Merge branch 'bench-fast-in-kernel-triggering-benchmarks'

Andrii Nakryiko says:

====================
bench: fast in-kernel triggering benchmarks

Remove "legacy" triggering benchmarks which rely on syscalls (and thus syscall
overhead is a noticeable part of benchmark, unfortunately). Replace them with
faster versions that rely on triggering BPF programs in-kernel through another
simple "driver" BPF program. See patch #2 with comparison results.

raw_tp/tp/fmodret benchmarks required adding a simple kfunc in kernel to be
able to trigger a simple tracepoint from BPF program (plus it is also allowed
to be replaced by fmod_ret programs). This limits raw_tp/tp/fmodret benchmarks
to new kernels only, but it keeps bench tool itself very portable and most of
other benchmarks will still work on wide variety of kernels without the need
to worry about building and deploying custom kernel module. See patches #5
and #6 for details.

v1->v2:
  - move new TP closer to BPF test run code;
  - rename/move kfunc and register it for fmod_rets (Alexei);
  - limit --trig-batch-iters param to [1, 1000] (Alexei).
====================

Link: https://lore.kernel.org/r/20240326162151.3981687-1-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>

8 files changed, 314 insertions, 228 deletions

diff --git a/include/trace/events/bpf_test_run.h b/include/trace/events/bpf_test_run.h
index 265447e3f71abe..0c924d39b7cbaf 100644
--- a/include/trace/events/bpf_test_run.h
+++ b/include/trace/events/bpf_test_run.h
@@ -7,6 +7,23 @@
 
 #include <linux/tracepoint.h>
 
+TRACE_EVENT(bpf_trigger_tp,
+
+	TP_PROTO(int nonce),
+
+	TP_ARGS(nonce),
+
+	TP_STRUCT__entry(
+		__field(int, nonce)
+	),
+
+	TP_fast_assign(
+		__entry->nonce = nonce;
+	),
+
+	TP_printk("nonce %d", __entry->nonce)
+);
+
 DECLARE_EVENT_CLASS(bpf_test_finish,
 
 	TP_PROTO(int *err),
diff --git a/kernel/bpf/helpers.c b/kernel/bpf/helpers.c
index f860adab3eb960..d9e7aca8ae9e86 100644
--- a/kernel/bpf/helpers.c
+++ b/kernel/bpf/helpers.c
@@ -2625,6 +2625,7 @@ BTF_ID_FLAGS(func, bpf_dynptr_is_null)
 BTF_ID_FLAGS(func, bpf_dynptr_is_rdonly)
 BTF_ID_FLAGS(func, bpf_dynptr_size)
 BTF_ID_FLAGS(func, bpf_dynptr_clone)
+BTF_ID_FLAGS(func, bpf_modify_return_test_tp)
 BTF_KFUNCS_END(common_btf_ids)
 
 static const struct btf_kfunc_id_set common_kfunc_set = {
diff --git a/net/bpf/test_run.c b/net/bpf/test_run.c
index 61efeadaff8db0..f6aad4ed2ab2f6 100644
--- a/net/bpf/test_run.c
+++ b/net/bpf/test_run.c
@@ -575,6 +575,13 @@ __bpf_kfunc int bpf_modify_return_test2(int a, int *b, short c, int d,
 	return a + *b + c + d + (long)e + f + g;
 }
 
+__bpf_kfunc int bpf_modify_return_test_tp(int nonce)
+{
+	trace_bpf_trigger_tp(nonce);
+
+	return nonce;
+}
+
 int noinline bpf_fentry_shadow_test(int a)
 {
 	return a + 1;
@@ -622,6 +629,7 @@ __bpf_kfunc_end_defs();
 BTF_KFUNCS_START(bpf_test_modify_return_ids)
 BTF_ID_FLAGS(func, bpf_modify_return_test)
 BTF_ID_FLAGS(func, bpf_modify_return_test2)
+BTF_ID_FLAGS(func, bpf_modify_return_test_tp)
 BTF_ID_FLAGS(func, bpf_fentry_test1, KF_SLEEPABLE)
 BTF_KFUNCS_END(bpf_test_modify_return_ids)
 
diff --git a/tools/testing/selftests/bpf/bench.c b/tools/testing/selftests/bpf/bench.c
index b2b4c391eb0ac2..82de56c8162e1a 100644
--- a/tools/testing/selftests/bpf/bench.c
+++ b/tools/testing/selftests/bpf/bench.c
@@ -280,6 +280,7 @@ extern struct argp bench_strncmp_argp;
 extern struct argp bench_hashmap_lookup_argp;
 extern struct argp bench_local_storage_create_argp;
 extern struct argp bench_htab_mem_argp;
+extern struct argp bench_trigger_batch_argp;
 
