From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
stable@vger.kernel.org, Jin Yao <yao.jin@linux.intel.com>,
Alexander Shishkin <alexander.shishkin@linux.intel.com>,
Andi Kleen <ak@linux.intel.com>, Jiri Olsa <jolsa@kernel.org>,
Kan Liang <kan.liang@linux.intel.com>,
Peter Zijlstra <peterz@infradead.org>,
Arnaldo Carvalho de Melo <acme@redhat.com>,
Sasha Levin <sashal@kernel.org>
Subject: [PATCH 4.19 23/85] perf report: Fix wrong iteration count in --branch-history
Date: Mon, 18 Feb 2019 14:42:49 +0100 [thread overview]
Message-ID: <20190218133502.144952478@linuxfoundation.org> (raw)
In-Reply-To: <20190218133459.758004711@linuxfoundation.org>
4.19-stable review patch. If anyone has any objections, please let me know.
------------------
[ Upstream commit a3366db06bb656cef2e03f30f780d93059bcc594 ]
By calculating the removed loops, we can get the iteration count.
But the iteration count could be reported incorrectly, reporting
impossibly high counts.
That's because previous code uses the number of removed LBR entries for
the iteration count. That's not good. Fix this by increasing the
iteration count when a loop is detected.
When matching the chain, the iteration count would be added up, finally we need
to compute the average value when printing out.
For example,
$ perf report --branch-history --stdio --no-children
Before:
---f2 +0
|
|--33.62%--f1 +9 (cycles:1)
| f1 +0
| main +22 (cycles:1)
| main +17
| main +38 (cycles:1)
| main +27
| f1 +26 (cycles:1)
| f1 +24
| f2 +27 (cycles:7)
| f2 +0
| f1 +19 (cycles:1)
| f1 +14
| f2 +27 (cycles:11)
| f2 +0
| f1 +9 (cycles:1 iter:2968 avg_cycles:3)
| f1 +0
| main +22 (cycles:1 iter:2968 avg_cycles:3)
| main +17
| main +38 (cycles:1 iter:2968 avg_cycles:3)
2968 is an impossible high iteration count and avg_cycles is too small.
After:
---f2 +0
|
|--33.62%--f1 +9 (cycles:1)
| f1 +0
| main +22 (cycles:1)
| main +17
| main +38 (cycles:1)
| main +27
| f1 +26 (cycles:1)
| f1 +24
| f2 +27 (cycles:7)
| f2 +0
| f1 +19 (cycles:1)
| f1 +14
| f2 +27 (cycles:11)
| f2 +0
| f1 +9 (cycles:1 iter:1 avg_cycles:23)
| f1 +0
| main +22 (cycles:1 iter:1 avg_cycles:23)
| main +17
| main +38 (cycles:1 iter:1 avg_cycles:23)
avg_cycles:23 is the average cycles of this iteration.
Fixes: c4ee06251d42 ("perf report: Calculate the average cycles of iterations")
Signed-off-by: Jin Yao <yao.jin@linux.intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1546582230-17507-1-git-send-email-yao.jin@linux.intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
tools/perf/util/callchain.c | 32 ++++++++++++++++++++------------
tools/perf/util/callchain.h | 1 +
tools/perf/util/machine.c | 2 +-
3 files changed, 22 insertions(+), 13 deletions(-)
diff --git a/tools/perf/util/callchain.c b/tools/perf/util/callchain.c
index 32ef7bdca1cf..dc2212e12184 100644
--- a/tools/perf/util/callchain.c
+++ b/tools/perf/util/callchain.c
@@ -766,6 +766,7 @@ static enum match_result match_chain(struct callchain_cursor_node *node,
cnode->cycles_count += node->branch_flags.cycles;
cnode->iter_count += node->nr_loop_iter;
cnode->iter_cycles += node->iter_cycles;
+ cnode->from_count++;
}
}
@@ -1345,10 +1346,10 @@ static int branch_to_str(char *bf, int bfsize,
static int branch_from_str(char *bf, int bfsize,
u64 branch_count,
u64 cycles_count, u64 iter_count,
- u64 iter_cycles)
+ u64 iter_cycles, u64 from_count)
{
