/core] " tip-bot for Wang Nan
2016-02-22  9:10 ` [PATCH 08/48] perf tools: Enable indices setting syntax for BPF map Wang Nan
2016-02-25  5:40   ` [tip:perf/core] " tip-bot for Wang Nan
2016-02-22  9:10 ` [PATCH 09/48] perf tools: Pass tracepoint options to BPF script Wang Nan
2016-02-25  5:41   ` [tip:perf/core] perf tools: Apply tracepoint event definition " tip-bot for Wang Nan
2016-02-22  9:10 ` [PATCH 10/48] perf tools: Introduce bpf-output event Wang Nan
2016-02-23 17:45   ` Arnaldo Carvalho de Melo
2016-02-24  1:58     ` Wangnan (F)
2016-02-24  2:04       ` Wangnan (F)
2016-02-24  4:03         ` Wangnan (F)
2016-02-24  5:03           ` Wangnan (F)
2016-02-24 13:36       ` Arnaldo Carvalho de Melo
2016-02-25  5:41   ` [tip:perf/core] " tip-bot for Wang Nan
2016-02-22  9:10 ` [PATCH 11/48] perf data: Support converting data from bpf_perf_event_output() Wang Nan
2016-02-23 16:14   ` Arnaldo Carvalho de Melo
2016-02-23 17:23   ` Jiri Olsa
2016-02-23 17:24     ` Jiri Olsa
2016-02-23 19:22   ` Jiri Olsa
2016-02-22  9:10 ` [PATCH 12/48] perf data: Explicitly set byte order for integer types Wang Nan
2016-02-22  9:10 ` [PATCH 13/48] perf core: Introduce new ioctl options to pause and resume ring buffer Wang Nan
2016-02-22  9:10 ` [PATCH 14/48] perf core: Set event's default overflow_handler Wang Nan
2016-02-22  9:10 ` [PATCH 15/48] perf core: Prepare writing into ring buffer from end Wang Nan
2016-02-22  9:10 ` [PATCH 16/48] perf core: Add backward attribute to perf event Wang Nan
2016-02-24 13:08   ` Jiri Olsa
2016-02-24 13:21     ` Jiri Olsa
2016-02-22  9:10 ` [PATCH 17/48] perf core: Reduce perf event output overhead by new overflow handler Wang Nan
2016-02-22  9:10 ` [PATCH 18/48] perf tools: Only validate is_pos for tracking evsels Wang Nan
2016-02-24 14:21   ` Jiri Olsa
2016-02-22  9:10 ` [PATCH 19/48] perf tools: Print write_backward value in perf_event_attr__fprintf Wang Nan
2016-02-22  9:10 ` [PATCH 20/48] perf tools: Make ordered_events reusable Wang Nan
2016-02-24 14:18   ` Jiri Olsa
2016-02-22  9:10 ` [PATCH 21/48] perf record: Extract synthesize code to record__synthesize() Wang Nan
2016-02-24 14:29   ` Jiri Olsa
2016-02-22  9:10 ` [PATCH 22/48] perf tools: Add perf_data_file__switch() helper Wang Nan
2016-02-24 14:34   ` Jiri Olsa
2016-02-22  9:10 ` [PATCH 23/48] perf record: Turns auxtrace_snapshot_enable into 3 states Wang Nan
2016-02-24 14:43   ` Jiri Olsa
2016-02-22  9:10 ` [PATCH 24/48] perf record: Introduce record__finish_output() to finish a perf.data Wang Nan
2016-02-22  9:10 ` [PATCH 25/48] perf record: Add '--timestamp-filename' option to append timestamp to output filename Wang Nan
2016-02-22  9:10 ` [PATCH 26/48] perf record: Split output into multiple files via '--switch-output' Wang Nan
2016-02-22  9:10 ` [PATCH 27/48] perf record: Force enable --timestamp-filename when --switch-output is provided Wang Nan
2016-02-22  9:10 ` [PATCH 28/48] perf record: Disable buildid cache options by default in switch output mode Wang Nan
2016-02-22  9:10 ` [PATCH 29/48] perf record: Re-synthesize tracking events after output switching Wang Nan
2016-02-24 14:57   ` Jiri Olsa
2016-02-22  9:10 ` [PATCH 30/48] perf record: Generate tracking events for process forked by perf Wang Nan
2016-02-24 15:01   ` Jiri Olsa
2016-02-22  9:10 ` [PATCH 31/48] perf record: Ensure return non-zero rc when mmap fail Wang Nan
2016-02-22  9:10 ` [PATCH 32/48] perf record: Prevent reading invalid data in record__mmap_read Wang Nan
2016-02-22  9:11 ` Wang Nan [this message]
2016-02-22  9:11 ` [PATCH 34/48] perf tools: Automatically add new channel according to evlist Wang Nan
2016-02-22  9:11 ` [PATCH 35/48] perf tools: Operate multiple channels Wang Nan
2016-02-22  9:11 ` [PATCH 36/48] perf tools: Squash overwrite setting into channel Wang Nan
2016-02-22  9:11 ` [PATCH 37/48] perf record: Don't read from and poll overwrite channel Wang Nan
2016-02-22  9:11 ` [PATCH 38/48] perf record: Don't poll on " Wang Nan
2016-02-22  9:11 ` [PATCH 39/48] perf tools: Detect avalibility of write_backward Wang Nan
2016-02-22  9:11 ` [PATCH 40/48] perf tools: Enable overwrite settings Wang Nan
2016-02-22  9:11 ` [PATCH 41/48] perf tools: Set write_backward attribut bit for overwrite events Wang Nan
2016-02-22  9:11 ` [PATCH 42/48] perf tools: Record fd into perf_mmap Wang Nan
2016-02-22  9:11 ` [PATCH 43/48] perf tools: Add API to pause a channel Wang Nan
2016-02-22  9:11 ` [PATCH 44/48] perf record: Toggle overwrite ring buffer for reading Wang Nan
2016-02-22  9:11 ` [PATCH 45/48] perf record: Rename variable to make code clear Wang Nan
2016-02-22  9:11 ` [PATCH 46/48] perf record: Read from backward ring buffer Wang Nan
2016-02-22  9:11 ` [PATCH 47/48] perf record: Allow generate tracking events at the end of output Wang Nan
2016-02-22  9:11 ` [PATCH 48/48] perf tools: Don't warn about out of order event if write_backward is used Wang Nan

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1456132275-98875-34-git-send-email-wangnan0@huawei.com \
    --to=wangnan0@huawei.com \
    --cc=acme@kernel.org \
    --cc=acme@redhat.com \
    --cc=adrian.hunter@intel.com \
    --cc=ast@kernel.org \
    --cc=brendan.d.gregg@gmail.com \
    --cc=davem@davemloft.net \
    --cc=dev@codyps.com \
    --cc=hekuang@huawei.com \
    --cc=jeremie.galarneau@efficios.com \
    --cc=jolsa@kernel.org \
    --cc=kirr@nexedi.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=lizefan@huawei.com \
    --cc=masami.hiramatsu.pt@hitachi.com \
    --cc=namhyung@kernel.org \
    --cc=peterz@infradead.org \
    --cc=pi3orama@163.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®