From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
stable@vger.kernel.org,
Srikar Dronamraju <srikar@linux.vnet.ibm.com>,
Jiri Olsa <jolsa@kernel.org>,
Arnaldo Carvalho de Melo <acme@redhat.com>,
Ravi Bangoria <ravi.bangoria@linux.ibm.com>,
Namhyung Kim <namhyung@kernel.org>,
"Naveen N. Rao" <naveen.n.rao@linux.vnet.ibm.com>,
Stephane Eranian <eranian@google.com>,
Sasha Levin <sashal@kernel.org>
Subject: [PATCH 4.14 54/61] perf stat: Reset previous counts on repeat with interval
Date: Thu, 10 Oct 2019 10:37:19 +0200 [thread overview]
Message-ID: <20191010083523.606396136@linuxfoundation.org> (raw)
In-Reply-To: <20191010083449.500442342@linuxfoundation.org>
From: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
[ Upstream commit b63fd11cced17fcb8e133def29001b0f6aaa5e06 ]
When using 'perf stat' with repeat and interval option, it shows wrong
values for events.
The wrong values will be shown for the first interval on the second and
subsequent repetitions.
Without the fix:
# perf stat -r 3 -I 2000 -e faults -e sched:sched_switch -a sleep 5
2.000282489 53 faults
2.000282489 513 sched:sched_switch
4.005478208 3,721 faults
4.005478208 2,666 sched:sched_switch
5.025470933 395 faults
5.025470933 1,307 sched:sched_switch
2.009602825 1,84,46,74,40,73,70,95,47,520 faults <------
2.009602825 1,84,46,74,40,73,70,95,49,568 sched:sched_switch <------
4.019612206 4,730 faults
4.019612206 2,746 sched:sched_switch
5.039615484 3,953 faults
5.039615484 1,496 sched:sched_switch
2.000274620 1,84,46,74,40,73,70,95,47,520 faults <------
2.000274620 1,84,46,74,40,73,70,95,47,520 sched:sched_switch <------
4.000480342 4,282 faults
4.000480342 2,303 sched:sched_switch
5.000916811 1,322 faults
5.000916811 1,064 sched:sched_switch
#
prev_raw_counts is allocated when using intervals. This is used when
calculating the difference in the counts of events when using interval.
The current counts are stored in prev_raw_counts to calculate the
differences in the next iteration.
On the first interval of the second and subsequent repetitions,
prev_raw_counts would be the values stored in the last interval of the
previous repetitions, while the current counts will only be for the
first interval of the current repetition.
Hence there is a possibility of events showing up as big number.
Fix this by resetting prev_raw_counts whenever perf stat repeats the
command.
With the fix:
# perf stat -r 3 -I 2000 -e faults -e sched:sched_switch -a sleep 5
2.019349347 2,597 faults
2.019349347 2,753 sched:sched_switch
4.019577372 3,098 faults
4.019577372 2,532 sched:sched_switch
5.019415481 1,879 faults
5.019415481 1,356 sched:sched_switch
2.000178813 8,468 faults
2.000178813 2,254 sched:sched_switch
4.000404621 7,440 faults
4.000404621 1,266 sched:sched_switch
5.040196079 2,458 faults
5.040196079 556 sched:sched_switch
2.000191939 6,870 faults
2.000191939 1,170 sched:sched_switch
4.000414103 541 faults
4.000414103 902 sched:sched_switch
5.000809863 450 faults
5.000809863 364 sched:sched_switch
#
Committer notes:
This was broken since the cset introducing the --interval feature, i.e.
--repeat + --interval wasn't tested at that point, add the Fixes tag so
that automatic scripts can pick this up.