int printed = 0, i = 0;
- u64 cycles;
+ u64 cycles, v = 0;
cycles = cycles_count / branch_count;
if (cycles) {
@@ -1357,14 +1358,16 @@ static int branch_from_str(char *bf, int bfsize,
bf + printed, bfsize - printed);
}
- if (iter_count) {
- printed += count_pri64_printf(i++, "iter",
- iter_count,
- bf + printed, bfsize - printed);
+ if (iter_count && from_count) {
+ v = iter_count / from_count;
+ if (v) {
+ printed += count_pri64_printf(i++, "iter",
+ v, bf + printed, bfsize - printed);
- printed += count_pri64_printf(i++, "avg_cycles",
- iter_cycles / iter_count,
- bf + printed, bfsize - printed);
+ printed += count_pri64_printf(i++, "avg_cycles",
+ iter_cycles / iter_count,
+ bf + printed, bfsize - printed);
+ }
}
if (i)
@@ -1377,6 +1380,7 @@ static int counts_str_build(char *bf, int bfsize,
u64 branch_count, u64 predicted_count,
u64 abort_count, u64 cycles_count,
u64 iter_count, u64 iter_cycles,
+ u64 from_count,
struct branch_type_stat *brtype_stat)
{
int printed;
@@ -1389,7 +1393,8 @@ static int counts_str_build(char *bf, int bfsize,
predicted_count, abort_count, brtype_stat);
} else {
printed = branch_from_str(bf, bfsize, branch_count,
- cycles_count, iter_count, iter_cycles);
+ cycles_count, iter_count, iter_cycles,
+ from_count);
}
if (!printed)
@@ -1402,13 +1407,14 @@ static int callchain_counts_printf(FILE *fp, char *bf, int bfsize,
u64 branch_count, u64 predicted_count,
u64 abort_count, u64 cycles_count,
u64 iter_count, u64 iter_cycles,
+ u64 from_count,
struct branch_type_stat *brtype_stat)
{
char str[256];
counts_str_build(str, sizeof(str), branch_count,
predicted_count, abort_count, cycles_count,
- iter_count, iter_cycles, brtype_stat);
+ iter_count, iter_cycles, from_count, brtype_stat);
if (fp)
return fprintf(fp, "%s", str);
@@ -1422,6 +1428,7 @@ int callchain_list_counts__printf_value(struct callchain_list *clist,
u64 branch_count, predicted_count;
u64 abort_count, cycles_count;
u64 iter_count, iter_cycles;
+ u64 from_count;
branch_count = clist->branch_count;
predicted_count = clist->predicted_count;
@@ -1429,11 +1436,12 @@ int callchain_list_counts__printf_value(struct callchain_list *clist,
cycles_count = clist->cycles_count;
iter_count = clist->iter_count;
iter_cycles = clist->iter_cycles;
+ from_count = clist->from_count;
return callchain_counts_printf(fp, bf, bfsize, branch_count,
predicted_count, abort_count,
cycles_count, iter_count, iter_cycles,
- &clist->brtype_stat);
+ from_count, &clist->brtype_stat);
}
static void free_callchain_node(struct callchain_node *node)
diff --git a/tools/perf/util/callchain.h b/tools/perf/util/callchain.h
index 154560b1eb65..99d38ac019b8 100644
--- a/tools/perf/util/callchain.h
+++ b/tools/perf/util/callchain.h
@@ -118,6 +118,7 @@ struct callchain_list {
bool has_children;
};
u64 branch_count;
+ u64 from_count;
u64 predicted_count;
u64 abort_count;
u64 cycles_count;
diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c
index d7403d1207d7..b1508ce3e412 100644
--- a/tools/perf/util/machine.c
+++ b/tools/perf/util/machine.c
@@ -1988,7 +1988,7 @@ static void save_iterations(struct iterations *iter,
{
int i;
- iter->nr_loop_iter = nr;
+ iter->nr_loop_iter++;
iter->cycles = 0;
for (i = 0; i < nr; i++)
--
2.19.1
next prev parent reply other threads:[~2019-02-18 13:51 UTC|newest]
Thread overview: 90+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-02-18 13:42 [PATCH 4.19 00/85] 4.19.24-stable review Greg Kroah-Hartman
