From: Arnaldo Carvalho de Melo <acme@kernel.org>
To: Ingo Molnar <mingo@kernel.org>
Cc: Clark Williams <williams@redhat.com>,
linux-kernel@vger.kernel.org, linux-perf-users@vger.kernel.org,
Adrian Hunter <adrian.hunter@intel.com>,
stable@vger.kernel.org,
Arnaldo Carvalho de Melo <acme@redhat.com>
Subject: [PATCH 46/46] perf intel-pt: Fix "Unexpected indirect branch" error
Date: Tue, 5 Jun 2018 14:50:30 -0300 [thread overview]
Message-ID: <20180605175030.32549-47-acme@kernel.org> (raw)
In-Reply-To: <20180605175030.32549-1-acme@kernel.org>
From: Adrian Hunter <adrian.hunter@intel.com>
Some Atom CPUs can produce FUP packets that contain NLIP (next linear
instruction pointer) instead of CLIP (current linear instruction
pointer). That will result in "Unexpected indirect branch" errors. Fix
by comparing IP to NLIP in that case.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: stable@vger.kernel.org
Link: http://lkml.kernel.org/r/1527762225-26024-5-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
tools/perf/util/intel-pt-decoder/intel-pt-decoder.c | 17 +++++++++++++++--
tools/perf/util/intel-pt-decoder/intel-pt-decoder.h | 9 +++++++++
tools/perf/util/intel-pt.c | 4 ++++
3 files changed, 28 insertions(+), 2 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 881d7c5e5e2a..d404bed7003a 100644
--- a/tools/perf/util/intel-pt-decoder/intel-pt-decoder.c
+++ b/tools/perf/util/intel-pt-decoder/intel-pt-decoder.c
@@ -113,6 +113,7 @@ struct intel_pt_decoder {
bool have_cyc;
bool fixup_last_mtc;
bool have_last_ip;
+ enum intel_pt_param_flags flags;
uint64_t pos;
uint64_t last_ip;
uint64_t ip;
@@ -226,6 +227,8 @@ struct intel_pt_decoder *intel_pt_decoder_new(struct intel_pt_params *params)
decoder->return_compression = params->return_compression;
decoder->branch_enable = params->branch_enable;
+ decoder->flags = params->flags;
+
decoder->period = params->period;
decoder->period_type = params->period_type;
@@ -1097,6 +1100,15 @@ static bool intel_pt_fup_event(struct intel_pt_decoder *decoder)
return ret;
}
+static inline bool intel_pt_fup_with_nlip(struct intel_pt_decoder *decoder,
+ struct intel_pt_insn *intel_pt_insn,
+ uint64_t ip, int err)
+{
+ return decoder->flags & INTEL_PT_FUP_WITH_NLIP && !err &&
+ intel_pt_insn->branch == INTEL_PT_BR_INDIRECT &&
+ ip == decoder->ip + intel_pt_insn->length;
+}
+
static int intel_pt_walk_fup(struct intel_pt_decoder *decoder)
{
struct intel_pt_insn intel_pt_insn;
@@ -1109,10 +1121,11 @@ static int intel_pt_walk_fup(struct intel_pt_decoder *decoder)
err = intel_pt_walk_insn(decoder, &intel_pt_insn, ip);
if (err == INTEL_PT_RETURN)
return 0;
- if (err == -EAGAIN) {
+ if (err == -EAGAIN ||
+ intel_pt_fup_with_nlip(decoder, &intel_pt_insn, ip, err)) {
if (intel_pt_fup_event(decoder))
return 0;
- return err;
+ return -EAGAIN;
}
decoder->set_fup_tx_flags = false;
if (err)
diff --git a/tools/perf/util/intel-pt-decoder/intel-pt-decoder.h b/tools/perf/util/intel-pt-decoder/intel-pt-decoder.h
index fc1752d50019..51c18d67f4ca 100644
--- a/tools/perf/util/intel-pt-decoder/intel-pt-decoder.h
+++ b/tools/perf/util/intel-pt-decoder/intel-pt-decoder.h
@@ -60,6 +60,14 @@ enum {
INTEL_PT_ERR_MAX,
};
+enum intel_pt_param_flags {
+ /*
+ * FUP packet can contain next linear instruction pointer instead of
+ * current linear instruction pointer.