Fixes: 13370a9b5bb8 ("perf stat: Add interval printing")
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Tested-by: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com>
Cc: Stephane Eranian <eranian@google.com>
Cc: stable@vger.kernel.org # v3.9+
Link: http://lore.kernel.org/lkml/20190904094738.9558-2-srikar@linux.vnet.ibm.com
[ Fixed up conflicts with libperf, i.e. some perf_{evsel,evlist} lost the 'perf' prefix ]
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
tools/perf/builtin-stat.c | 3 +++
tools/perf/util/stat.c | 17 +++++++++++++++++
tools/perf/util/stat.h | 1 +
3 files changed, 21 insertions(+)
diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c
index b6c1c9939c2f1..0801e0ffba4ae 100644
--- a/tools/perf/builtin-stat.c
+++ b/tools/perf/builtin-stat.c
@@ -2769,6 +2769,9 @@ int cmd_stat(int argc, const char **argv)
fprintf(output, "[ perf stat: executing run #%d ... ]\n",
run_idx + 1);
+ if (run_idx != 0)
+ perf_evlist__reset_prev_raw_counts(evsel_list);
+
status = run_perf_stat(argc, argv);
if (forever && status != -1 && !interval) {
print_counters(NULL, argc, argv);
diff --git a/tools/perf/util/stat.c b/tools/perf/util/stat.c
index c9bae5fb8b479..d028c2786802e 100644
--- a/tools/perf/util/stat.c
+++ b/tools/perf/util/stat.c
@@ -154,6 +154,15 @@ static void perf_evsel__free_prev_raw_counts(struct perf_evsel *evsel)
evsel->prev_raw_counts = NULL;
}
+static void perf_evsel__reset_prev_raw_counts(struct perf_evsel *evsel)
+{
+ if (evsel->prev_raw_counts) {
+ evsel->prev_raw_counts->aggr.val = 0;
+ evsel->prev_raw_counts->aggr.ena = 0;
+ evsel->prev_raw_counts->aggr.run = 0;
+ }
+}
+
static int perf_evsel__alloc_stats(struct perf_evsel *evsel, bool alloc_raw)
{
int ncpus = perf_evsel__nr_cpus(evsel);
@@ -204,6 +213,14 @@ void perf_evlist__reset_stats(struct perf_evlist *evlist)
}
}
+void perf_evlist__reset_prev_raw_counts(struct perf_evlist *evlist)
+{
+ struct perf_evsel *evsel;
+
+ evlist__for_each_entry(evlist, evsel)
+ perf_evsel__reset_prev_raw_counts(evsel);
+}
+
static void zero_per_pkg(struct perf_evsel *counter)
{
if (counter->per_pkg_mask)
diff --git a/tools/perf/util/stat.h b/tools/perf/util/stat.h
index 96326b1f94438..bdfbed8e2df28 100644
--- a/tools/perf/util/stat.h
+++ b/tools/perf/util/stat.h
@@ -100,6 +100,7 @@ void perf_stat__collect_metric_expr(struct perf_evlist *);
int perf_evlist__alloc_stats(struct perf_evlist *evlist, bool alloc_raw);
void perf_evlist__free_stats(struct perf_evlist *evlist);
void perf_evlist__reset_stats(struct perf_evlist *evlist);
+void perf_evlist__reset_prev_raw_counts(struct perf_evlist *evlist);
int perf_stat_process_counter(struct perf_stat_config *config,
struct perf_evsel *counter);
--
2.20.1
next prev parent reply other threads:[~2019-10-10 8:53 UTC|newest]
Thread overview: 72+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-10-10 8:36 [PATCH 4.14 00/61] 4.14.149-stable review Greg Kroah-Hartman
2019-10-10 8:36 ` [PATCH 4.14 01/61] s390/process: avoid potential reading of freed stack Greg Kroah-Hartman
2019-10-10 8:36 ` [PATCH 4.14 02/61] KVM: s390: Test for bad access register and size at the start of S390_MEM_OP Greg Kroah-Hartman
2019-10-10 8:36 ` [PATCH 4.14 03/61] s390/topology: avoid firing events before kobjs are created Greg Kroah-Hartman
2019-10-10 8:36 ` [PATCH 4.14 04/61] s390/cio: avoid calling strlen on null pointer Greg Kroah-Hartman
2019-10-10 8:36 ` [PATCH 4.14 05/61] s390/cio: exclude subchannels with no parent from pseudo check Greg Kroah-Hartman
