aboutsummaryrefslogtreecommitdiffstats
path: root/tools/perf/util/intel-pt.c
diff options
context:
space:
mode:
authorAdrian Hunter <adrian.hunter@intel.com>2021-07-01 20:51:32 +0300
committerArnaldo Carvalho de Melo <acme@redhat.com>2021-07-07 11:40:56 -0300
commitb4b046ff9e3410593db7b35326cf696810af8f4f (patch)
treee70198ffe60371758acf44486501fc2560c1c82e /tools/perf/util/intel-pt.c
parentperf stat: Disable the NMI watchdog message on hybrid (diff)
downloadlinux-dev-b4b046ff9e3410593db7b35326cf696810af8f4f.tar.xz
linux-dev-b4b046ff9e3410593db7b35326cf696810af8f4f.zip
perf intel-pt: Add a config for max loops without consuming a packet
The Intel PT decoder limits the number of unconditional branches (e.g. jmps) decoded without consuming any trace packets. Generally, a loop needs a conditional branch which generates a TNT packet, whereas a "ret" instruction will generate a TIP or TNT packet. So exceeding the limit is assumed to be a never-ending loop, which can happen if there has been a decoding error putting the decoder at the wrong place in the code. Up until now, the limit of 10000 has been enough but some analytic purposes have been reported to exceed that. Increase the limit to 100000, and make it configurable via perf config intel-pt.max-loops. Also amend the "Never-ending loop" message to mention the configuration entry. Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Reviewed-by: Andi Kleen <ak@linux.intel.com> Cc: Jiri Olsa <jolsa@redhat.com> Link: http://lore.kernel.org/lkml/20210701175132.3977-1-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Diffstat (limited to '')
-rw-r--r--tools/perf/util/intel-pt.c5
1 files changed, 5 insertions, 0 deletions
diff --git a/tools/perf/util/intel-pt.c b/tools/perf/util/intel-pt.c
index 154a1077f22e..6f852b305e92 100644
--- a/tools/perf/util/intel-pt.c
+++ b/tools/perf/util/intel-pt.c
@@ -123,6 +123,7 @@ struct intel_pt {
u64 noretcomp_bit;
unsigned max_non_turbo_ratio;
unsigned cbr2khz;
+ int max_loops;
unsigned long num_events;
@@ -1200,6 +1201,7 @@ static struct intel_pt_queue *intel_pt_alloc_queue(struct intel_pt *pt,
params.vm_time_correlation = pt->synth_opts.vm_time_correlation;
params.vm_tm_corr_dry_run = pt->synth_opts.vm_tm_corr_dry_run;
params.first_timestamp = pt->first_timestamp;
+ params.max_loops = pt->max_loops;
if (pt->filts.cnt > 0)
params.pgd_ip = intel_pt_pgd_ip;
@@ -3431,6 +3433,9 @@ static int intel_pt_perf_config(const char *var, const char *value, void *data)
if (!strcmp(var, "intel-pt.mispred-all"))
pt->mispred_all = perf_config_bool(var, value);
+ if (!strcmp(var, "intel-pt.max-loops"))
+ perf_config_int(&pt->max_loops, var, value);
+
return 0;
}