+ */
+ INTEL_PT_FUP_WITH_NLIP = 1 << 0,
+};
+
struct intel_pt_state {
enum intel_pt_sample_type type;
int err;
@@ -106,6 +114,7 @@ struct intel_pt_params {
unsigned int mtc_period;
uint32_t tsc_ctc_ratio_n;
uint32_t tsc_ctc_ratio_d;
+ enum intel_pt_param_flags flags;
};
struct intel_pt_decoder;
diff --git a/tools/perf/util/intel-pt.c b/tools/perf/util/intel-pt.c
index 3db7f0ee52a8..aec68908d604 100644
--- a/tools/perf/util/intel-pt.c
+++ b/tools/perf/util/intel-pt.c
@@ -749,6 +749,7 @@ static struct intel_pt_queue *intel_pt_alloc_queue(struct intel_pt *pt,
unsigned int queue_nr)
{
struct intel_pt_params params = { .get_trace = 0, };
+ struct perf_env *env = pt->machine->env;
struct intel_pt_queue *ptq;
ptq = zalloc(sizeof(struct intel_pt_queue));
@@ -830,6 +831,9 @@ static struct intel_pt_queue *intel_pt_alloc_queue(struct intel_pt *pt,
}
}
+ if (env->cpuid && !strncmp(env->cpuid, "GenuineIntel,6,92,", 18))
+ params.flags |= INTEL_PT_FUP_WITH_NLIP;
+
ptq->decoder = intel_pt_decoder_new(¶ms);
if (!ptq->decoder)
goto out_free;
--
2.14.3
next prev parent reply other threads:[~2018-06-05 17:50 UTC|newest]
Thread overview: 52+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-06-05 17:49 [GIT PULL 00/46] perf/core fixes and improvements Arnaldo Carvalho de Melo
2018-06-05 17:49 ` [PATCH 01/46] perf tools: Remove dead quote.[ch] code Arnaldo Carvalho de Melo
2018-06-05 17:49 ` [PATCH 02/46] perf probe: Use return of map__get() to make code more compact Arnaldo Carvalho de Melo
2018-06-05 17:49 ` [PATCH 03/46] perf cgroup: Make evlist__find_cgroup() " Arnaldo Carvalho de Melo
2018-06-05 17:49 ` [PATCH 04/46] perf tools: No need to check if the argument to __get() function is NULL Arnaldo Carvalho de Melo
2018-06-05 17:49 ` [PATCH 05/46] perf annotate: Pass perf_evsel instead of just evsel->idx Arnaldo Carvalho de Melo
2018-06-05 17:49 ` [PATCH 06/46] perf annotate: __symbol__acount_cycles doesn't need notes Arnaldo Carvalho de Melo
2018-06-05 17:49 ` [PATCH 07/46] perf annotate: Split allocation of annotated_source struct Arnaldo Carvalho de Melo
2018-06-05 17:49 ` [PATCH 08/46] perf annotate: Introduce constructor/destructor for annotated_source Arnaldo Carvalho de Melo
2018-06-05 17:49 ` [PATCH 09/46] perf annotate: Introduce annotated_source__alloc_histograms Arnaldo Carvalho de Melo
2018-06-05 17:49 ` [PATCH 10/46] perf annotate: __symbol__inc_addr_samples() needs just annotated_source Arnaldo Carvalho de Melo
2018-06-05 17:49 ` [PATCH 11/46] perf annotate: Introduce symbol__hists() Arnaldo Carvalho de Melo
2018-06-05 17:49 ` [PATCH 12/46] perf annotate: Introduce symbol__cycle_hists() Arnaldo Carvalho de Melo
2018-06-05 17:49 ` [PATCH 13/46] perf annotate: Stop using symbol_conf.nr_events global in symbol__hists() Arnaldo Carvalho de Melo
2018-06-05 17:49 ` [PATCH 14/46] perf annotate: Replace symbol__alloc_hists() with symbol__hists() Arnaldo Carvalho de Melo
2018-06-05 17:49 ` [PATCH 15/46] perf tools: Ditch the symbol_conf.nr_events global Arnaldo Carvalho de Melo
2018-06-05 17:50 ` [PATCH 16/46] perf annotate: Add comment about annotated_src->nr_histograms Arnaldo Carvalho de Melo
2018-06-05 17:50 ` [PATCH 17/46] perf annotate stdio: Use annotation_options consistently Arnaldo Carvalho de Melo
2018-06-05 17:50 ` [PATCH 18/46] perf srcline: Introduce map__srcline() to make code more compact Arnaldo Carvalho de Melo
2018-06-05 17:50 ` [PATCH 19/46] perf sort: Introduce addr_map_symbol__srcline() " Arnaldo Carvalho de Melo
2018-06-05 17:50 ` [PATCH 20/46] perf srcline: Make hist_entry srcline helper consistent with map's Arnaldo Carvalho de Melo