2019-10-10 8:36 ` [PATCH 4.14 06/61] KVM: PPC: Book3S HV: Dont lose pending doorbell request on migration on P9 Greg Kroah-Hartman
2019-10-10 8:36 ` [PATCH 4.14 07/61] KVM: nVMX: handle page fault in vmread fix Greg Kroah-Hartman
2019-10-10 8:36 ` [PATCH 4.14 08/61] PM / devfreq: tegra: Fix kHz to Hz conversion Greg Kroah-Hartman
2019-10-10 8:36 ` [PATCH 4.14 09/61] ASoC: Define a set of DAPM pre/post-up events Greg Kroah-Hartman
2019-10-10 8:36 ` [PATCH 4.14 10/61] powerpc/powernv: Restrict OPAL symbol map to only be readable by root Greg Kroah-Hartman
2019-10-10 8:36 ` [PATCH 4.14 11/61] can: mcp251x: mcp251x_hw_reset(): allow more time after a reset Greg Kroah-Hartman
2019-10-10 8:36 ` [PATCH 4.14 12/61] tools lib traceevent: Fix "robust" test of do_generate_dynamic_list_file Greg Kroah-Hartman
2019-10-10 8:36 ` [PATCH 4.14 13/61] crypto: qat - Silence smp_processor_id() warning Greg Kroah-Hartman
2019-10-10 8:36 ` [PATCH 4.14 14/61] crypto: skcipher - Unmap pages after an external error Greg Kroah-Hartman
2019-10-10 8:36 ` [PATCH 4.14 15/61] crypto: cavium/zip - Add missing single_release() Greg Kroah-Hartman
2019-10-10 8:36 ` [PATCH 4.14 16/61] crypto: caam - fix concurrency issue in givencrypt descriptor Greg Kroah-Hartman
2019-10-10 8:36 ` [PATCH 4.14 17/61] MIPS: Treat Loongson Extensions as ASEs Greg Kroah-Hartman
2019-10-11 4:30 ` Greg Kroah-Hartman
2019-10-10 8:36 ` [PATCH 4.14 18/61] usercopy: Avoid HIGHMEM pfn warning Greg Kroah-Hartman
2019-10-10 8:36 ` [PATCH 4.14 19/61] timer: Read jiffies once when forwarding base clk Greg Kroah-Hartman
2019-10-10 8:36 ` [PATCH 4.14 20/61] watchdog: imx2_wdt: fix min() calculation in imx2_wdt_set_timeout Greg Kroah-Hartman
2019-10-10 8:36 ` [PATCH 4.14 21/61] drm/omap: fix max fclk divider for omap36xx Greg Kroah-Hartman
2019-10-10 8:36 ` [PATCH 4.14 22/61] mmc: sdhci: improve ADMA error reporting Greg Kroah-Hartman
2019-10-10 8:36 ` [PATCH 4.14 23/61] mmc: sdhci-of-esdhc: set DMA snooping based on DMA coherence Greg Kroah-Hartman
2019-10-10 8:36 ` [PATCH 4.14 24/61] Revert "locking/pvqspinlock: Dont wait if vCPU is preempted" Greg Kroah-Hartman
2019-10-10 8:36 ` [PATCH 4.14 25/61] xen/xenbus: fix self-deadlock after killing user process Greg Kroah-Hartman
2019-10-10 8:36 ` [PATCH 4.14 26/61] ieee802154: atusb: fix use-after-free at disconnect Greg Kroah-Hartman
2019-10-10 8:36 ` [PATCH 4.14 27/61] cfg80211: initialize on-stack chandefs Greg Kroah-Hartman
2019-10-10 8:36 ` [PATCH 4.14 28/61] ima: always return negative code for error Greg Kroah-Hartman
2019-10-10 8:36 ` [PATCH 4.14 29/61] fs: nfs: Fix possible null-pointer dereferences in encode_attrs() Greg Kroah-Hartman
2019-10-10 8:36 ` [PATCH 4.14 30/61] 9p: avoid attaching writeback_fid on mmap with type PRIVATE Greg Kroah-Hartman
2019-10-10 8:36 ` [PATCH 4.14 31/61] xen/pci: reserve MCFG areas earlier Greg Kroah-Hartman
2019-10-10 8:36 ` [PATCH 4.14 32/61] ceph: fix directories inode i_blkbits initialization Greg Kroah-Hartman
2019-10-10 8:36 ` [PATCH 4.14 33/61] ceph: reconnect connection if session hang in opening state Greg Kroah-Hartman
2019-10-10 8:36 ` [PATCH 4.14 34/61] watchdog: aspeed: Add support for AST2600 Greg Kroah-Hartman
2019-10-10 8:37 ` [PATCH 4.14 35/61] netfilter: nf_tables: allow lookups in dynamic sets Greg Kroah-Hartman
2019-10-10 8:37 ` [PATCH 4.14 36/61] drm/amdgpu: Check for valid number of registers to read Greg Kroah-Hartman
