From: tip-bot for Adrian Hunter <tipbot@zytor.com>
To: linux-tip-commits@vger.kernel.org
Cc: leo.yan@linaro.org, tglx@linutronix.de, adrian.hunter@intel.com,
davem@davemloft.net, linux-kernel@vger.kernel.org,
jolsa@kernel.org, acme@redhat.com, hpa@zytor.com,
mingo@kernel.org, ak@linux.intel.com, mathieu.poirier@linaro.org
Subject: [tip:perf/urgent] perf intel-pt/bts: Calculate cpumode for synthesized samples
Date: Wed, 31 Oct 2018 15:13:21 -0700 [thread overview]
Message-ID: <tip-5d4f0edaa3ac4f1844ed7c64cd2bae6f1912bac5@git.kernel.org> (raw)
In-Reply-To: <20181031091043.23465-6-adrian.hunter@intel.com>
Commit-ID: 5d4f0edaa3ac4f1844ed7c64cd2bae6f1912bac5
Gitweb: https://git.kernel.org/tip/5d4f0edaa3ac4f1844ed7c64cd2bae6f1912bac5
Author: Adrian Hunter <adrian.hunter@intel.com>
AuthorDate: Wed, 31 Oct 2018 11:10:43 +0200
Committer: Arnaldo Carvalho de Melo <acme@redhat.com>
CommitDate: Wed, 31 Oct 2018 12:56:26 -0300
perf intel-pt/bts: Calculate cpumode for synthesized samples
In the absence of a fallback, samples must provide a correct cpumode for
the 'ip'. Do that now there is no fallback.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Reviewed-by: Jiri Olsa <jolsa@kernel.org>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: David S. Miller <davem@davemloft.net>
Cc: Leo Yan <leo.yan@linaro.org>
Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
Cc: stable@vger.kernel.org # 4.19
Link: http://lkml.kernel.org/r/20181031091043.23465-6-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
tools/perf/util/intel-bts.c | 17 ++++++++++++-----
tools/perf/util/intel-pt.c | 22 +++++++++++++---------
2 files changed, 25 insertions(+), 14 deletions(-)
diff --git a/tools/perf/util/intel-bts.c b/tools/perf/util/intel-bts.c
index 3b3a3d55dca1..7b27d77306c2 100644
--- a/tools/perf/util/intel-bts.c
+++ b/tools/perf/util/intel-bts.c
@@ -269,6 +269,13 @@ static int intel_bts_do_fix_overlap(struct auxtrace_queue *queue,
return 0;
}
+static inline u8 intel_bts_cpumode(struct intel_bts *bts, uint64_t ip)
+{
+ return machine__kernel_ip(bts->machine, ip) ?
+ PERF_RECORD_MISC_KERNEL :
+ PERF_RECORD_MISC_USER;
+}
+
static int intel_bts_synth_branch_sample(struct intel_bts_queue *btsq,
struct branch *branch)
{
@@ -281,12 +288,8 @@ static int intel_bts_synth_branch_sample(struct intel_bts_queue *btsq,
bts->num_events++ <= bts->synth_opts.initial_skip)
return 0;
- event.sample.header.type = PERF_RECORD_SAMPLE;
- event.sample.header.misc = PERF_RECORD_MISC_USER;
- event.sample.header.size = sizeof(struct perf_event_header);
-
- sample.cpumode = PERF_RECORD_MISC_USER;
sample.ip = le64_to_cpu(branch->from);
+ sample.cpumode = intel_bts_cpumode(bts, sample.ip);
sample.pid = btsq->pid;
sample.tid = btsq->tid;
sample.addr = le64_to_cpu(branch->to);
@@ -298,6 +301,10 @@ static int intel_bts_synth_branch_sample(struct intel_bts_queue *btsq,
sample.insn_len = btsq->intel_pt_insn.length;
memcpy(sample.insn, btsq->intel_pt_insn.buf, INTEL_PT_INSN_BUF_SZ);
+ event.sample.header.type = PERF_RECORD_SAMPLE;
+ event.sample.header.misc = sample.cpumode;
+ event.sample.header.size = sizeof(struct perf_event_header);
+
if (bts->synth_opts.inject) {
event.sample.header.size = bts->branches_event_size;
ret = perf_event__synthesize_sample(&event,
diff --git a/tools/perf/util/intel-pt.c b/tools/perf/util/intel-pt.c
index 60732213d16a..86cc9a64e982 100644
--- a/tools/perf/util/intel-pt.c
+++ b/tools/perf/util/intel-pt.c
@@ -407,6 +407,13 @@ intel_pt_cache_lookup(struct dso *dso, struct machine *machine, u64 offset)
return auxtrace_cache__lookup(dso->auxtrace_cache, offset);
}
+static inline u8 intel_pt_cpumode(struct intel_pt *pt, uint64_t ip)
+{
+ return ip >= pt->kernel_start ?