2019-02-18 13:42 ` [PATCH 4.19 01/85] dt-bindings: eeprom: at24: add "atmel,24c2048" compatible string Greg Kroah-Hartman
2019-02-18 13:42 ` [PATCH 4.19 02/85] eeprom: at24: add support for 24c2048 Greg Kroah-Hartman
2019-02-18 13:42 ` [PATCH 4.19 03/85] blk-mq: fix a hung issue when fsync Greg Kroah-Hartman
2019-02-18 13:42 ` [PATCH 4.19 04/85] ARM: 8789/1: signal: copy registers using __copy_to_user() Greg Kroah-Hartman
2019-02-18 13:42 ` [PATCH 4.19 05/85] ARM: 8790/1: signal: always use __copy_to_user to save iwmmxt context Greg Kroah-Hartman
2019-02-18 13:42 ` [PATCH 4.19 06/85] ARM: 8791/1: vfp: use __copy_to_user() when saving VFP state Greg Kroah-Hartman
2019-02-18 13:42 ` [PATCH 4.19 07/85] ARM: 8792/1: oabi-compat: copy oabi events using __copy_to_user() Greg Kroah-Hartman
2019-02-18 13:42 ` [PATCH 4.19 08/85] ARM: 8793/1: signal: replace __put_user_error with __put_user Greg Kroah-Hartman
2019-02-18 13:42 ` [PATCH 4.19 09/85] ARM: 8794/1: uaccess: Prevent speculative use of the current addr_limit Greg Kroah-Hartman
2019-02-18 13:42 ` [PATCH 4.19 10/85] ARM: 8795/1: spectre-v1.1: use put_user() for __put_user() Greg Kroah-Hartman
2019-02-18 13:42 ` [PATCH 4.19 11/85] ARM: 8796/1: spectre-v1,v1.1: provide helpers for address sanitization Greg Kroah-Hartman
2019-02-18 13:42 ` [PATCH 4.19 12/85] ARM: 8797/1: spectre-v1.1: harden __copy_to_user Greg Kroah-Hartman
2019-02-18 13:42 ` [PATCH 4.19 13/85] ARM: 8810/1: vfp: Fix wrong assignement to ufp_exc Greg Kroah-Hartman
2019-02-18 13:42 ` [PATCH 4.19 14/85] ARM: make lookup_processor_type() non-__init Greg Kroah-Hartman
2019-02-18 13:42 ` [PATCH 4.19 15/85] ARM: split out processor lookup Greg Kroah-Hartman
2019-02-18 13:42 ` [PATCH 4.19 16/85] ARM: clean up per-processor check_bugs method call Greg Kroah-Hartman
2019-02-18 13:42 ` [PATCH 4.19 17/85] ARM: add PROC_VTABLE and PROC_TABLE macros Greg Kroah-Hartman
2019-02-18 13:42 ` [PATCH 4.19 18/85] ARM: spectre-v2: per-CPU vtables to work around big.Little systems Greg Kroah-Hartman
2019-02-18 13:42 ` [PATCH 4.19 19/85] ARM: ensure that processor vtables is not lost after boot Greg Kroah-Hartman
2019-02-18 13:42 ` [PATCH 4.19 20/85] ARM: fix the cockup in the previous patch Greg Kroah-Hartman
2019-02-18 13:42 ` [PATCH 4.19 21/85] drm/amdgpu/sriov:Correct pfvf exchange logic Greg Kroah-Hartman
2019-02-18 13:42 ` [PATCH 4.19 22/85] ACPI: NUMA: Use correct type for printing addresses on i386-PAE Greg Kroah-Hartman
2019-02-18 13:42 ` Greg Kroah-Hartman [this message]
2019-02-18 13:42 ` [PATCH 4.19 24/85] perf test shell: Use a fallback to get the pathname in vfs_getname Greg Kroah-Hartman
2019-02-18 13:42 ` [PATCH 4.19 25/85] tools uapi: fix RISC-V 64-bit support Greg Kroah-Hartman
2019-02-18 13:42 ` [PATCH 4.19 26/85] riscv: fix trace_sys_exit hook Greg Kroah-Hartman
2019-02-18 13:42 ` [PATCH 4.19 27/85] cpufreq: check if policy is inactive early in __cpufreq_get() Greg Kroah-Hartman
2019-02-18 13:42 ` [PATCH 4.19 28/85] drm/bridge: tc358767: add bus flags Greg Kroah-Hartman
2019-02-18 13:42 ` [PATCH 4.19 29/85] drm/bridge: tc358767: add defines for DP1_SRCCTRL & PHY_2LANE Greg Kroah-Hartman
2019-02-18 13:42 ` [PATCH 4.19 30/85] drm/bridge: tc358767: fix single lane configuration Greg Kroah-Hartman