 static const struct argp_child bench_parsers[] = {
 	{ &bench_ringbufs_argp, 0, "Ring buffers benchmark", 0 },
@@ -292,6 +293,7 @@ static const struct argp_child bench_parsers[] = {
 	{ &bench_hashmap_lookup_argp, 0, "Hashmap lookup benchmark", 0 },
 	{ &bench_local_storage_create_argp, 0, "local-storage-create benchmark", 0 },
 	{ &bench_htab_mem_argp, 0, "hash map memory benchmark", 0 },
+	{ &bench_trigger_batch_argp, 0, "BPF triggering benchmark", 0 },
 	{},
 };
 
@@ -491,24 +493,31 @@ extern const struct bench bench_rename_kretprobe;
 extern const struct bench bench_rename_rawtp;
 extern const struct bench bench_rename_fentry;
 extern const struct bench bench_rename_fexit;
-extern const struct bench bench_trig_base;
-extern const struct bench bench_trig_tp;
-extern const struct bench bench_trig_rawtp;
+
+/* pure counting benchmarks to establish theoretical lmits */
+extern const struct bench bench_trig_usermode_count;
+extern const struct bench bench_trig_syscall_count;
+extern const struct bench bench_trig_kernel_count;
+
+/* batched, staying mostly in-kernel benchmarks */
 extern const struct bench bench_trig_kprobe;
 extern const struct bench bench_trig_kretprobe;
 extern const struct bench bench_trig_kprobe_multi;
 extern const struct bench bench_trig_kretprobe_multi;
 extern const struct bench bench_trig_fentry;
 extern const struct bench bench_trig_fexit;
-extern const struct bench bench_trig_fentry_sleep;
 extern const struct bench bench_trig_fmodret;
-extern const struct bench bench_trig_uprobe_base;
+extern const struct bench bench_trig_tp;
+extern const struct bench bench_trig_rawtp;
+
+/* uprobe/uretprobe benchmarks */
 extern const struct bench bench_trig_uprobe_nop;
 extern const struct bench bench_trig_uretprobe_nop;
 extern const struct bench bench_trig_uprobe_push;
 extern const struct bench bench_trig_uretprobe_push;
 extern const struct bench bench_trig_uprobe_ret;
 extern const struct bench bench_trig_uretprobe_ret;
+
 extern const struct bench bench_rb_libbpf;
 extern const struct bench bench_rb_custom;
 extern const struct bench bench_pb_libbpf;
@@ -539,24 +548,28 @@ static const struct bench *benchs[] = {
 	&bench_rename_rawtp,
 	&bench_rename_fentry,
 	&bench_rename_fexit,
-	&bench_trig_base,
-	&bench_trig_tp,
-	&bench_trig_rawtp,
+	/* pure counting benchmarks for establishing theoretical limits */
+	&bench_trig_usermode_count,
+	&bench_trig_kernel_count,
+	&bench_trig_syscall_count,
+	/* batched, staying mostly in-kernel triggers */
 	&bench_trig_kprobe,
 	&bench_trig_kretprobe,
 	&bench_trig_kprobe_multi,
 	&bench_trig_kretprobe_multi,
 	&bench_trig_fentry,
 	&bench_trig_fexit,
-	&bench_trig_fentry_sleep,
 	&bench_trig_fmodret,
-	&bench_trig_uprobe_base,
+	&bench_trig_tp,
+	&bench_trig_rawtp,
+	/* uprobes */
 	&bench_trig_uprobe_nop,
 	&bench_trig_uretprobe_nop,
 	&bench_trig_uprobe_push,
 	&bench_trig_uretprobe_push,
 	&bench_trig_uprobe_ret,
 	&bench_trig_uretprobe_ret,
+	/* ringbuf/perfbuf benchmarks */
 	&bench_rb_libbpf,
 	&bench_rb_custom,
 	&bench_pb_libbpf,
diff --git a/tools/testing/selftests/bpf/benchs/bench_trigger.c b/tools/testing/selftests/bpf/benchs/bench_trigger.c
index d66eddacd64266..4b05539f167dff 100644
--- a/tools/testing/selftests/bpf/benchs/bench_trigger.c
+++ b/tools/testing/selftests/bpf/benchs/bench_trigger.c
@@ -1,11 +1,57 @@
 // SPDX-License-Identifier: GPL-2.0
 /* Copyright (c) 2020 Facebook */
 #define _GNU_SOURCE
+#include <argp.h>
 #include <unistd.h>
+#include <stdint.h>
 #include "bench.h"
 #include "trigger_bench.skel.h"
 #include "trace_helpers.h"
 
