From: Steven Rostedt <rostedt@goodmis.org>
To: linux-kernel@vger.kernel.org
Cc: Ingo Molnar <mingo@kernel.org>,
Andrew Morton <akpm@linux-foundation.org>,
Tom Zanussi <tom.zanussi@linux.intel.com>,
Rajvi Jingar <rajvi.jingar@intel.com>
Subject: [for-next][PATCH 34/46] tracing: Add a clock attribute for hist triggers
Date: Tue, 20 Mar 2018 22:21:26 -0400 [thread overview]
Message-ID: <20180321022129.700959357@goodmis.org> (raw)
In-Reply-To: 20180321022052.030055576@goodmis.org
[-- Attachment #1: 0034-tracing-Add-a-clock-attribute-for-hist-triggers.patch --]
[-- Type: text/plain, Size: 4449 bytes --]
From: Tom Zanussi <tom.zanussi@linux.intel.com>
The default clock if timestamps are used in a histogram is "global".
If timestamps aren't used, the clock is irrelevant.
Use the "clock=" param only if you want to override the default
"global" clock for a histogram with timestamps.
Link: http://lkml.kernel.org/r/427bed1389c5d22aa40c3e0683e30cc3d151e260.1516069914.git.tom.zanussi@linux.intel.com
Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com>
Signed-off-by: Rajvi Jingar <rajvi.jingar@intel.com>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
---
Documentation/trace/histogram.txt | 11 +++++++++-
kernel/trace/trace_events_hist.c | 42 ++++++++++++++++++++++++++++++++++++---
2 files changed, 49 insertions(+), 4 deletions(-)
diff --git a/Documentation/trace/histogram.txt b/Documentation/trace/histogram.txt
index df08882d091c..6e05510afc28 100644
--- a/Documentation/trace/histogram.txt
+++ b/Documentation/trace/histogram.txt
@@ -1671,7 +1671,16 @@ features have been added to the hist trigger support:
it is in units of nanoseconds; appending '.usecs' to a
common_timestamp field changes the units to microseconds.
-These features are decribed in more detail in the following sections.
+A note on inter-event timestamps: If common_timestamp is used in a
+histogram, the trace buffer is automatically switched over to using
+absolute timestamps and the "global" trace clock, in order to avoid
+bogus timestamp differences with other clocks that aren't coherent
+across CPUs. This can be overridden by specifying one of the other
+trace clocks instead, using the "clock=XXX" hist trigger attribute,
+where XXX is any of the clocks listed in the tracing/trace_clock
+pseudo-file.
+
+These features are described in more detail in the following sections.
2.2.1 Histogram Variables
-------------------------
diff --git a/kernel/trace/trace_events_hist.c b/kernel/trace/trace_events_hist.c
index 8719b0ea672f..f7d0da20c5c8 100644
--- a/kernel/trace/trace_events_hist.c
+++ b/kernel/trace/trace_events_hist.c
@@ -242,6 +242,7 @@ struct hist_trigger_attrs {
char *vals_str;
char *sort_key_str;
char *name;
+ char *clock;
bool pause;
bool cont;
bool clear;
@@ -1776,6 +1777,7 @@ static void destroy_hist_trigger_attrs(struct hist_trigger_attrs *attrs)
kfree(attrs->sort_key_str);
kfree(attrs->keys_str);
kfree(attrs->vals_str);
+ kfree(attrs->clock);
kfree(attrs);
}
@@ -1831,6 +1833,19 @@ static int parse_assignment(char *str, struct hist_trigger_attrs *attrs)
ret = -ENOMEM;
goto out;
}
+ } else if (strncmp(str, "clock=", strlen("clock=")) == 0) {
+ strsep(&str, "=");
+ if (!str) {
+ ret = -EINVAL;
+ goto out;
+ }
+
+ str = strstrip(str);
+ attrs->clock = kstrdup(str, GFP_KERNEL);
+ if (!attrs->clock) {
+ ret = -ENOMEM;
+ goto out;
+ }
} else if (strncmp(str, "size=", strlen("size=")) == 0) {
int map_bits = parse_map_size(str);