2019-02-18 13:42 ` [PATCH 4.19 31/85] drm/bridge: tc358767: fix initial DP0/1_SRCCTRL value Greg Kroah-Hartman
2019-02-18 13:42 ` [PATCH 4.19 32/85] drm/bridge: tc358767: reject modes which require too much BW Greg Kroah-Hartman
2019-02-18 13:42 ` [PATCH 4.19 33/85] drm/bridge: tc358767: fix output H/V syncs Greg Kroah-Hartman
2019-02-18 13:43 ` [PATCH 4.19 34/85] nvme-pci: use the same attributes when freeing host_mem_desc_bufs Greg Kroah-Hartman
2019-02-18 13:43 ` [PATCH 4.19 35/85] nvme-pci: fix out of bounds access in nvme_cqe_pending Greg Kroah-Hartman
2019-02-18 13:43 ` [PATCH 4.19 36/85] nvme-multipath: zero out ANA log buffer Greg Kroah-Hartman
2019-02-18 13:43 ` [PATCH 4.19 37/85] nvme: pad fake subsys NQN vid and ssvid with zeros Greg Kroah-Hartman
2019-02-18 13:43 ` [PATCH 4.19 38/85] drm/amdgpu: set WRITE_BURST_LENGTH to 64B to workaround SDMA1 hang Greg Kroah-Hartman
2019-02-18 13:43 ` [PATCH 4.19 39/85] ARM: dts: da850-evm: Correct the audio codec regulators Greg Kroah-Hartman
2019-02-18 13:43 ` [PATCH 4.19 40/85] ARM: dts: da850-evm: Correct the sound card name Greg Kroah-Hartman
2019-02-18 13:43 ` [PATCH 4.19 41/85] ARM: dts: da850-lcdk: Correct the audio codec regulators Greg Kroah-Hartman
2019-02-18 13:43 ` [PATCH 4.19 42/85] ARM: dts: da850-lcdk: Correct the sound card name Greg Kroah-Hartman
2019-02-18 13:43 ` [PATCH 4.19 43/85] ARM: dts: kirkwood: Fix polarity of GPIO fan lines Greg Kroah-Hartman
2019-02-18 13:43 ` [PATCH 4.19 44/85] gpio: pl061: handle failed allocations Greg Kroah-Hartman
2019-02-18 13:43 ` [PATCH 4.19 45/85] drm/nouveau: Dont disable polling in fallback mode Greg Kroah-Hartman
2019-02-18 13:43 ` [PATCH 4.19 46/85] drm/nouveau/falcon: avoid touching registers if engine is off Greg Kroah-Hartman
2019-02-18 13:43 ` [PATCH 4.19 47/85] cifs: Limit memory used by lock request calls to a page Greg Kroah-Hartman
2019-02-18 13:43 ` [PATCH 4.19 48/85] kvm: sev: Fail KVM_SEV_INIT if already initialized Greg Kroah-Hartman
2019-02-18 13:43 ` [PATCH 4.19 49/85] CIFS: Do not assume one credit for async responses Greg Kroah-Hartman
2019-02-18 13:43 ` [PATCH 4.19 50/85] gpio: mxc: move gpio noirq suspend/resume to syscore phase Greg Kroah-Hartman
2019-02-18 13:43 ` [PATCH 4.19 51/85] Revert "Input: elan_i2c - add ACPI ID for touchpad in ASUS Aspire F5-573G" Greg Kroah-Hartman
2019-02-18 13:43 ` [PATCH 4.19 52/85] Input: elan_i2c - add ACPI ID for touchpad in Lenovo V330-15ISK Greg Kroah-Hartman
2019-02-18 13:43 ` [PATCH 4.19 53/85] ARM: OMAP5+: Fix inverted nirq pin interrupts with irq_set_type Greg Kroah-Hartman
2019-02-18 13:43 ` [PATCH 4.19 54/85] perf/core: Fix impossible ring-buffer sizes warning Greg Kroah-Hartman
2019-02-18 13:43 ` [PATCH 4.19 55/85] perf/x86: Add check_period PMU callback Greg Kroah-Hartman
2019-02-18 13:43 ` [PATCH 4.19 56/85] ALSA: hda - Add quirk for HP EliteBook 840 G5 Greg Kroah-Hartman
2019-02-18 13:43 ` [PATCH 4.19 57/85] ALSA: usb-audio: Fix implicit fb endpoint setup by quirk Greg Kroah-Hartman
2019-02-18 13:43 ` [PATCH 4.19 58/85] ASoC: hdmi-codec: fix oops on re-probe Greg Kroah-Hartman
2019-02-18 13:43 ` [PATCH 4.19 59/85] tools uapi: fix Alpha support Greg Kroah-Hartman
2019-02-18 13:43 ` [PATCH 4.19 60/85] riscv: Add pte bit to distinguish swap from invalid Greg Kroah-Hartman
2019-02-18 13:43 ` [PATCH 4.19 61/85] x86/kvm/nVMX: read from MSR_IA32_VMX_PROCBASED_CTLS2 only when it is available Greg Kroah-Hartman
