From: Ian Rogers <irogers@google.com>
To: acme@kernel.org
Cc: adrian.hunter@intel.com, ak@linux.intel.com,
alexander.shishkin@linux.intel.com, andres@anarazel.de,
dapeng1.mi@linux.intel.com, irogers@google.com,
james.clark@linaro.org, jolsa@kernel.org,
linux-kernel@vger.kernel.org, linux-perf-users@vger.kernel.org,
linux@treblig.org, mingo@redhat.com, namhyung@kernel.org,
peterz@infradead.org, thomas.falcon@intel.com,
tmricht@linux.ibm.com, yang.lee@linux.alibaba.com
Subject: [PATCH v8 1/6] Revert "perf tool_pmu: More accurately set the cpus for tool events"
Date: Fri, 6 Feb 2026 14:25:04 -0800 [thread overview]
Message-ID: <20260206222509.982489-2-irogers@google.com> (raw)
In-Reply-To: <20260206222509.982489-1-irogers@google.com>
This reverts commit d8d8a0b3603a9a8fa207cf9e4f292e81dc5d1008.
The setting of a user CPU map can cause an empty intersection when
combined with CPU 0 and the event removed. This later triggers a segv
in the stat-shadow logic. Let's put back a full online CPU map for now
by reverting this patch.
Reported-by: Andres Freund <andres@anarazel.de>
Closes: https://lore.kernel.org/linux-perf-users/cgja46br2smmznxs7kbeabs6zgv3b4olfqgh2fdp5mxk2yom4v@w6jjgov6hdi6/
Signed-off-by: Ian Rogers <irogers@google.com>
---
tools/perf/util/parse-events.c | 9 ++-------
tools/perf/util/tool_pmu.c | 19 -------------------
tools/perf/util/tool_pmu.h | 1 -
3 files changed, 2 insertions(+), 27 deletions(-)
diff --git a/tools/perf/util/parse-events.c b/tools/perf/util/parse-events.c
index d4647ded340f..f631bf7a919f 100644
--- a/tools/perf/util/parse-events.c
+++ b/tools/perf/util/parse-events.c
@@ -30,7 +30,6 @@
#include "util/event.h"
#include "util/bpf-filter.h"
#include "util/stat.h"
-#include "util/tool_pmu.h"
#include "util/util.h"
#include "tracepoint.h"
#include <api/fs/tracing_path.h>
@@ -230,12 +229,8 @@ __add_event(struct list_head *list, int *idx,
if (pmu) {
is_pmu_core = pmu->is_core;
pmu_cpus = perf_cpu_map__get(pmu->cpus);
- if (perf_cpu_map__is_empty(pmu_cpus)) {
- if (perf_pmu__is_tool(pmu))
- pmu_cpus = tool_pmu__cpus(attr);
- else
- pmu_cpus = cpu_map__online();
- }
+ if (perf_cpu_map__is_empty(pmu_cpus))
+ pmu_cpus = cpu_map__online();
} else {
is_pmu_core = (attr->type == PERF_TYPE_HARDWARE ||
attr->type == PERF_TYPE_HW_CACHE);
diff --git a/tools/perf/util/tool_pmu.c b/tools/perf/util/tool_pmu.c
index 37c4eae0bef1..6a9df3dc0e07 100644
--- a/tools/perf/util/tool_pmu.c
+++ b/tools/perf/util/tool_pmu.c
@@ -2,7 +2,6 @@
#include "cgroup.h"
#include "counts.h"
#include "cputopo.h"
-#include "debug.h"
#include "evsel.h"
#include "pmu.h"
#include "print-events.h"
@@ -14,7 +13,6 @@
#include <api/fs/fs.h>
#include <api/io.h>
#include <internal/threadmap.h>
-#include <perf/cpumap.h>
#include <perf/threadmap.h>
#include <fcntl.h>
#include <strings.h>
@@ -111,23 +109,6 @@ const char *evsel__tool_pmu_event_name(const struct evsel *evsel)
return tool_pmu__event_to_str(evsel->core.attr.config);
}
-struct perf_cpu_map *tool_pmu__cpus(struct perf_event_attr *attr)
-{
- static struct perf_cpu_map *cpu0_map;
- enum tool_pmu_event event = (enum tool_pmu_event)attr->config;
-
- if (event <= TOOL_PMU__EVENT_NONE || event >= TOOL_PMU__EVENT_MAX) {
- pr_err("Invalid tool PMU event config %llx\n", attr->config);
- return NULL;
- }
- if (event == TOOL_PMU__EVENT_USER_TIME || event == TOOL_PMU__EVENT_SYSTEM_TIME)
- return cpu_map__online();
-
- if (!cpu0_map)
- cpu0_map = perf_cpu_map__new_int(0);
- return perf_cpu_map__get(cpu0_map);
-}
-
static bool read_until_char(struct io *io, char e)
{
int c;
diff --git a/tools/perf/util/tool_pmu.h b/tools/perf/util/tool_pmu.h
index ea343d1983d3..f1714001bc1d 100644
--- a/tools/perf/util/tool_pmu.h
