From: Arnaldo Carvalho de Melo <acme@kernel.org>
To: Ingo Molnar <mingo@kernel.org>
Cc: linux-kernel@vger.kernel.org, linux-perf-users@vger.kernel.org,
Kan Liang <Kan.liang@intel.com>,
Dan Williams <dan.j.williams@intel.com>,
Jiri Olsa <jolsa@kernel.org>,
Peter Zijlstra <peterz@infradead.org>,
Philippe Ombredanne <pombredanne@nexb.com>,
Stephane Eranian <eranian@google.com>,
Arnaldo Carvalho de Melo <acme@redhat.com>
Subject: [PATCH 03/32] perf script python: Add script to profile and resolve physical mem type
Date: Wed, 17 Jan 2018 13:11:53 -0300 [thread overview]
Message-ID: <20180117161222.15611-4-acme@kernel.org> (raw)
In-Reply-To: <20180117161222.15611-1-acme@kernel.org>
From: Kan Liang <Kan.liang@intel.com>
There could be different types of memory in the system. E.g normal
System Memory, Persistent Memory. To understand how the workload maps to
those memories, it's important to know the I/O statistics of them. Perf
can collect physical addresses, but those are raw data. It still needs
extra work to resolve the physical addresses. Provide a script to
facilitate the physical addresses resolving and I/O statistics.
Profile with MEM_INST_RETIRED.ALL_LOADS or MEM_UOPS_RETIRED.ALL_LOADS
event if any of them is available.
Look up the /proc/iomem and resolve the physical address. Provide
memory type summary.
Here is an example output:
# perf script report mem-phys-addr
Event: mem_inst_retired.all_loads:P
Memory type count percentage
---------------------------------------- ----------- -----------
System RAM 74 53.2%
Persistent Memory 55 39.6%
N/A
---
Changes since V2:
- Apply the new license rules.
- Add comments for globals
Changes since V1:
- Do not mix DLA and Load Latency. Do not compare the loads and stores.
Only profile the loads.
- Use event name to replace the RAW event
Signed-off-by: Kan Liang <Kan.liang@intel.com>
Reviewed-by: Andi Kleen <ak@linux.intel.com>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Philippe Ombredanne <pombredanne@nexb.com>
Cc: Stephane Eranian <eranian@google.com>
Link: https://lkml.kernel.org/r/1515099595-34770-1-git-send-email-kan.liang@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
tools/perf/scripts/python/bin/mem-phys-addr-record | 19 +++++
tools/perf/scripts/python/bin/mem-phys-addr-report | 3 +
tools/perf/scripts/python/mem-phys-addr.py | 95 ++++++++++++++++++++++
.../util/scripting-engines/trace-event-python.c | 2 +
4 files changed, 119 insertions(+)
create mode 100644 tools/perf/scripts/python/bin/mem-phys-addr-record
create mode 100644 tools/perf/scripts/python/bin/mem-phys-addr-report
create mode 100644 tools/perf/scripts/python/mem-phys-addr.py
diff --git a/tools/perf/scripts/python/bin/mem-phys-addr-record b/tools/perf/scripts/python/bin/mem-phys-addr-record
new file mode 100644
index 000000000000..5a875122a904
--- /dev/null
+++ b/tools/perf/scripts/python/bin/mem-phys-addr-record
@@ -0,0 +1,19 @@
+#!/bin/bash
+
+#
+# Profiling physical memory by all retired load instructions/uops event
+# MEM_INST_RETIRED.ALL_LOADS or MEM_UOPS_RETIRED.ALL_LOADS
+#
+
+load=`perf list | grep mem_inst_retired.all_loads`
+if [ -z "$load" ]; then
+ load=`perf list | grep mem_uops_retired.all_loads`
+fi
+if [ -z "$load" ]; then
+ echo "There is no event to count all retired load instructions/uops."
+ exit 1
+fi
+
+arg=$(echo $load | tr -d ' ')
+arg="$arg:P"
+perf record --phys-data -e $arg $@
diff --git a/tools/perf/scripts/python/bin/mem-phys-addr-report b/tools/perf/scripts/python/bin/mem-phys-addr-report
new file mode 100644
index 000000000000..3f2b847e2eab
--- /dev/null
+++ b/tools/perf/scripts/python/bin/mem-phys-addr-report
@@ -0,0 +1,3 @@
+#!/bin/bash
+# description: resolve physical address samples
+perf script $@ -s "$PERF_EXEC_PATH"/scripts/python/mem-phys-addr.py
diff --git a/tools/perf/scripts/python/mem-phys-addr.py b/tools/perf/scripts/python/mem-phys-addr.py
new file mode 100644
index 000000000000..ebee2c5ae496
--- /dev/null
+++ b/tools/perf/scripts/python/mem-phys-addr.py
@@ -0,0 +1,95 @@
+# mem-phys-addr.py: Resolve physical address samples
+# SPDX-License-Identifier: GPL-2.0
+#
+# Copyright (c) 2018, Intel Corporation.