+#define MAX_TRIG_BATCH_ITERS 1000
+
+static struct {
+	__u32 batch_iters;
+} args = {
+	.batch_iters = 100,
+};
+
+enum {
+	ARG_TRIG_BATCH_ITERS = 7000,
+};
+
+static const struct argp_option opts[] = {
+	{ "trig-batch-iters", ARG_TRIG_BATCH_ITERS, "BATCH_ITER_CNT", 0,
+		"Number of in-kernel iterations per one driver test run"},
+	{},
+};
+
+static error_t parse_arg(int key, char *arg, struct argp_state *state)
+{
+	long ret;
+
+	switch (key) {
+	case ARG_TRIG_BATCH_ITERS:
+		ret = strtol(arg, NULL, 10);
+		if (ret < 1 || ret > MAX_TRIG_BATCH_ITERS) {
+			fprintf(stderr, "invalid --trig-batch-iters value (should be between %d and %d)\n",
+				1, MAX_TRIG_BATCH_ITERS);
+			argp_usage(state);
+		}
+		args.batch_iters = ret;
+		break;
+	default:
+		return ARGP_ERR_UNKNOWN;
+	}
+
+	return 0;
+}
+
+const struct argp bench_trigger_batch_argp = {
+	.options = opts,
+	.parser = parse_arg,
+};
+
 /* adjust slot shift in inc_hits() if changing */
 #define MAX_BUCKETS 256
 
@@ -14,6 +60,8 @@
 /* BPF triggering benchmarks */
 static struct trigger_ctx {
 	struct trigger_bench *skel;
+	bool usermode_counters;
+	int driver_prog_fd;
 } ctx;
 
 static struct counter base_hits[MAX_BUCKETS];
@@ -51,41 +99,63 @@ static void trigger_validate(void)
 	}
 }
 
-static void *trigger_base_producer(void *input)
+static void *trigger_producer(void *input)
 {
-	while (true) {
-		(void)syscall(__NR_getpgid);
-		inc_counter(base_hits);
+	if (ctx.usermode_counters) {
+		while (true) {
+			(void)syscall(__NR_getpgid);
+			inc_counter(base_hits);
+		}
+	} else {
+		while (true)
+			(void)syscall(__NR_getpgid);
 	}
 	return NULL;
 }
 
-static void trigger_base_measure(struct bench_res *res)
+static void *trigger_producer_batch(void *input)
 {
-	res->hits = sum_and_reset_counters(base_hits);
-}
+	int fd = ctx.driver_prog_fd ?: bpf_program__fd(ctx.skel->progs.trigger_driver);
 
-static void *trigger_producer(void *input)
-{
 	while (true)
-		(void)syscall(__NR_getpgid);
+		bpf_prog_test_run_opts(fd, NULL);
+
 	return NULL;
 }
 
 static void trigger_measure(struct bench_res *res)
 {
-	res->hits = sum_and_reset_counters(ctx.skel->bss->hits);
+	if (ctx.usermode_counters)
+		res->hits = sum_and_reset_counters(base_hits);
+	else
+		res->hits = sum_and_reset_counters(ctx.skel->bss->hits);
 }
 
 static void setup_ctx(void)
 {
 	setup_libbpf();
 
-	ctx.skel = trigger_bench__open_and_load();
+	ctx.skel = trigger_bench__open();
 	if (!ctx.skel) {
 		fprintf(stderr, "failed to open skeleton\n");
 		exit(1);
 	}
+
+	/* default "driver" BPF program */
+	bpf_program__set_autoload(ctx.skel->progs.trigger_driver, true);
+
+	ctx.skel->rodata->batch_iters = args.batch_iters;
+}
+
+static void load_ctx(void)
+{
+	int err;
+
+	err = trigger_bench__load(ctx.skel);
+	if (err) {
+		fprintf(stderr, "failed to open skeleton\n");
+		exit(1);
+	}
 }
 
 static void attach_bpf(struct bpf_program *prog)
@@ -99,64 +169,104 @@ static void attach_bpf(struct bpf_program *prog)
 	}
 }
 
-static void trigger_tp_setup(void)
+static void trigger_syscall_count_setup(void)
 {
-	setup_ctx();
-	attach_bpf(ctx.skel->progs.bench_trigger_tp);
+	ctx.usermode_counters = true;
 }
 