+++ b/tools/perf/util/tool_pmu.h
@@ -46,7 +46,6 @@ bool tool_pmu__read_event(enum tool_pmu_event ev,
u64 tool_pmu__cpu_slots_per_cycle(void);
bool perf_pmu__is_tool(const struct perf_pmu *pmu);
-struct perf_cpu_map *tool_pmu__cpus(struct perf_event_attr *attr);
bool evsel__is_tool(const struct evsel *evsel);
enum tool_pmu_event evsel__tool_event(const struct evsel *evsel);
--
2.53.0.rc2.204.g2597b5adb4-goog
next prev parent reply other threads:[~2026-02-06 22:25 UTC|newest]
Thread overview: 29+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-02-03 22:51 [PATCH v7 0/6] perf stat affinity changes Ian Rogers
2026-02-03 22:51 ` [PATCH v7 1/6] Revert "perf tool_pmu: More accurately set the cpus for tool events" Ian Rogers
2026-02-03 22:51 ` [PATCH v7 2/6] perf stat-shadow: In prepare_metric fix guard on reading NULL perf_stat_evsel Ian Rogers
2026-02-03 22:51 ` [PATCH v7 3/6] perf evlist: Special map propagation for tool events that read on 1 CPU Ian Rogers
2026-02-03 22:51 ` [PATCH v7 4/6] perf evlist: Missing TPEBS close in evlist__close Ian Rogers
2026-02-03 22:51 ` [PATCH v7 5/6] perf evlist: Reduce affinity use and move into iterator, fix no affinity Ian Rogers
2026-02-03 22:51 ` [PATCH v7 6/6] perf stat: Add no-affinity flag Ian Rogers
2026-02-06 21:35 ` [PATCH v7 0/6] perf stat affinity changes Arnaldo Carvalho de Melo
2026-02-06 22:01 ` Ian Rogers
2026-02-06 22:13 ` Arnaldo Carvalho de Melo
2026-02-06 22:25 ` [PATCH v8 " Ian Rogers
2026-02-06 22:25 ` Ian Rogers [this message]
2026-02-09 23:10 ` [PATCH v8 1/6] Revert "perf tool_pmu: More accurately set the cpus for tool events" Ian Rogers
2026-02-06 22:25 ` [PATCH v8 2/6] perf stat-shadow: In prepare_metric fix guard on reading NULL perf_stat_evsel Ian Rogers
2026-02-06 22:25 ` [PATCH v8 3/6] perf evlist: Special map propagation for tool events that read on 1 CPU Ian Rogers
2026-02-06 22:25 ` [PATCH v8 4/6] perf evlist: Missing TPEBS close in evlist__close Ian Rogers
2026-02-06 22:25 ` [PATCH v8 5/6] perf evlist: Reduce affinity use and move into iterator, fix no affinity Ian Rogers
2026-02-06 22:25 ` [PATCH v8 6/6] perf stat: Add no-affinity flag Ian Rogers
2026-02-07 12:51 ` Arnaldo Carvalho de Melo
2026-02-07 15:54 ` Ian Rogers
2026-02-10 1:18 ` Namhyung Kim
2026-02-10 6:03 ` [PATCH v9 0/6] perf stat affinity changes Ian Rogers
2026-02-10 6:03 ` Ian Rogers
2026-02-10 6:03 ` [PATCH v9 1/6] Revert "perf tool_pmu: More accurately set the cpus for tool events" Ian Rogers
2026-02-10 6:03 ` [PATCH v9 2/6] perf stat-shadow: In prepare_metric fix guard on reading NULL perf_stat_evsel Ian Rogers
2026-02-10 6:03 ` [PATCH v9 3/6] perf evlist: Special map propagation for tool events that read on 1 CPU Ian Rogers
2026-02-10 6:03 ` [PATCH v9 4/6] perf evlist: Missing TPEBS close in evlist__close Ian Rogers
2026-02-10 6:03 ` [PATCH v9 5/6] perf evlist: Reduce affinity use and move into iterator, fix no affinity Ian Rogers
2026-02-10 6:03 ` [PATCH v9 6/6] perf stat: Add no-affinity flag Ian Rogers
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=20260206222509.982489-2-irogers@google.com \
--to=irogers@google.com \
--cc=acme@kernel.org \
--cc=adrian.hunter@intel.com \
--cc=ak@linux.intel.com \
--cc=alexander.shishkin@linux.intel.com \
--cc=andres@anarazel.de \
--cc=dapeng1.mi@linux.intel.com \
--cc=james.clark@linaro.org \
--cc=jolsa@kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-perf-users@vger.kernel.org \
--cc=linux@treblig.org \
--cc=mingo@redhat.com \
--cc=namhyung@kernel.org \
--cc=peterz@infradead.org \
--cc=thomas.falcon@intel.com \
--cc=tmricht@linux.ibm.com \
--cc=yang.lee@linux.alibaba.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.