+
+from __future__ import division
+import os
+import sys
+import struct
+import re
+import bisect
+import collections
+
+sys.path.append(os.environ['PERF_EXEC_PATH'] + \
+ '/scripts/python/Perf-Trace-Util/lib/Perf/Trace')
+
+#physical address ranges for System RAM
+system_ram = []
+#physical address ranges for Persistent Memory
+pmem = []
+#file object for proc iomem
+f = None
+#Count for each type of memory
+load_mem_type_cnt = collections.Counter()
+#perf event name
+event_name = None
+
+def parse_iomem():
+ global f
+ f = open('/proc/iomem', 'r')
+ for i, j in enumerate(f):
+ m = re.split('-|:',j,2)
+ if m[2].strip() == 'System RAM':
+ system_ram.append(long(m[0], 16))
+ system_ram.append(long(m[1], 16))
+ if m[2].strip() == 'Persistent Memory':
+ pmem.append(long(m[0], 16))
+ pmem.append(long(m[1], 16))
+
+def print_memory_type():
+ print "Event: %s" % (event_name)
+ print "%-40s %10s %10s\n" % ("Memory type", "count", "percentage"),
+ print "%-40s %10s %10s\n" % ("----------------------------------------", \
+ "-----------", "-----------"),
+ total = sum(load_mem_type_cnt.values())
+ for mem_type, count in sorted(load_mem_type_cnt.most_common(), \
+ key = lambda(k, v): (v, k), reverse = True):
+ print "%-40s %10d %10.1f%%\n" % (mem_type, count, 100 * count / total),
+
+def trace_begin():
+ parse_iomem()
+
+def trace_end():
+ print_memory_type()
+ f.close()
+
+def is_system_ram(phys_addr):
+ #/proc/iomem is sorted
+ position = bisect.bisect(system_ram, phys_addr)
+ if position % 2 == 0:
+ return False
+ return True
+
+def is_persistent_mem(phys_addr):
+ position = bisect.bisect(pmem, phys_addr)
+ if position % 2 == 0:
+ return False
+ return True
+
+def find_memory_type(phys_addr):
+ if phys_addr == 0:
+ return "N/A"
+ if is_system_ram(phys_addr):
+ return "System RAM"
+
+ if is_persistent_mem(phys_addr):
+ return "Persistent Memory"
+
+ #slow path, search all
+ f.seek(0, 0)
+ for j in f:
+ m = re.split('-|:',j,2)
+ if long(m[0], 16) <= phys_addr <= long(m[1], 16):
+ return m[2]
+ return "N/A"
+
+def process_event(param_dict):
+ name = param_dict["ev_name"]
+ sample = param_dict["sample"]
+ phys_addr = sample["phys_addr"]
+
+ global event_name
+ if event_name == None:
+ event_name = name
+ load_mem_type_cnt[find_memory_type(phys_addr)] += 1
diff --git a/tools/perf/util/scripting-engines/trace-event-python.c b/tools/perf/util/scripting-engines/trace-event-python.c
index c1848b543f27..ea070883c593 100644
--- a/tools/perf/util/scripting-engines/trace-event-python.c
+++ b/tools/perf/util/scripting-engines/trace-event-python.c
@@ -499,6 +499,8 @@ static PyObject *get_perf_sample_dict(struct perf_sample *sample,
PyLong_FromUnsignedLongLong(sample->time));
pydict_set_item_string_decref(dict_sample, "period",
PyLong_FromUnsignedLongLong(sample->period));
+ pydict_set_item_string_decref(dict_sample, "phys_addr",
+ PyLong_FromUnsignedLongLong(sample->phys_addr));
set_sample_read_in_dict(dict_sample, sample, evsel);
pydict_set_item_string_decref(dict, "sample", dict_sample);
--
2.14.3
next prev parent reply other threads:[~2018-01-17 16:12 UTC|newest]
Thread overview: 34+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-01-17 16:11 [GIT PULL 00/32] perf/core improvements and fixes Arnaldo Carvalho de Melo
2018-01-17 16:11 ` [PATCH 01/32] perf evsel: Fix incorrect handling of type _TERM_DRV_CFG Arnaldo Carvalho de Melo
2018-01-17 16:11 ` [PATCH 02/32] perf evlist: Remove trailing semicolon Arnaldo Carvalho de Melo
2018-01-17 16:11 ` Arnaldo Carvalho de Melo [this message]
2018-01-17 16:11 ` [PATCH 04/32] perf trace: No need to set PERF_SAMPLE_IDENTIFIER explicitely Arnaldo Carvalho de Melo
2018-01-17 16:11 ` [PATCH 05/32] perf tools: Fix copyfile_offset update of output offset Arnaldo Carvalho de Melo