2018-06-05 17:50 ` [PATCH 21/46] perf annotate: Pass annotation_options to symbol__annotate() Arnaldo Carvalho de Melo
2018-06-05 17:50 ` [PATCH 22/46] perf annotate: Adopt anotation options from symbol_conf Arnaldo Carvalho de Melo
2018-06-05 17:50 ` [PATCH 23/46] perf annotate: Move disassembler_style global to annotation_options Arnaldo Carvalho de Melo
2018-06-05 17:50 ` [PATCH 24/46] perf hists browser: Pass annotation_options from tool to browser Arnaldo Carvalho de Melo
2018-06-05 17:50 ` [PATCH 25/46] perf annotate: Move objdump_path to struct annotation_options Arnaldo Carvalho de Melo
2018-06-05 17:50 ` [PATCH 26/46] perf test: Use header file util/debug.h Arnaldo Carvalho de Melo
2018-06-05 17:50 ` [PATCH 27/46] perf report: No need to have report_callchain_help as a global Arnaldo Carvalho de Melo
2018-06-05 17:50 ` [PATCH 28/46] perf evsel: Add has_callchain() helper to make code more compact/clear Arnaldo Carvalho de Melo
2018-06-05 17:50 ` [PATCH 29/46] perf script: Check if evsel has callchains before trying to use it Arnaldo Carvalho de Melo
2018-06-05 17:50 ` [PATCH 30/46] perf sched: Use sched->show_callchain where appropriate Arnaldo Carvalho de Melo
2018-06-05 17:50 ` [PATCH 31/46] perf hists: Do not allocate space for callchains for evsels without them Arnaldo Carvalho de Melo
2018-06-05 17:50 ` [PATCH 32/46] perf hists: Introduce hist_entry__has_callchain() method Arnaldo Carvalho de Melo
2018-06-05 17:50 ` [PATCH 33/46] perf hists: Check if a hist_entry has callchains before using them Arnaldo Carvalho de Melo
2018-06-05 17:50 ` [PATCH 34/46] perf tests kmod-path: Add tests for vdso32 and vdsox32 Arnaldo Carvalho de Melo
2018-06-05 17:50 ` [PATCH 35/46] perf tools: Fix symbol and object code resolution " Arnaldo Carvalho de Melo
2018-06-05 17:50 ` [PATCH 36/46] perf record: Enable arbitrary event names thru name= modifier Arnaldo Carvalho de Melo
2018-06-05 17:50 ` [PATCH 37/46] perf stat: Display user and system time Arnaldo Carvalho de Melo
2018-06-05 17:50 ` [PATCH 38/46] perf tools: Fix pmu events parsing rule Arnaldo Carvalho de Melo
2018-06-05 17:50 ` [PATCH 39/46] perf test code-reading: Fix perf_env setup for PTI entry trampolines Arnaldo Carvalho de Melo
2018-06-05 17:50 ` [PATCH 40/46] perf map: Consider PTI entry trampolines in rip_2objdump() Arnaldo Carvalho de Melo
2018-06-05 17:50 ` [PATCH 41/46] perf test record+probe_libc_inet_pton: Ask 'nm' for dynamic symbols Arnaldo Carvalho de Melo
2018-06-05 17:50 ` [PATCH 42/46] perf script powerpc: Python script for hypervisor call statistics Arnaldo Carvalho de Melo
2018-06-06 14:53 ` Paul Clarke
2018-06-07 5:34 ` Ravi Bangoria
2018-06-07 13:41 ` Paul Clarke
2018-06-07 13:45 ` Arnaldo Carvalho de Melo
2018-06-05 17:50 ` [PATCH 43/46] perf intel-pt: Fix sync_switch INTEL_PT_SS_NOT_TRACING Arnaldo Carvalho de Melo
2018-06-05 17:50 ` [PATCH 44/46] perf intel-pt: Fix decoding to accept CBR between FUP and corresponding TIP Arnaldo Carvalho de Melo
2018-06-05 17:50 ` [PATCH 45/46] perf intel-pt: Fix MTC timing after overflow Arnaldo Carvalho de Melo
2018-06-05 17:50 ` Arnaldo Carvalho de Melo [this message]
2018-06-07 5:22 ` [GIT PULL 00/46] perf/core fixes and improvements 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=20180605175030.32549-47-acme@kernel.org \
--to=acme@kernel.org \
--cc=acme@redhat.com \
--cc=adrian.hunter@intel.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-perf-users@vger.kernel.org \
--cc=mingo@kernel.org \
--cc=stable@vger.kernel.org \
--cc=williams@redhat.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).