-static void trigger_rawtp_setup(void)
+/* Batched, staying mostly in-kernel triggering setups */
+static void trigger_kernel_count_setup(void)
 {
 	setup_ctx();
-	attach_bpf(ctx.skel->progs.bench_trigger_raw_tp);
+	bpf_program__set_autoload(ctx.skel->progs.trigger_driver, false);
+	bpf_program__set_autoload(ctx.skel->progs.trigger_count, true);
+	load_ctx();
+	/* override driver program */
+	ctx.driver_prog_fd = bpf_program__fd(ctx.skel->progs.trigger_count);
 }
 
 static void trigger_kprobe_setup(void)
 {
 	setup_ctx();
+	bpf_program__set_autoload(ctx.skel->progs.bench_trigger_kprobe, true);
+	load_ctx();
 	attach_bpf(ctx.skel->progs.bench_trigger_kprobe);
 }
 
 static void trigger_kretprobe_setup(void)
 {
 	setup_ctx();
+	bpf_program__set_autoload(ctx.skel->progs.bench_trigger_kretprobe, true);
+	load_ctx();
 	attach_bpf(ctx.skel->progs.bench_trigger_kretprobe);
 }
 
 static void trigger_kprobe_multi_setup(void)
 {
 	setup_ctx();
+	bpf_program__set_autoload(ctx.skel->progs.bench_trigger_kprobe_multi, true);
+	load_ctx();
 	attach_bpf(ctx.skel->progs.bench_trigger_kprobe_multi);
 }
 
 static void trigger_kretprobe_multi_setup(void)
 {
 	setup_ctx();
+	bpf_program__set_autoload(ctx.skel->progs.bench_trigger_kretprobe_multi, true);
+	load_ctx();
 	attach_bpf(ctx.skel->progs.bench_trigger_kretprobe_multi);
 }
 
 static void trigger_fentry_setup(void)
 {
 	setup_ctx();
+	bpf_program__set_autoload(ctx.skel->progs.bench_trigger_fentry, true);
+	load_ctx();
 	attach_bpf(ctx.skel->progs.bench_trigger_fentry);
 }
 
 static void trigger_fexit_setup(void)
 {
 	setup_ctx();
+	bpf_program__set_autoload(ctx.skel->progs.bench_trigger_fexit, true);
+	load_ctx();
 	attach_bpf(ctx.skel->progs.bench_trigger_fexit);
 }
 
-static void trigger_fentry_sleep_setup(void)
+static void trigger_fmodret_setup(void)
 {
 	setup_ctx();
-	attach_bpf(ctx.skel->progs.bench_trigger_fentry_sleep);
+	bpf_program__set_autoload(ctx.skel->progs.trigger_driver, false);
+	bpf_program__set_autoload(ctx.skel->progs.trigger_driver_kfunc, true);
+	bpf_program__set_autoload(ctx.skel->progs.bench_trigger_fmodret, true);
+	load_ctx();
+	/* override driver program */
+	ctx.driver_prog_fd = bpf_program__fd(ctx.skel->progs.trigger_driver_kfunc);
+	attach_bpf(ctx.skel->progs.bench_trigger_fmodret);
 }
 
-static void trigger_fmodret_setup(void)
+static void trigger_tp_setup(void)
 {
 	setup_ctx();
-	attach_bpf(ctx.skel->progs.bench_trigger_fmodret);
+	bpf_program__set_autoload(ctx.skel->progs.trigger_driver, false);
+	bpf_program__set_autoload(ctx.skel->progs.trigger_driver_kfunc, true);
+	bpf_program__set_autoload(ctx.skel->progs.bench_trigger_tp, true);
+	load_ctx();
+	/* override driver program */
+	ctx.driver_prog_fd = bpf_program__fd(ctx.skel->progs.trigger_driver_kfunc);
+	attach_bpf(ctx.skel->progs.bench_trigger_tp);
+}
+
+static void trigger_rawtp_setup(void)
+{
+	setup_ctx();
+	bpf_program__set_autoload(ctx.skel->progs.trigger_driver, false);
+	bpf_program__set_autoload(ctx.skel->progs.trigger_driver_kfunc, true);
+	bpf_program__set_autoload(ctx.skel->progs.bench_trigger_rawtp, true);
+	load_ctx();
+	/* override driver program */
+	ctx.driver_prog_fd = bpf_program__fd(ctx.skel->progs.trigger_driver_kfunc);
+	attach_bpf(ctx.skel->progs.bench_trigger_rawtp);
 }
 
 /* make sure call is not inlined and not avoided by compiler, so __weak and
@@ -192,7 +302,7 @@ __nocf_check __weak void uprobe_target_ret(void)
 	asm volatile ("");
 }
 
