From: Arnaldo Carvalho de Melo <acme@kernel.org>
To: Ingo Molnar <mingo@kernel.org>
Cc: linux-kernel@vger.kernel.org,
Adrian Hunter <adrian.hunter@intel.com>,
Andi Kleen <ak@linux.intel.com>,
stable@vger.kernel.org,
Arnaldo Carvalho de Melo <acme@redhat.com>
Subject: [PATCH 08/25] perf intel-pt: Improve sample timestamp
Date: Wed, 21 Jun 2017 15:02:28 -0300 [thread overview]
Message-ID: <20170621180245.23134-9-acme@kernel.org> (raw)
In-Reply-To: <20170621180245.23134-1-acme@kernel.org>
From: Adrian Hunter <adrian.hunter@intel.com>
The decoder uses its current timestamp in samples. Usually that is a
timestamp that has already passed, but in some cases it is a timestamp
for a branch that the decoder is walking towards, and consequently
hasn't reached. Improve that situation by using the pkt_state to
determine when to use the current or previous timestamp.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: stable@vger.kernel.org
Link: http://lkml.kernel.org/r/1495786658-18063-3-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
.../perf/util/intel-pt-decoder/intel-pt-decoder.c | 34 ++++++++++++++++++++--
1 file changed, 31 insertions(+), 3 deletions(-)
diff --git a/tools/perf/util/intel-pt-decoder/intel-pt-decoder.c b/tools/perf/util/intel-pt-decoder/intel-pt-decoder.c
index 5a9676c6e23f..d5c69e822282 100644
--- a/tools/perf/util/intel-pt-decoder/intel-pt-decoder.c
+++ b/tools/perf/util/intel-pt-decoder/intel-pt-decoder.c
@@ -64,6 +64,25 @@ enum intel_pt_pkt_state {
INTEL_PT_STATE_FUP_NO_TIP,
};
+static inline bool intel_pt_sample_time(enum intel_pt_pkt_state pkt_state)
+{
+ switch (pkt_state) {
+ case INTEL_PT_STATE_NO_PSB:
+ case INTEL_PT_STATE_NO_IP:
+ case INTEL_PT_STATE_ERR_RESYNC:
+ case INTEL_PT_STATE_IN_SYNC:
+ case INTEL_PT_STATE_TNT:
+ return true;
+ case INTEL_PT_STATE_TIP:
+ case INTEL_PT_STATE_TIP_PGD:
+ case INTEL_PT_STATE_FUP:
+ case INTEL_PT_STATE_FUP_NO_TIP:
+ return false;
+ default:
+ return true;
+ };
+}
+
#ifdef INTEL_PT_STRICT
#define INTEL_PT_STATE_ERR1 INTEL_PT_STATE_NO_PSB
#define INTEL_PT_STATE_ERR2 INTEL_PT_STATE_NO_PSB
@@ -99,6 +118,7 @@ struct intel_pt_decoder {
uint64_t timestamp;
uint64_t tsc_timestamp;
uint64_t ref_timestamp;
+ uint64_t sample_timestamp;
uint64_t ret_addr;
uint64_t ctc_timestamp;
uint64_t ctc_delta;
@@ -139,6 +159,7 @@ struct intel_pt_decoder {
unsigned int fup_tx_flags;
unsigned int tx_flags;
uint64_t timestamp_insn_cnt;
+ uint64_t sample_insn_cnt;
uint64_t stuck_ip;
int no_progress;
int stuck_ip_prd;
@@ -898,6 +919,7 @@ static int intel_pt_walk_insn(struct intel_pt_decoder *decoder,
decoder->tot_insn_cnt += insn_cnt;
decoder->timestamp_insn_cnt += insn_cnt;
+ decoder->sample_insn_cnt += insn_cnt;
decoder->period_insn_cnt += insn_cnt;
if (err) {
@@ -2069,7 +2091,7 @@ static int intel_pt_sync(struct intel_pt_decoder *decoder)
static uint64_t intel_pt_est_timestamp(struct intel_pt_decoder *decoder)
{
- uint64_t est = decoder->timestamp_insn_cnt << 1;
+ uint64_t est = decoder->sample_insn_cnt << 1;
if (!decoder->cbr || !decoder->max_non_turbo_ratio)
goto out;
@@ -2077,7 +2099,7 @@ static uint64_t intel_pt_est_timestamp(struct intel_pt_decoder *decoder)
est *= decoder->max_non_turbo_ratio;
est /= decoder->cbr;
out:
- return decoder->timestamp + est;
+ return decoder->sample_timestamp + est;
}
const struct intel_pt_state *intel_pt_decode(struct intel_pt_decoder *decoder)
@@ -2133,11 +2155,17 @@ const struct intel_pt_state *intel_pt_decode(struct intel_pt_decoder *decoder)
if (err) {
decoder->state.err = intel_pt_ext_err(err);
decoder->state.from_ip = decoder->ip;
+ decoder->sample_timestamp = decoder->timestamp;
+ decoder->sample_insn_cnt = decoder->timestamp_insn_cnt;
} else {
decoder->state.err = 0;
