提交 66ec1191 编写于 作者: M Mark Rutland 提交者: Arnaldo Carvalho de Melo

perf pmu: Unbreak perf record for arm/arm64 with events with explicit PMU

Currently, perf record is broken on arm/arm64 systems when the PMU is
specified explicitly as part of the event, e.g.

$ ./perf record -e armv8_cortex_a53/cpu_cycles/u true

In such cases, perf record fails to open events unless
perf_event_paranoid is set to -1, even if the PMU in question supports
mode exclusion. Further, even when perf_event_paranoid is toggled, no
samples are recorded.

This is an unintended side effect of commit:

  e3ba76de ("perf tools: Force uncore events to system wide monitoring)

... which assumes that if a PMU has an associated cpu_map, it is an
uncore PMU, and forces events for such PMUs to be system-wide.

This is not true for arm/arm64 systems, which can have heterogeneous
CPUs. To account for this, multiple CPU PMUs are exposed, each with a
"cpus" field under sysfs, which the perf tool parses into a cpu_map. ARM
PMUs do not have a "cpumask" file, and only have a "cpus" file. For the
gory details as to why, see commit:

 7e3fcffe ("perf pmu: Support alternative sysfs cpumask")

Given all of this, we can instead identify uncore PMUs by explicitly
checking for a "cpumask" file, and restore arm/arm64 PMU support back to
a working state. This patch does so, adding a new perf_pmu::is_uncore
field, and splitting the existing cpumask parsing so that it can be
reused.
Signed-off-by: NMark Rutland <mark.rutland@arm.com>
Tested-by Will Deacon <will.deacon@arm.com>
Acked-by: NJiri Olsa <jolsa@kernel.org>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@alien8.de>
Cc: David Ahern <dsahern@gmail.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: 4.12+ <stable@vger.kernel.org>
Fixes: e3ba76de ("perf tools: Force uncore events to system wide monitoring)
Link: http://lkml.kernel.org/r/1507315102-5942-1-git-send-email-mark.rutland@arm.comSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
上级 e9516c08
......@@ -309,10 +309,11 @@ static char *get_config_name(struct list_head *head_terms)
static struct perf_evsel *
__add_event(struct list_head *list, int *idx,
struct perf_event_attr *attr,
char *name, struct cpu_map *cpus,
char *name, struct perf_pmu *pmu,
struct list_head *config_terms, bool auto_merge_stats)
{
struct perf_evsel *evsel;
struct cpu_map *cpus = pmu ? pmu->cpus : NULL;
event_attr_init(attr);
......@@ -323,7 +324,7 @@ __add_event(struct list_head *list, int *idx,
(*idx)++;
evsel->cpus = cpu_map__get(cpus);
evsel->own_cpus = cpu_map__get(cpus);
evsel->system_wide = !!cpus;
evsel->system_wide = pmu ? pmu->is_uncore : false;
evsel->auto_merge_stats = auto_merge_stats;
if (name)
......@@ -1233,7 +1234,7 @@ static int __parse_events_add_pmu(struct parse_events_state *parse_state,
if (!head_config) {
attr.type = pmu->type;
evsel = __add_event(list, &parse_state->idx, &attr, NULL, pmu->cpus, NULL, auto_merge_stats);
evsel = __add_event(list, &parse_state->idx, &attr, NULL, pmu, NULL, auto_merge_stats);
return evsel ? 0 : -ENOMEM;
}
......@@ -1254,7 +1255,7 @@ static int __parse_events_add_pmu(struct parse_events_state *parse_state,
return -EINVAL;
evsel = __add_event(list, &parse_state->idx, &attr,
get_config_name(head_config), pmu->cpus,
get_config_name(head_config), pmu,
&config_terms, auto_merge_stats);
if (evsel) {
evsel->unit = info.unit;
......
......@@ -470,17 +470,36 @@ static void pmu_read_sysfs(void)
closedir(dir);
}
static struct cpu_map *__pmu_cpumask(const char *path)
{
FILE *file;
struct cpu_map *cpus;
file = fopen(path, "r");
if (!file)
return NULL;
cpus = cpu_map__read(file);
fclose(file);
return cpus;
}
/*
* Uncore PMUs have a "cpumask" file under sysfs. CPU PMUs (e.g. on arm/arm64)
* may have a "cpus" file.
*/
#define CPUS_TEMPLATE_UNCORE "%s/bus/event_source/devices/%s/cpumask"
#define CPUS_TEMPLATE_CPU "%s/bus/event_source/devices/%s/cpus"
static struct cpu_map *pmu_cpumask(const char *name)
{
struct stat st;
char path[PATH_MAX];
FILE *file;
struct cpu_map *cpus;
const char *sysfs = sysfs__mountpoint();
const char *templates[] = {
"%s/bus/event_source/devices/%s/cpumask",
"%s/bus/event_source/devices/%s/cpus",
NULL
CPUS_TEMPLATE_UNCORE,
CPUS_TEMPLATE_CPU,
NULL
};
const char **template;
......@@ -489,20 +508,25 @@ static struct cpu_map *pmu_cpumask(const char *name)
for (template = templates; *template; template++) {
snprintf(path, PATH_MAX, *template, sysfs, name);
if (stat(path, &st) == 0)
break;
cpus = __pmu_cpumask(path);
if (cpus)
return cpus;
}
if (!*template)
return NULL;
return NULL;
}
file = fopen(path, "r");
if (!file)
return NULL;
static bool pmu_is_uncore(const char *name)
{
char path[PATH_MAX];
struct cpu_map *cpus;
const char *sysfs = sysfs__mountpoint();
cpus = cpu_map__read(file);
fclose(file);
return cpus;
snprintf(path, PATH_MAX, CPUS_TEMPLATE_UNCORE, sysfs, name);
cpus = __pmu_cpumask(path);
cpu_map__put(cpus);
return !!cpus;
}
/*
......@@ -617,6 +641,8 @@ static struct perf_pmu *pmu_lookup(const char *name)
pmu->cpus = pmu_cpumask(name);
pmu->is_uncore = pmu_is_uncore(name);
INIT_LIST_HEAD(&pmu->format);
INIT_LIST_HEAD(&pmu->aliases);
list_splice(&format, &pmu->format);
......
......@@ -22,6 +22,7 @@ struct perf_pmu {
char *name;
__u32 type;
bool selectable;
bool is_uncore;
struct perf_event_attr *default_config;
struct cpu_map *cpus;
struct list_head format; /* HEAD struct perf_pmu_format -> list */
......
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册