-static void *uprobe_base_producer(void *input)
+static void *uprobe_producer_count(void *input)
 {
 	while (true) {
 		uprobe_target_nop();
@@ -226,15 +336,24 @@ static void usetup(bool use_retprobe, void *target_addr)
 {
 	size_t uprobe_offset;
 	struct bpf_link *link;
+	int err;
 
 	setup_libbpf();
 
-	ctx.skel = trigger_bench__open_and_load();
+	ctx.skel = trigger_bench__open();
 	if (!ctx.skel) {
 		fprintf(stderr, "failed to open skeleton\n");
 		exit(1);
 	}
 
+	bpf_program__set_autoload(ctx.skel->progs.bench_trigger_uprobe, true);
+
+	err = trigger_bench__load(ctx.skel);
+	if (err) {
+		fprintf(stderr, "failed to load skeleton\n");
+		exit(1);
+	}
+
 	uprobe_offset = get_uprobe_offset(target_addr);
 	link = bpf_program__attach_uprobe(ctx.skel->progs.bench_trigger_uprobe,
 					  use_retprobe,
@@ -248,204 +367,90 @@ static void usetup(bool use_retprobe, void *target_addr)
 	ctx.skel->links.bench_trigger_uprobe = link;
 }
 
-static void uprobe_setup_nop(void)
+static void usermode_count_setup(void)
+{
+	ctx.usermode_counters = true;
+}
+
+static void uprobe_nop_setup(void)
 {
 	usetup(false, &uprobe_target_nop);
 }
 
-static void uretprobe_setup_nop(void)
+static void uretprobe_nop_setup(void)
 {
 	usetup(true, &uprobe_target_nop);
 }
 
-static void uprobe_setup_push(void)
+static void uprobe_push_setup(void)
 {
 	usetup(false, &uprobe_target_push);
 }
 
-static void uretprobe_setup_push(void)
+static void uretprobe_push_setup(void)
 {
 	usetup(true, &uprobe_target_push);
 }
 
-static void uprobe_setup_ret(void)
+static void uprobe_ret_setup(void)
 {
 	usetup(false, &uprobe_target_ret);
 }
 
-static void uretprobe_setup_ret(void)
+static void uretprobe_ret_setup(void)
 {
 	usetup(true, &uprobe_target_ret);
 }
 
-const struct bench bench_trig_base = {
-	.name = "trig-base",
-	.validate = trigger_validate,
-	.producer_thread = trigger_base_producer,
-	.measure = trigger_base_measure,
-	.report_progress = hits_drops_report_progress,
-	.report_final = hits_drops_report_final,
-};
-
-const struct bench bench_trig_tp = {
-	.name = "trig-tp",
-	.validate = trigger_validate,
-	.setup = trigger_tp_setup,
-	.producer_thread = trigger_producer,
-	.measure = trigger_measure,
-	.report_progress = hits_drops_report_progress,
-	.report_final = hits_drops_report_final,
-};
-
-const struct bench bench_trig_rawtp = {
-	.name = "trig-rawtp",
-	.validate = trigger_validate,
-	.setup = trigger_rawtp_setup,
-	.producer_thread = trigger_producer,
-	.measure = trigger_measure,
-	.report_progress = hits_drops_report_progress,
-	.report_final = hits_drops_report_final,
-};
-
-const struct bench bench_trig_kprobe = {
-	.name = "trig-kprobe",
-	.validate = trigger_validate,
-	.setup = trigger_kprobe_setup,
-	.producer_thread = trigger_producer,
-	.measure = trigger_measure,
-	.report_progress = hits_drops_report_progress,
-	.report_final = hits_drops_report_final,
-};
-
-const struct bench bench_trig_kretprobe = {
-	.name = "trig-kretprobe",
-	.validate = trigger_validate,
-	.setup = trigger_kretprobe_setup,
-	.producer_thread = trigger_producer,
-	.measure = trigger_measure,
-	.report_progress = hits_drops_report_progress,
-	.report_final = hits_drops_report_final,
-};
-
-const struct bench bench_trig_kprobe_multi = {
-	.name = "trig-kprobe-multi",
-	.validate = trigger_validate,
-	.setup = trigger_kprobe_multi_setup,
-	.producer_thread = trigger_producer,
-	.measure = trigger_measure,
-	.report_progress = hits_drops_report_progress,
-	.report_final = hits_drops_report_final,
-};
-
-const struct bench bench_trig_kretprobe_multi = {
-	.name = "trig-kretprobe-multi",
-	.validate = trigger_validate,
-	.setup = trigger_kretprobe_multi_setup,
-	.producer_thread = trigger_producer,
-	.measure = trigger_measure,
-	.report_progress = hits_drops_report_progress,
-	.report_final = hits_drops_report_final,
-};
-
-const struct bench bench_trig_fentry = {
-	.name = "trig-fentry",
-	.validate = trigger_validate,
-	.setup = trigger_fentry_setup,
-	.producer_thread = trigger_producer,
-	.measure = trigger_measure,
-	.report_progress = hits_drops_report_progress,
-	.report_final = hits_drops_report_final,
-};
-
-const struct bench bench_trig_fexit = {
-	.name = "trig-fexit",
-	.validate = trigger_validate,
-	.setup = trigger_fexit_setup,
-	.producer_thread = trigger_producer,
-	.measure = trigger_measure,
-	.report_progress = hits_drops_report_progress,
-	.report_final = hits_drops_report_final,
-};
-
-const struct bench bench_trig_fentry_sleep = {
-	.name = "trig-fentry-sleep",
-	.validate = trigger_validate,
-	.setup = trigger_fentry_sleep_setup,
-	.producer_thread = trigger_producer,
-	.measure = trigger_measure,
-	.report_progress = hits_drops_report_progress,
-	.report_final = hits_drops_report_final,
-};
-
-const struct bench bench_trig_fmodret = {
-	.name = "trig-fmodret",
+const struct bench bench_trig_syscall_count = {
+	.name = "trig-syscall-count",
 	.validate = trigger_validate,
-	.setup = trigger_fmodret_setup,
+	.setup = trigger_syscall_count_setup,
 	.producer_thread = trigger_producer,
 	.measure = trigger_measure,
 	.report_progress = hits_drops_report_progress,
 	.report_final = hits_drops_report_final,
 };
 