@@ -1895,6 +1910,14 @@ static struct hist_trigger_attrs *parse_hist_trigger_attrs(char *trigger_str)
goto free;
}
+ if (!attrs->clock) {
+ attrs->clock = kstrdup("global", GFP_KERNEL);
+ if (!attrs->clock) {
+ ret = -ENOMEM;
+ goto free;
+ }
+ }
+
return attrs;
free:
destroy_hist_trigger_attrs(attrs);
@@ -4934,6 +4957,8 @@ static int event_hist_trigger_print(struct seq_file *m,
seq_puts(m, ".descending");
}
seq_printf(m, ":size=%u", (1 << hist_data->map->map_bits));
+ if (hist_data->enable_timestamps)
+ seq_printf(m, ":clock=%s", hist_data->attrs->clock);
print_actions_spec(m, hist_data);
@@ -5201,7 +5226,6 @@ static int hist_register_trigger(char *glob, struct event_trigger_ops *ops,
data->paused = true;
if (named_data) {
- destroy_hist_data(data->private_data);
data->private_data = named_data->private_data;
set_named_trigger_data(data, named_data);
data->ops = &event_hist_trigger_named_ops;
@@ -5213,10 +5237,22 @@ static int hist_register_trigger(char *glob, struct event_trigger_ops *ops,
goto out;
}
- ret++;
+ if (hist_data->enable_timestamps) {
+ char *clock = hist_data->attrs->clock;
+
+ ret = tracing_set_clock(file->tr, hist_data->attrs->clock);
+ if (ret) {
+ hist_err("Couldn't set trace_clock: ", clock);
+ goto out;
+ }
- if (hist_data->enable_timestamps)
tracing_set_time_stamp_abs(file->tr, true);
+ }
+
+ if (named_data)
+ destroy_hist_data(hist_data);
+
+ ret++;
out:
return ret;
}
--
2.15.1
next prev parent reply other threads:[~2018-03-21 2:24 UTC|newest]
Thread overview: 47+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-03-21 2:20 [for-next][PATCH 00/46] tracing: Updates for v4.17 Steven Rostedt
2018-03-21 2:20 ` [for-next][PATCH 01/46] tracing: Move hist trigger Documentation to histogram.txt Steven Rostedt
2018-03-21 2:20 ` [for-next][PATCH 02/46] tracing: Add Documentation for log2 modifier Steven Rostedt
2018-03-21 2:20 ` [for-next][PATCH 03/46] tracing: Add support to detect and avoid duplicates Steven Rostedt
2018-03-21 2:20 ` [for-next][PATCH 04/46] tracing: Remove code which merges duplicates Steven Rostedt
2018-03-21 2:20 ` [for-next][PATCH 05/46] ring-buffer: Add interface for setting absolute time stamps Steven Rostedt
2018-03-21 2:20 ` [for-next][PATCH 06/46] ring-buffer: Redefine the unimplemented RINGBUF_TYPE_TIME_STAMP Steven Rostedt
2018-03-21 2:20 ` [for-next][PATCH 07/46] tracing: Add timestamp_mode trace file Steven Rostedt
2018-03-21 2:21 ` [for-next][PATCH 08/46] tracing: Give event triggers access to ring_buffer_event Steven Rostedt
2018-03-21 2:21 ` [for-next][PATCH 09/46] tracing: Add ring buffer event param to hist field functions Steven Rostedt
2018-03-21 2:21 ` [for-next][PATCH 10/46] tracing: Break out hist trigger assignment parsing Steven Rostedt
2018-03-21 2:21 ` [for-next][PATCH 11/46] tracing: Add hist trigger timestamp support Steven Rostedt
2018-03-21 2:21 ` [for-next][PATCH 12/46] tracing: Add per-element variable support to tracing_map Steven Rostedt
2018-03-21 2:21 ` [for-next][PATCH 13/46] tracing: Add hist_data member to hist_field Steven Rostedt
2018-03-21 2:21 ` [for-next][PATCH 14/46] tracing: Add usecs modifier for hist trigger timestamps Steven Rostedt
2018-03-21 2:21 ` [for-next][PATCH 15/46] tracing: Add variable support to hist triggers Steven Rostedt
2018-03-21 2:21 ` [for-next][PATCH 16/46] tracing: Account for variables in named trigger compatibility Steven Rostedt
2018-03-21 2:21 ` [for-next][PATCH 17/46] tracing: Move get_hist_field_flags() Steven Rostedt
