perf lock: Implement cpu and task filters for BPF

Add -a/--all-cpus and -C/--cpu options for cpu filtering.  Also -p/--pid
and --tid options are added for task filtering.  The short -t option is
taken for --threads already.  Tracking the command line workload is
possible as well.

  $ sudo perf lock contention -a -b sleep 1

Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Cc: Blake Jones <blakejones@google.com>
Cc: Boqun Feng <boqun.feng@gmail.com>
Cc: Davidlohr Bueso <dave@stgolabs.net>
Cc: Ian Rogers <irogers@google.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Waiman Long <longman@redhat.com>
Cc: Will Deacon <will@kernel.org>
Link: https://lore.kernel.org/r/20220729200756.666106-4-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>

3 files changed, 99 insertions, 4 deletions

diff --git a/tools/perf/util/bpf_lock_contention.c b/tools/perf/util/bpf_lock_contention.c
index 8eb33e6f5029..16b7451b4b09 100644
--- a/tools/perf/util/bpf_lock_contention.c
+++ b/tools/perf/util/bpf_lock_contention.c
@@ -1,8 +1,11 @@
 // SPDX-License-Identifier: GPL-2.0
 #include "util/debug.h"
+#include "util/evlist.h"
 #include "util/machine.h"
 #include "util/map.h"
 #include "util/symbol.h"
+#include "util/target.h"
+#include "util/thread_map.h"
 #include "util/lock-contention.h"
 #include <linux/zalloc.h>
 #include <bpf/bpf.h>
@@ -24,19 +27,65 @@ struct lock_contention_data {
 	u32 flags;
 };
 
-int lock_contention_prepare(void)
+int lock_contention_prepare(struct evlist *evlist, struct target *target)
 {
+	int i, fd;
+	int ncpus = 1, ntasks = 1;
+
 	skel = lock_contention_bpf__open();
 	if (!skel) {
 		pr_err("Failed to open lock-contention BPF skeleton\n");
 		return -1;
 	}
 
+	if (target__has_cpu(target))
+		ncpus = perf_cpu_map__nr(evlist->core.user_requested_cpus);
+	if (target__has_task(target))
+		ntasks = perf_thread_map__nr(evlist->core.threads);
+
+	bpf_map__set_max_entries(skel->maps.cpu_filter, ncpus);
+	bpf_map__set_max_entries(skel->maps.task_filter, ntasks);
+
 	if (lock_contention_bpf__load(skel) < 0) {
 		pr_err("Failed to load lock-contention BPF skeleton\n");
 		return -1;
 	}
 
+	if (target__has_cpu(target)) {
+		u32 cpu;
+		u8 val = 1;
+
+		skel->bss->has_cpu = 1;
+		fd = bpf_map__fd(skel->maps.cpu_filter);
+
+		for (i = 0; i < ncpus; i++) {
+			cpu = perf_cpu_map__cpu(evlist->core.user_requested_cpus, i).cpu;
+			bpf_map_update_elem(fd, &cpu, &val, BPF_ANY);
+		}
+	}
+
+	if (target__has_task(target)) {
+		u32 pid;
+		u8 val = 1;
+
+		skel->bss->has_task = 1;
+		fd = bpf_map__fd(skel->maps.task_filter);
+
+		for (i = 0; i < ntasks; i++) {
+			pid = perf_thread_map__pid(evlist->core.threads, i);
+			bpf_map_update_elem(fd, &pid, &val, BPF_ANY);
+		}
+	}
+
+	if (target__none(target) && evlist->workload.pid > 0) {
+		u32 pid = evlist->workload.pid;
+		u8 val = 1;
+
+		skel->bss->has_task = 1;
+		fd = bpf_map__fd(skel->maps.task_filter);
+		bpf_map_update_elem(fd, &pid, &val, BPF_ANY);
+	}
+
 	lock_contention_bpf__attach(skel);
 	return 0;
 }
diff --git a/tools/perf/util/bpf_skel/lock_contention.bpf.c b/tools/perf/util/bpf_skel/lock_contention.bpf.c
index 5d1c7641223f..67d46533e518 100644
--- a/tools/perf/util/bpf_skel/lock_contention.bpf.c
+++ b/tools/perf/util/bpf_skel/lock_contention.bpf.c
@@ -54,8 +54,47 @@ struct {
 	__uint(max_entries, MAX_ENTRIES);
 } lock_stat SEC(".maps");
 
+struct {
+	__uint(type, BPF_MAP_TYPE_HASH);
+	__uint(key_size, sizeof(__u32));
+	__uint(value_size, sizeof(__u8));
+	__uint(max_entries, 1);
+} cpu_filter SEC(".maps");
+
+struct {
+	__uint(type, BPF_MAP_TYPE_HASH);
+	__uint(key_size, sizeof(__u32));
+	__uint(value_size, sizeof(__u8));
+	__uint(max_entries, 1);
+} task_filter SEC(".maps");
+
 /* control flags */
 int enabled;
+int has_cpu;
+int has_task;
+
+static inline int can_record(void)
+{
+	if (has_cpu) {
+		__u32 cpu = bpf_get_smp_processor_id();
+		__u8 *ok;
+
+		ok = bpf_map_lookup_elem(&cpu_filter, &cpu);
+		if (!ok)
+			return 0;
+	}
+
+	if (has_task) {
+		__u8 *ok;
+		__u32 pid = bpf_get_current_pid_tgid();
+
+		ok = bpf_map_lookup_elem(&task_filter, &pid);
+		if (!ok)
+			return 0;
+	}
+
+	return 1;
+}
 
 SEC("tp_btf/contention_begin")
 int contention_begin(u64 *ctx)
@@ -63,7 +102,7 @@ int contention_begin(u64 *ctx)
 	struct task_struct *curr;
 	struct tstamp_data *pelem;
 
-	if (!enabled)
+	if (!enabled || !can_record())
 		return 0;
 
 	curr = bpf_get_current_task_btf();
diff --git a/tools/perf/util/lock-contention.h b/tools/perf/util/lock-contention.h
index c92db4a47d8d..092c84441f9f 100644
--- a/tools/perf/util/lock-contention.h
+++ b/tools/perf/util/lock-contention.h
@@ -103,11 +103,13 @@ struct thread_stat {
 #define LCB_F_PERCPU	(1U << 4)
 #define LCB_F_MUTEX	(1U << 5)
 
+struct evlist;
 struct machine;
+struct target;
 
 #ifdef HAVE_BPF_SKEL
 
-int lock_contention_prepare(void);
+int lock_contention_prepare(struct evlist *evlist, struct target *target);
 int lock_contention_start(void);
 int lock_contention_stop(void);
 int lock_contention_read(struct machine *machine, struct hlist_head *head);
@@ -115,7 +117,12 @@ int lock_contention_finish(void);
 
 #else  /* !HAVE_BPF_SKEL */
 
-static inline int lock_contention_prepare(void) { return 0; }
+static inline int lock_contention_prepare(struct evlist *evlist __maybe_unused,
+					  struct target *target __maybe_unused)
+{
+	return 0;
+}
+
 static inline int lock_contention_start(void) { return 0; }
 static inline int lock_contention_stop(void) { return 0; }
 static inline int lock_contention_finish(void) { return 0; }