-const struct bench bench_trig_uprobe_base = {
-	.name = "trig-uprobe-base",
-	.setup = NULL, /* no uprobe/uretprobe is attached */
-	.producer_thread = uprobe_base_producer,
-	.measure = trigger_base_measure,
-	.report_progress = hits_drops_report_progress,
-	.report_final = hits_drops_report_final,
-};
-
-const struct bench bench_trig_uprobe_nop = {
-	.name = "trig-uprobe-nop",
-	.setup = uprobe_setup_nop,
-	.producer_thread = uprobe_producer_nop,
-	.measure = trigger_measure,
-	.report_progress = hits_drops_report_progress,
-	.report_final = hits_drops_report_final,
-};
-
-const struct bench bench_trig_uretprobe_nop = {
-	.name = "trig-uretprobe-nop",
-	.setup = uretprobe_setup_nop,
-	.producer_thread = uprobe_producer_nop,
-	.measure = trigger_measure,
-	.report_progress = hits_drops_report_progress,
-	.report_final = hits_drops_report_final,
-};
-
-const struct bench bench_trig_uprobe_push = {
-	.name = "trig-uprobe-push",
-	.setup = uprobe_setup_push,
-	.producer_thread = uprobe_producer_push,
-	.measure = trigger_measure,
-	.report_progress = hits_drops_report_progress,
-	.report_final = hits_drops_report_final,
-};
-
-const struct bench bench_trig_uretprobe_push = {
-	.name = "trig-uretprobe-push",
-	.setup = uretprobe_setup_push,
-	.producer_thread = uprobe_producer_push,
-	.measure = trigger_measure,
-	.report_progress = hits_drops_report_progress,
-	.report_final = hits_drops_report_final,
-};
-
-const struct bench bench_trig_uprobe_ret = {
-	.name = "trig-uprobe-ret",
-	.setup = uprobe_setup_ret,
-	.producer_thread = uprobe_producer_ret,
-	.measure = trigger_measure,
-	.report_progress = hits_drops_report_progress,
-	.report_final = hits_drops_report_final,
-};
-
-const struct bench bench_trig_uretprobe_ret = {
-	.name = "trig-uretprobe-ret",
-	.setup = uretprobe_setup_ret,
-	.producer_thread = uprobe_producer_ret,
-	.measure = trigger_measure,
-	.report_progress = hits_drops_report_progress,
-	.report_final = hits_drops_report_final,
-};
+/* batched (staying mostly in kernel) kprobe/fentry benchmarks */
+#define BENCH_TRIG_KERNEL(KIND, NAME)					\
+const struct bench bench_trig_##KIND = {				\
+	.name = "trig-" NAME,						\
+	.setup = trigger_##KIND##_setup,				\
+	.producer_thread = trigger_producer_batch,			\
+	.measure = trigger_measure,					\
+	.report_progress = hits_drops_report_progress,			\
+	.report_final = hits_drops_report_final,			\
+	.argp = &bench_trigger_batch_argp,				\
+}
+
+BENCH_TRIG_KERNEL(kernel_count, "kernel-count");
+BENCH_TRIG_KERNEL(kprobe, "kprobe");
+BENCH_TRIG_KERNEL(kretprobe, "kretprobe");
+BENCH_TRIG_KERNEL(kprobe_multi, "kprobe-multi");
+BENCH_TRIG_KERNEL(kretprobe_multi, "kretprobe-multi");
+BENCH_TRIG_KERNEL(fentry, "fentry");
+BENCH_TRIG_KERNEL(fexit, "fexit");
+BENCH_TRIG_KERNEL(fmodret, "fmodret");
+BENCH_TRIG_KERNEL(tp, "tp");
+BENCH_TRIG_KERNEL(rawtp, "rawtp");
+
+/* uprobe benchmarks */
+#define BENCH_TRIG_USERMODE(KIND, PRODUCER, NAME)			\
+const struct bench bench_trig_##KIND = {				\
+	.name = "trig-" NAME,						\
+	.validate = trigger_validate,					\
+	.setup = KIND##_setup,						\
+	.producer_thread = uprobe_producer_##PRODUCER,			\
+	.measure = trigger_measure,					\
+	.report_progress = hits_drops_report_progress,			\
+	.report_final = hits_drops_report_final,			\
+}
+
+BENCH_TRIG_USERMODE(usermode_count, count, "usermode-count");
+BENCH_TRIG_USERMODE(uprobe_nop, nop, "uprobe-nop");
+BENCH_TRIG_USERMODE(uprobe_push, push, "uprobe-push");
+BENCH_TRIG_USERMODE(uprobe_ret, ret, "uprobe-ret");
+BENCH_TRIG_USERMODE(uretprobe_nop, nop, "uretprobe-nop");
+BENCH_TRIG_USERMODE(uretprobe_push, push, "uretprobe-push");
+BENCH_TRIG_USERMODE(uretprobe_ret, ret, "uretprobe-ret");
diff --git a/tools/testing/selftests/bpf/benchs/run_bench_trigger.sh b/tools/testing/selftests/bpf/benchs/run_bench_trigger.sh
index 78e83f24329463..a690f5a68b6b02 100755
--- a/tools/testing/selftests/bpf/benchs/run_bench_trigger.sh
+++ b/tools/testing/selftests/bpf/benchs/run_bench_trigger.sh
@@ -2,8 +2,22 @@
 