2018-03-21 2:21 ` [for-next][PATCH 18/46] tracing: Add simple expression support to hist triggers Steven Rostedt
2018-03-21 2:21 ` [for-next][PATCH 19/46] tracing: Generalize per-element hist trigger data Steven Rostedt
2018-03-21 2:21 ` [for-next][PATCH 20/46] tracing: Pass tracing_map_elt to hist_field accessor functions Steven Rostedt
2018-03-21 2:21 ` [for-next][PATCH 21/46] tracing: Add hist_field type field Steven Rostedt
2018-03-21 2:21 ` [for-next][PATCH 22/46] tracing: Add variable reference handling to hist triggers Steven Rostedt
2018-03-21 2:21 ` [for-next][PATCH 23/46] tracing: Add hist trigger action hook Steven Rostedt
2018-03-21 2:21 ` [for-next][PATCH 24/46] tracing: Add support for synthetic events Steven Rostedt
2018-03-21 2:21 ` [for-next][PATCH 25/46] tracing: Add support for field variables Steven Rostedt
2018-03-21 2:21 ` [for-next][PATCH 26/46] tracing: Add onmatch hist trigger action support Steven Rostedt
2018-03-21 2:21 ` [for-next][PATCH 27/46] tracing: Add onmax " Steven Rostedt
2018-03-21 2:21 ` [for-next][PATCH 28/46] tracing: Allow whitespace to surround hist trigger filter Steven Rostedt
2018-03-21 2:21 ` [for-next][PATCH 29/46] tracing: Add cpu field for hist triggers Steven Rostedt
2018-03-21 2:21 ` [for-next][PATCH 30/46] tracing: Add hist trigger support for variable reference aliases Steven Rostedt
2018-03-21 2:21 ` [for-next][PATCH 31/46] tracing: Add last error error facility for hist triggers Steven Rostedt
2018-03-21 2:21 ` [for-next][PATCH 32/46] tracing: Add inter-event hist trigger Documentation Steven Rostedt
2018-03-21 2:21 ` [for-next][PATCH 33/46] tracing: Make tracing_set_clock() non-static Steven Rostedt
2018-03-21 2:21 ` Steven Rostedt [this message]
2018-03-21 2:21 ` [for-next][PATCH 35/46] ring-buffer: Add nesting for adding events within events Steven Rostedt
2018-03-21 2:21 ` [for-next][PATCH 36/46] tracing: Use the ring-buffer nesting to allow synthetic events to be traced Steven Rostedt
2018-03-21 2:21 ` [for-next][PATCH 37/46] tracing: Add inter-event blurb to HIST_TRIGGERS config option Steven Rostedt
2018-03-21 2:21 ` [for-next][PATCH 38/46] selftests: ftrace: Add inter-event hist triggers testcases Steven Rostedt
2018-03-21 2:21 ` [for-next][PATCH 39/46] tracing: Remove BUG_ON() from append_filter_string() Steven Rostedt
2018-03-21 2:21 ` [for-next][PATCH 40/46] tracing: Use trace_seq instead of open code string appending Steven Rostedt
2018-03-21 2:21 ` [for-next][PATCH 41/46] tracing: Remove filter allocator helper Steven Rostedt
2018-03-21 2:21 ` [for-next][PATCH 42/46] tracing: Only add filter list when needed Steven Rostedt
2018-03-21 2:21 ` [for-next][PATCH 43/46] tracing: Embed replace_filter_string() helper function Steven Rostedt
2018-03-21 2:21 ` [for-next][PATCH 44/46] tracing: Combine enum and arrays into single macro in filter code Steven Rostedt
2018-03-21 2:21 ` [for-next][PATCH 45/46] tracing: Clean up and document pred_funcs_##type creation and use Steven Rostedt
2018-03-21 2:21 ` [for-next][PATCH 46/46] tracing: Rewrite filter logic to be simpler and faster Steven Rostedt
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=20180321022129.700959357@goodmis.org \
--to=rostedt@goodmis.org \
--cc=akpm@linux-foundation.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@kernel.org \
--cc=rajvi.jingar@intel.com \
--cc=tom.zanussi@linux.intel.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.