+ PERF_RECORD_MISC_KERNEL :
+ PERF_RECORD_MISC_USER;
+}
+
static int intel_pt_walk_next_insn(struct intel_pt_insn *intel_pt_insn,
uint64_t *insn_cnt_ptr, uint64_t *ip,
uint64_t to_ip, uint64_t max_insn_cnt,
@@ -429,10 +436,7 @@ static int intel_pt_walk_next_insn(struct intel_pt_insn *intel_pt_insn,
if (to_ip && *ip == to_ip)
goto out_no_cache;
- if (*ip >= ptq->pt->kernel_start)
- cpumode = PERF_RECORD_MISC_KERNEL;
- else
- cpumode = PERF_RECORD_MISC_USER;
+ cpumode = intel_pt_cpumode(ptq->pt, *ip);
thread = ptq->thread;
if (!thread) {
@@ -1059,15 +1063,11 @@ static void intel_pt_prep_b_sample(struct intel_pt *pt,
union perf_event *event,
struct perf_sample *sample)
{
- event->sample.header.type = PERF_RECORD_SAMPLE;
- event->sample.header.misc = PERF_RECORD_MISC_USER;
- event->sample.header.size = sizeof(struct perf_event_header);
-
if (!pt->timeless_decoding)
sample->time = tsc_to_perf_time(ptq->timestamp, &pt->tc);
- sample->cpumode = PERF_RECORD_MISC_USER;
sample->ip = ptq->state->from_ip;
+ sample->cpumode = intel_pt_cpumode(pt, sample->ip);
sample->pid = ptq->pid;
sample->tid = ptq->tid;
sample->addr = ptq->state->to_ip;
@@ -1076,6 +1076,10 @@ static void intel_pt_prep_b_sample(struct intel_pt *pt,
sample->flags = ptq->flags;
sample->insn_len = ptq->insn_len;
memcpy(sample->insn, ptq->insn, INTEL_PT_INSN_BUF_SZ);
+
+ event->sample.header.type = PERF_RECORD_SAMPLE;
+ event->sample.header.misc = sample->cpumode;
+ event->sample.header.size = sizeof(struct perf_event_header);
}
static int intel_pt_inject_event(union perf_event *event,
next prev parent reply other threads:[~2018-10-31 22:13 UTC|newest]
Thread overview: 25+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-10-31 9:10 [PATCH 0/5] perf tools: Fix for cases where cpumode is incorrect or insufficient Adrian Hunter
2018-10-31 9:10 ` [PATCH 1/5] perf tools: Add fallback functions for cases where cpumode is insufficient Adrian Hunter
2018-11-05 17:29 ` Arnaldo Carvalho de Melo
2018-11-05 18:19 ` David Miller
2018-11-05 19:21 ` Hunter, Adrian
2018-11-05 19:36 ` Arnaldo Carvalho de Melo
2018-11-05 19:53 ` Hunter, Adrian
2018-11-05 20:31 ` Arnaldo Carvalho de Melo
2018-11-26 19:02 ` Arnaldo Carvalho de Melo
2018-11-26 19:41 ` Arnaldo Carvalho de Melo
2018-11-26 19:44 ` Arnaldo Carvalho de Melo
2018-10-31 9:10 ` [PATCH 2/5] perf tools: Use fallback for sample_addr_correlates_sym() cases Adrian Hunter
2018-11-03 17:46 ` Hunter, Adrian
2018-10-31 9:10 ` [PATCH 3/5] perf tools: Use fallbacks for branch stacks Adrian Hunter
2018-11-05 17:56 ` Arnaldo Carvalho de Melo
2018-10-31 9:10 ` [PATCH 4/5] perf intel-pt: Insert callchain context into synthesized callchains Adrian Hunter
2018-10-31 13:28 ` Arnaldo Carvalho de Melo
2018-10-31 14:15 ` Adrian Hunter
2018-10-31 14:20 ` Adrian Hunter
2018-10-31 14:27 ` Arnaldo Carvalho de Melo
2018-10-31 22:12 ` [tip:perf/urgent] " tip-bot for Adrian Hunter
2018-10-31 14:26 ` [PATCH 4/5] " Arnaldo Carvalho de Melo
2018-10-31 9:10 ` [PATCH 5/5] perf intel-pt/bts: Calculate cpumode for synthesized samples Adrian Hunter
2018-10-31 22:13 ` tip-bot for Adrian Hunter [this message]
2018-10-31 13:35 ` [PATCH 0/5] perf tools: Fix for cases where cpumode is incorrect or insufficient Jiri Olsa
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=tip-5d4f0edaa3ac4f1844ed7c64cd2bae6f1912bac5@git.kernel.org \
--to=tipbot@zytor.com \
--cc=acme@redhat.com \
--cc=adrian.hunter@intel.com \
--cc=ak@linux.intel.com \
--cc=davem@davemloft.net \
--cc=hpa@zytor.com \
--cc=jolsa@kernel.org \
--cc=leo.yan@linaro.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-tip-commits@vger.kernel.org \
--cc=mathieu.poirier@linaro.org \
--cc=mingo@kernel.org \
--cc=tglx@linutronix.de \
/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).