2019-02-18 13:43 ` [PATCH 4.19 62/85] kvm: vmx: Fix entry number check for add_atomic_switch_msr() Greg Kroah-Hartman
2019-02-18 13:43 ` [PATCH 4.19 63/85] mmc: sunxi: Filter out unsupported modes declared in the device tree Greg Kroah-Hartman
2019-02-18 13:43 ` [PATCH 4.19 64/85] mmc: block: handle complete_work on separate workqueue Greg Kroah-Hartman
2019-02-18 13:43 ` [PATCH 4.19 65/85] Input: bma150 - register input device after setting private data Greg Kroah-Hartman
2019-02-18 13:43 ` [PATCH 4.19 66/85] Input: elantech - enable 3rd button support on Fujitsu CELSIUS H780 Greg Kroah-Hartman
2019-02-18 13:43 ` [PATCH 4.19 67/85] Revert "nfsd4: return default lease period" Greg Kroah-Hartman
2019-02-18 13:43 ` [PATCH 4.19 68/85] Revert "mm: dont reclaim inodes with many attached pages" Greg Kroah-Hartman
2019-02-18 13:43 ` [PATCH 4.19 69/85] Revert "mm: slowly shrink slabs with a relatively small number of objects" Greg Kroah-Hartman
2019-02-18 13:43 ` [PATCH 4.19 70/85] alpha: fix page fault handling for r16-r18 targets Greg Kroah-Hartman
2019-02-18 13:43 ` [PATCH 4.19 71/85] alpha: Fix Eiger NR_IRQS to 128 Greg Kroah-Hartman
2019-02-18 13:43 ` [PATCH 4.19 72/85] s390/zcrypt: fix specification exception on z196 during ap probe Greg Kroah-Hartman
2019-02-18 13:43 ` [PATCH 4.19 73/85] tracing/uprobes: Fix output for multiple string arguments Greg Kroah-Hartman
2019-02-18 13:43 ` [PATCH 4.19 74/85] x86/platform/UV: Use efi_runtime_lock to serialise BIOS calls Greg Kroah-Hartman
2019-02-18 13:43 ` [PATCH 4.19 75/85] scsi: sd: fix entropy gathering for most rotational disks Greg Kroah-Hartman
2019-02-18 13:43 ` [PATCH 4.19 76/85] signal: Restore the stop PTRACE_EVENT_EXIT Greg Kroah-Hartman
2019-02-18 13:43 ` [PATCH 4.19 77/85] md/raid1: dont clear bitmap bits on interrupted recovery Greg Kroah-Hartman
2019-02-18 13:43 ` [PATCH 4.19 78/85] x86/a.out: Clear the dump structure initially Greg Kroah-Hartman
2019-02-18 13:43 ` [PATCH 4.19 79/85] dm crypt: dont overallocate the integrity tag space Greg Kroah-Hartman
2019-02-18 13:43 ` [PATCH 4.19 80/85] dm thin: fix bug where bio that overwrites thin block ignores FUA Greg Kroah-Hartman
2019-02-18 13:43 ` [PATCH 4.19 81/85] drm: Use array_size() when creating lease Greg Kroah-Hartman
2019-02-18 13:43 ` [PATCH 4.19 82/85] drm/vkms: Fix license inconsistent Greg Kroah-Hartman
2019-02-18 13:43 ` [PATCH 4.19 83/85] drm/i915: Block fbdev HPD processing during suspend Greg Kroah-Hartman
2019-02-18 13:43 ` Greg Kroah-Hartman
2019-02-18 13:43 ` [PATCH 4.19 84/85] drm/i915: Prevent a race during I915_GEM_MMAP ioctl with WC set Greg Kroah-Hartman
2019-02-18 13:43 ` [PATCH 4.19 85/85] mm: proc: smaps_rollup: fix pss_locked calculation Greg Kroah-Hartman
2019-02-19 5:41 ` [PATCH 4.19 00/85] 4.19.24-stable review Naresh Kamboju
2019-02-19 17:27 ` Guenter Roeck
2019-02-20 0:17 ` shuah
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=20190218133502.144952478@linuxfoundation.org \
--to=gregkh@linuxfoundation.org \
--cc=acme@redhat.com \
--cc=ak@linux.intel.com \
--cc=alexander.shishkin@linux.intel.com \
--cc=jolsa@kernel.org \
--cc=kan.liang@linux.intel.com \
--cc=linux-kernel@vger.kernel.org \
--cc=peterz@infradead.org \
--cc=sashal@kernel.org \
--cc=stable@vger.kernel.org \
--cc=yao.jin@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.