2019-10-10 8:37 ` [PATCH 4.14 37/61] pNFS: Ensure we do clear the return-on-close layout stateid on fatal errors Greg Kroah-Hartman
2019-10-10 8:37 ` [PATCH 4.14 38/61] pwm: stm32-lp: Add check in case requested period cannot be achieved Greg Kroah-Hartman
2019-10-10 8:37 ` [PATCH 4.14 39/61] thermal: Fix use-after-free when unregistering thermal zone device Greg Kroah-Hartman
2019-10-10 8:37 ` [PATCH 4.14 40/61] fuse: fix memleak in cuse_channel_open Greg Kroah-Hartman
2019-10-10 8:37 ` [PATCH 4.14 41/61] sched/core: Fix migration to invalid CPU in __set_cpus_allowed_ptr() Greg Kroah-Hartman
2019-10-10 8:37 ` [PATCH 4.14 42/61] perf build: Add detection of java-11-openjdk-devel package Greg Kroah-Hartman
2019-10-10 8:37 ` [PATCH 4.14 43/61] kernel/elfcore.c: include proper prototypes Greg Kroah-Hartman
2019-10-10 8:37 ` [PATCH 4.14 44/61] perf unwind: Fix libunwind build failure on i386 systems Greg Kroah-Hartman
2019-10-10 8:37 ` [PATCH 4.14 45/61] KVM: PPC: Book3S HV: XIVE: Free escalation interrupts before disabling the VP Greg Kroah-Hartman
2019-10-10 8:37 ` [PATCH 4.14 46/61] nbd: fix crash when the blksize is zero Greg Kroah-Hartman
2019-10-10 8:37 ` [PATCH 4.14 47/61] block/ndb: add WQ_UNBOUND to the knbd-recv workqueue Greg Kroah-Hartman
2019-10-10 8:37 ` [PATCH 4.14 48/61] nbd: fix max number of supported devs Greg Kroah-Hartman
2019-10-10 8:37 ` [PATCH 4.14 49/61] powerpc/pseries: Fix cpu_hotplug_lock acquisition in resize_hpt() Greg Kroah-Hartman
2019-10-10 8:37 ` [PATCH 4.14 50/61] tools lib traceevent: Do not free tep->cmdlines in add_new_comm() on failure Greg Kroah-Hartman
2019-10-10 8:37 ` [PATCH 4.14 51/61] tick: broadcast-hrtimer: Fix a race in bc_set_next Greg Kroah-Hartman
2019-10-10 8:37 ` [PATCH 4.14 53/61] perf stat: Fix a segmentation fault when using repeat forever Greg Kroah-Hartman
2019-10-10 8:37 ` Greg Kroah-Hartman [this message]
2019-10-10 8:37 ` [PATCH 4.14 55/61] drm/i915/userptr: Acquire the page lock around set_page_dirty() Greg Kroah-Hartman
2019-10-10 8:37 ` [PATCH 4.14 56/61] vfs: Fix EOVERFLOW testing in put_compat_statfs64 Greg Kroah-Hartman
2019-10-10 8:37 ` [PATCH 4.14 57/61] coresight: etm4x: Use explicit barriers on enable/disable Greg Kroah-Hartman
2019-10-10 8:37 ` [PATCH 4.14 58/61] cfg80211: add and use strongly typed element iteration macros Greg Kroah-Hartman
2019-10-10 8:37 ` [PATCH 4.14 59/61] cfg80211: Use const more consistently in for_each_element macros Greg Kroah-Hartman
2019-10-10 8:37 ` [PATCH 4.14 60/61] nl80211: validate beacon head Greg Kroah-Hartman
2019-10-10 8:37 ` [PATCH 4.14 61/61] ASoC: sgtl5000: Improve VAG power and mute control Greg Kroah-Hartman
2019-10-10 13:27 ` [PATCH 4.14 00/61] 4.14.149-stable review Naresh Kamboju
2019-10-10 14:21 ` kernelci.org bot
2019-10-10 17:12 ` Guenter Roeck
2019-10-11 4:29 ` Greg Kroah-Hartman
2019-10-11 13:14 ` Guenter Roeck
2019-10-11 14:01 ` Greg Kroah-Hartman
2019-10-10 22:18 ` Guenter Roeck
2019-10-10 23:50 ` Didik Setiawan
2019-10-11 3:09 ` shuah
2019-10-11 8:33 ` Jon Hunter
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=20191010083523.606396136@linuxfoundation.org \
--to=gregkh@linuxfoundation.org \
--cc=acme@redhat.com \
--cc=eranian@google.com \
--cc=jolsa@kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=namhyung@kernel.org \
--cc=naveen.n.rao@linux.vnet.ibm.com \
--cc=ravi.bangoria@linux.ibm.com \
--cc=sashal@kernel.org \
--cc=srikar@linux.vnet.ibm.com \
--cc=stable@vger.kernel.org \
/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).