[v6] Combine perf and bpf for fast eval of hw breakpoint conditions

[PATCH v6 6/7] perf/bpf: Allow a bpf program to suppress all sample side effects

Posted by Kyle Huey 1 year, 6 months ago

Returning zero from a bpf program attached to a perf event already
suppresses any data output. Return early from __perf_event_overflow() in
this case so it will also suppress event_limit accounting, SIGTRAP
generation, and F_ASYNC signalling.

Signed-off-by: Kyle Huey <khuey@kylehuey.com>
Acked-by: Song Liu <song@kernel.org>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Acked-by: Andrii Nakryiko <andrii@kernel.org>
---
 kernel/events/core.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/kernel/events/core.c b/kernel/events/core.c
index fd601d509cea..cd88d1e89eb8 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -9658,6 +9658,9 @@ static int __perf_event_overflow(struct perf_event *event,
 
 	ret = __perf_event_account_interrupt(event, throttle);
 
+	if (event->prog && !bpf_overflow_handler(event, data, regs))
+		return ret;
+
 	/*
 	 * XXX event_limit might not quite work as expected on inherited
 	 * events
@@ -9707,8 +9710,7 @@ static int __perf_event_overflow(struct perf_event *event,
 		irq_work_queue(&event->pending_irq);
 	}
 
-	if (!(event->prog && !bpf_overflow_handler(event, data, regs)))
-		READ_ONCE(event->overflow_handler)(event, data, regs);
+	READ_ONCE(event->overflow_handler)(event, data, regs);
 
 	if (*perf_event_fasync(event) && event->pending_kill) {
 		event->pending_wakeup = 1;
-- 
2.34.1

[tip: perf/core] perf/bpf: Allow a BPF program to suppress all sample side effects

Posted by tip-bot2 for Kyle Huey 1 year, 6 months ago

The following commit has been merged into the perf/core branch of tip:

Commit-ID:     c4fcc7d1f41532e878087c7c43f4cf247604d68b
Gitweb:        https://git.kernel.org/tip/c4fcc7d1f41532e878087c7c43f4cf247604d68b
Author:        Kyle Huey <me@kylehuey.com>
AuthorDate:    Thu, 11 Apr 2024 18:50:18 -07:00
Committer:     Ingo Molnar <mingo@kernel.org>
CommitterDate: Fri, 12 Apr 2024 11:49:50 +02:00

perf/bpf: Allow a BPF program to suppress all sample side effects

Returning zero from a BPF program attached to a perf event already
suppresses any data output. Return early from __perf_event_overflow() in
this case so it will also suppress event_limit accounting, SIGTRAP
generation, and F_ASYNC signalling.

Signed-off-by: Kyle Huey <khuey@kylehuey.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Song Liu <song@kernel.org>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Acked-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20240412015019.7060-7-khuey@kylehuey.com
---
 kernel/events/core.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/kernel/events/core.c b/kernel/events/core.c
index c6a6936..2212670 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -9677,6 +9677,9 @@ static int __perf_event_overflow(struct perf_event *event,
 
 	ret = __perf_event_account_interrupt(event, throttle);
 
+	if (event->prog && !bpf_overflow_handler(event, data, regs))
+		return ret;
+
 	/*
 	 * XXX event_limit might not quite work as expected on inherited
 	 * events
@@ -9726,8 +9729,7 @@ static int __perf_event_overflow(struct perf_event *event,
 		irq_work_queue(&event->pending_irq);
 	}
 
-	if (!(event->prog && !bpf_overflow_handler(event, data, regs)))
-		READ_ONCE(event->overflow_handler)(event, data, regs);
+	READ_ONCE(event->overflow_handler)(event, data, regs);
 
 	if (*perf_event_fasync(event) && event->pending_kill) {
 		event->pending_wakeup = 1;