+ if (intel_pt_sample_time(decoder->pkt_state)) {
+ decoder->sample_timestamp = decoder->timestamp;
+ decoder->sample_insn_cnt = decoder->timestamp_insn_cnt;
+ }
}
- decoder->state.timestamp = decoder->timestamp;
+ decoder->state.timestamp = decoder->sample_timestamp;
decoder->state.est_timestamp = intel_pt_est_timestamp(decoder);
decoder->state.cr3 = decoder->cr3;
decoder->state.tot_insn_cnt = decoder->tot_insn_cnt;
--
2.9.4
next prev parent reply other threads:[~2017-06-21 18:07 UTC|newest]
Thread overview: 27+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-06-21 18:02 [GIT PULL 00/25] perf/core improvements and fixes Arnaldo Carvalho de Melo
2017-06-21 18:02 ` [PATCH 01/25] perf evsel: Adopt find_process() Arnaldo Carvalho de Melo
2017-06-21 18:02 ` [PATCH 02/25] perf tools: Do parameter validation earlier on fetch_kernel_version() Arnaldo Carvalho de Melo
2017-06-21 18:02 ` [PATCH 03/25] perf tools: Remove unused _ALL_SOURCE define Arnaldo Carvalho de Melo
2017-06-21 18:02 ` [PATCH 04/25] tools lib api fs: Add sysfs__write_int function Arnaldo Carvalho de Melo
2017-06-21 18:02 ` [PATCH 05/25] perf stat: Add support to measure SMI cost Arnaldo Carvalho de Melo
2017-06-21 18:02 ` [PATCH 06/25] perf unwind: Support for powerpc Arnaldo Carvalho de Melo
2017-06-21 18:02 ` [PATCH 07/25] perf intel-pt: Move decoder error setting into one condition Arnaldo Carvalho de Melo
2017-06-21 18:02 ` Arnaldo Carvalho de Melo [this message]
2017-06-21 18:02 ` [PATCH 09/25] perf intel-pt: Fix missing stack clear Arnaldo Carvalho de Melo
2017-06-21 18:02 ` [PATCH 10/25] perf intel-pt: Ensure IP is zero when state is INTEL_PT_STATE_NO_IP Arnaldo Carvalho de Melo
2017-06-21 18:02 ` [PATCH 11/25] perf intel-pt: Fix last_ip usage Arnaldo Carvalho de Melo
2017-06-21 18:02 ` [PATCH 12/25] perf intel-pt: Ensure never to set 'last_ip' when packet 'count' is zero Arnaldo Carvalho de Melo
2017-06-21 18:02 ` [PATCH 13/25] perf intel-pt: Use FUP always when scanning for an IP Arnaldo Carvalho de Melo
2017-06-21 18:02 ` [PATCH 14/25] perf intel-pt: Clear FUP flag on error Arnaldo Carvalho de Melo
2017-06-21 18:02 ` [PATCH 15/25] perf intel-pt: Add missing __fallthrough Arnaldo Carvalho de Melo
2017-06-21 18:02 ` [PATCH 16/25] perf intel-pt: Allow decoding with branch tracing disabled Arnaldo Carvalho de Melo
2017-06-21 18:02 ` [PATCH 17/25] perf intel-pt: Add default config for pass-through branch enable Arnaldo Carvalho de Melo
2017-06-21 18:02 ` [PATCH 18/25] perf intel-pt: Add documentation for new config terms Arnaldo Carvalho de Melo
2017-06-21 18:02 ` [PATCH 19/25] perf intel-pt: Add decoder support for ptwrite and power event packets Arnaldo Carvalho de Melo
2017-06-21 18:02 ` [PATCH 20/25] perf intel-pt: Add reserved byte to CBR packet payload Arnaldo Carvalho de Melo
2017-06-21 18:02 ` [PATCH 21/25] perf intel-pt: Add decoder support for CBR events Arnaldo Carvalho de Melo
2017-06-21 18:02 ` [PATCH 22/25] perf intel-pt: Remove redundant initial_skip checks Arnaldo Carvalho de Melo
2017-06-21 18:02 ` [PATCH 23/25] perf intel-pt: Fix transactions_sample_type Arnaldo Carvalho de Melo
2017-06-21 18:02 ` [PATCH 24/25] perf tools: Fix message because cpu list option is -C not -c Arnaldo Carvalho de Melo
2017-06-21 18:02 ` [PATCH 25/25] perf script: Fix message because field list option is -F not -f Arnaldo Carvalho de Melo
2017-06-21 18:13 ` [GIT PULL 00/25] 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=20170621180245.23134-9-acme@kernel.org \
--to=acme@kernel.org \
--cc=acme@redhat.com \
--cc=adrian.hunter@intel.com \
--cc=ak@linux.intel.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@kernel.org \
--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