2018-01-17 16:11 ` [PATCH 06/32] perf evsel: Check if callchain is enabled before setting it up Arnaldo Carvalho de Melo
2018-01-17 16:11 ` [PATCH 07/32] perf trace: Fix setting of --call-graph/--max-stack for non-syscall events Arnaldo Carvalho de Melo
2018-01-17 16:11 ` [PATCH 08/32] tools lib traceevent: Fix bad force_token escape sequence Arnaldo Carvalho de Melo
2018-01-17 16:11 ` [PATCH 09/32] tools lib traceevent: Show value of flags that have not been parsed Arnaldo Carvalho de Melo
2018-01-17 16:12 ` [PATCH 10/32] tools lib traceevent: Print value of unknown symbolic fields Arnaldo Carvalho de Melo
2018-01-17 16:12 ` [PATCH 11/32] tools lib traceevent: Simplify pointer print logic and fix %pF Arnaldo Carvalho de Melo
2018-01-17 16:12 ` [PATCH 12/32] tools lib traceevent: Handle new pointer processing of bprint strings Arnaldo Carvalho de Melo
2018-01-17 16:12 ` [PATCH 13/32] tools lib traceevent: Show contents (in hex) of data of unrecognized type records Arnaldo Carvalho de Melo
2018-01-17 16:12 ` [PATCH 14/32] tools lib traceevent: Use asprintf when possible Arnaldo Carvalho de Melo
2018-01-17 16:12 ` [PATCH 15/32] tools lib traceevent: Add UL suffix to MISSING_EVENTS Arnaldo Carvalho de Melo
2018-01-17 16:12 ` [PATCH 16/32] tools lib traceevent: Fix missing break in FALSE case of pevent_filter_clear_trivial() Arnaldo Carvalho de Melo
2018-01-17 16:12 ` [PATCH 17/32] tools lib traceevent: Fix get_field_str() for dynamic strings Arnaldo Carvalho de Melo
2018-01-17 16:12 ` [PATCH 18/32] perf tools: Add ARM Statistical Profiling Extensions (SPE) support Arnaldo Carvalho de Melo
2018-01-17 16:12 ` [PATCH 19/32] perf callchain: Fix attr.sample_max_stack setting Arnaldo Carvalho de Melo
2018-01-17 16:12 ` [PATCH 20/32] perf unwind: Do not look just at the global callchain_param.record_mode Arnaldo Carvalho de Melo
2018-01-17 16:12 ` [PATCH 21/32] perf trace: Setup DWARF callchains for non-syscall events when --max-stack is used Arnaldo Carvalho de Melo
2018-01-17 16:12 ` [PATCH 22/32] perf trace: Allow overriding global --max-stack per event Arnaldo Carvalho de Melo
2018-01-17 16:12 ` [PATCH 23/32] perf callchains: Ask for PERF_RECORD_MMAP for data mmaps for DWARF unwinding Arnaldo Carvalho de Melo
2018-01-17 16:12 ` [PATCH 24/32] perf report: Improve error msg when no first/last sample time found Arnaldo Carvalho de Melo
2018-01-17 16:12 ` [PATCH 25/32] perf script: " Arnaldo Carvalho de Melo
2018-01-17 16:12 ` [PATCH 26/32] perf util: Improve error checking for time percent input Arnaldo Carvalho de Melo
2018-01-17 16:12 ` [PATCH 27/32] perf util: Support no index time percent slice Arnaldo Carvalho de Melo
2018-01-17 16:12 ` [PATCH 28/32] perf report: Add an indication of what time slices are used Arnaldo Carvalho de Melo
2018-01-17 16:12 ` [PATCH 29/32] perf util: Allocate time slices buffer according to number of comma Arnaldo Carvalho de Melo
2018-01-17 16:12 ` [PATCH 30/32] perf report: Remove the time slices number limitation Arnaldo Carvalho de Melo
2018-01-17 16:12 ` [PATCH 31/32] perf script: " Arnaldo Carvalho de Melo
2018-01-17 16:12 ` [PATCH 32/32] perf record: Fix failed memory allocation for get_cpuid_str Arnaldo Carvalho de Melo
2018-01-17 16:22 ` [GIT PULL 00/32] perf/core improvements and fixes Ingo Molnar
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=20180117161222.15611-4-acme@kernel.org \
--to=acme@kernel.org \
--cc=Kan.liang@intel.com \
--cc=acme@redhat.com \
--cc=dan.j.williams@intel.com \
--cc=eranian@google.com \
--cc=jolsa@kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-perf-users@vger.kernel.org \
--cc=mingo@kernel.org \
--cc=peterz@infradead.org \
--cc=pombredanne@nexb.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;
as well as URLs for NNTP newsgroup(s).