 set -eufo pipefail
 
-for i in base tp rawtp kprobe fentry fmodret
-do
-	summary=$(sudo ./bench -w2 -d5 -a trig-$i | tail -n1 | cut -d'(' -f1 | cut -d' ' -f3-)
-	printf "%-10s: %s\n" $i "$summary"
+def_tests=( \
+	usermode-count kernel-count syscall-count \
+	fentry fexit fmodret \
+	rawtp tp \
+	kprobe kprobe-multi \
+	kretprobe kretprobe-multi \
+)
+
+tests=("$@")
+if [ ${#tests[@]} -eq 0 ]; then
+	tests=("${def_tests[@]}")
+fi
+
+p=${PROD_CNT:-1}
+
+for t in "${tests[@]}"; do
+	summary=$(sudo ./bench -w2 -d5 -a -p$p trig-$t | tail -n1 | cut -d'(' -f1 | cut -d' ' -f3-)
+	printf "%-15s: %s\n" $t "$summary"
 done
diff --git a/tools/testing/selftests/bpf/benchs/run_bench_uprobes.sh b/tools/testing/selftests/bpf/benchs/run_bench_uprobes.sh
index 9bdcc74e03a41f..af169f831f2f25 100755
--- a/tools/testing/selftests/bpf/benchs/run_bench_uprobes.sh
+++ b/tools/testing/selftests/bpf/benchs/run_bench_uprobes.sh
@@ -2,7 +2,7 @@
 
 set -eufo pipefail
 
-for i in base {uprobe,uretprobe}-{nop,push,ret}
+for i in usermode-count syscall-count {uprobe,uretprobe}-{nop,push,ret}
 do
 	summary=$(sudo ./bench -w2 -d5 -a trig-$i | tail -n1 | cut -d'(' -f1 | cut -d' ' -f3-)
 	printf "%-15s: %s\n" $i "$summary"
diff --git a/tools/testing/selftests/bpf/progs/trigger_bench.c b/tools/testing/selftests/bpf/progs/trigger_bench.c
index 42ec202015ed40..2619ed193c65a5 100644
--- a/tools/testing/selftests/bpf/progs/trigger_bench.c
+++ b/tools/testing/selftests/bpf/progs/trigger_bench.c
@@ -1,6 +1,5 @@
 // SPDX-License-Identifier: GPL-2.0
 // Copyright (c) 2020 Facebook
-
 #include <linux/bpf.h>
 #include <asm/unistd.h>
 #include <bpf/bpf_helpers.h>
@@ -26,79 +25,108 @@ static __always_inline void inc_counter(void)
 	__sync_add_and_fetch(&hits[cpu & CPU_MASK].value, 1);
 }
 
-SEC("tp/syscalls/sys_enter_getpgid")
-int bench_trigger_tp(void *ctx)
+SEC("?uprobe")
+int bench_trigger_uprobe(void *ctx)
 {
 	inc_counter();
 	return 0;
 }
 
-SEC("raw_tp/sys_enter")
-int BPF_PROG(bench_trigger_raw_tp, struct pt_regs *regs, long id)
+const volatile int batch_iters = 0;
+
+SEC("?raw_tp")
+int trigger_count(void *ctx)
 {
-	if (id == __NR_getpgid)
+	int i;
+
+	for (i = 0; i < batch_iters; i++)
 		inc_counter();
+
+	return 0;
+}
+
+SEC("?raw_tp")
+int trigger_driver(void *ctx)
+{
+	int i;
+
+	for (i = 0; i < batch_iters; i++)
+		(void)bpf_get_numa_node_id(); /* attach point for benchmarking */
+
 	return 0;
 }
 
-SEC("kprobe/" SYS_PREFIX "sys_getpgid")
+extern int bpf_modify_return_test_tp(int nonce) __ksym __weak;
+
+SEC("?raw_tp")
+int trigger_driver_kfunc(void *ctx)
+{
+	int i;
+
+	for (i = 0; i < batch_iters; i++)
+		(void)bpf_modify_return_test_tp(0); /* attach point for benchmarking */
+
+	return 0;
+}
+
+SEC("?kprobe/bpf_get_numa_node_id")
 int bench_trigger_kprobe(void *ctx)
 {
 	inc_counter();
 	return 0;
 }
 
-SEC("kretprobe/" SYS_PREFIX "sys_getpgid")
+SEC("?kretprobe/bpf_get_numa_node_id")
 int bench_trigger_kretprobe(void *ctx)
 {
 	inc_counter();
 	return 0;
 }
 
-SEC("kprobe.multi/" SYS_PREFIX "sys_getpgid")
+SEC("?kprobe.multi/bpf_get_numa_node_id")
 int bench_trigger_kprobe_multi(void *ctx)
 {
 	inc_counter();
 	return 0;
 }
 
-SEC("kretprobe.multi/" SYS_PREFIX "sys_getpgid")
+SEC("?kretprobe.multi/bpf_get_numa_node_id")
 int bench_trigger_kretprobe_multi(void *ctx)
 {
 	inc_counter();
 	return 0;
 }
 
-SEC("fentry/" SYS_PREFIX "sys_getpgid")
+SEC("?fentry/bpf_get_numa_node_id")
 int bench_trigger_fentry(void *ctx)
 {
 	inc_counter();
 	return 0;
 }
 
-SEC("fexit/" SYS_PREFIX "sys_getpgid")
+SEC("?fexit/bpf_get_numa_node_id")
 int bench_trigger_fexit(void *ctx)
 {
 	inc_counter();
 	return 0;
 }
 
-SEC("fentry.s/" SYS_PREFIX "sys_getpgid")
-int bench_trigger_fentry_sleep(void *ctx)
+SEC("?fmod_ret/bpf_modify_return_test_tp")
+int bench_trigger_fmodret(void *ctx)
 {
 	inc_counter();
-	return 0;
+	return -22;
 }
 
-SEC("fmod_ret/" SYS_PREFIX "sys_getpgid")
-int bench_trigger_fmodret(void *ctx)
+SEC("?tp/bpf_test_run/bpf_trigger_tp")
+int bench_trigger_tp(void *ctx)
 {
 	inc_counter();
-	return -22;
+	return 0;
 }
 
-SEC("uprobe")
-int bench_trigger_uprobe(void *ctx)
+SEC("?raw_tp/bpf_trigger_tp")
+int bench_trigger_rawtp(void *ctx)
 {
 	inc_counter();
 	return 0;