linux-kernel - [tip: perf/core] perf/x86/intel/pt: Add sampling support

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Wed, 13 Nov 2019 10:56:44 -0000
From:   "tip-bot2 for Alexander Shishkin" <tip-bot2@...utronix.de>
To:     linux-tip-commits@...r.kernel.org
Cc:     Alexander Shishkin <alexander.shishkin@...ux.intel.com>,
        "Peter Zijlstra (Intel)" <peterz@...radead.org>,
        Arnaldo Carvalho de Melo <acme@...hat.com>,
        David Ahern <dsahern@...il.com>, Jiri Olsa <jolsa@...hat.com>,
        Linus Torvalds <torvalds@...ux-foundation.org>,
        Mark Rutland <mark.rutland@....com>,
        Namhyung Kim <namhyung@...nel.org>,
        Stephane Eranian <eranian@...gle.com>,
        Thomas Gleixner <tglx@...utronix.de>,
        Vince Weaver <vincent.weaver@...ne.edu>,
        adrian.hunter@...el.com, mathieu.poirier@...aro.org,
        Ingo Molnar <mingo@...nel.org>, Borislav Petkov <bp@...en8.de>,
        linux-kernel@...r.kernel.org
Subject: [tip: perf/core] perf/x86/intel/pt: Add sampling support

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

Commit-ID:     25e8920b301c133aeaa9f57d81295bf4ac78e17b
Gitweb:        https://git.kernel.org/tip/25e8920b301c133aeaa9f57d81295bf4ac78e17b
Author:        Alexander Shishkin <alexander.shishkin@...ux.intel.com>
AuthorDate:    Fri, 25 Oct 2019 17:08:35 +03:00
Committer:     Ingo Molnar <mingo@...nel.org>
CommitterDate: Wed, 13 Nov 2019 11:06:16 +01:00

perf/x86/intel/pt: Add sampling support

Add AUX sampling support to the PT PMU: implement an NMI-safe callback
that takes a snapshot of the buffer without touching the event states.
This is done for PT events that don't use PMIs, that is, snapshot mode
(RO mapping of the AUX area).

Signed-off-by: Alexander Shishkin <alexander.shishkin@...ux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@...radead.org>
Cc: Arnaldo Carvalho de Melo <acme@...hat.com>
Cc: David Ahern <dsahern@...il.com>
Cc: Jiri Olsa <jolsa@...hat.com>
Cc: Linus Torvalds <torvalds@...ux-foundation.org>
Cc: Mark Rutland <mark.rutland@....com>
Cc: Namhyung Kim <namhyung@...nel.org>
Cc: Stephane Eranian <eranian@...gle.com>
Cc: Thomas Gleixner <tglx@...utronix.de>
Cc: Vince Weaver <vincent.weaver@...ne.edu>
Cc: adrian.hunter@...el.com
Cc: mathieu.poirier@...aro.org
Link: https://lkml.kernel.org/r/20191025140835.53665-4-alexander.shishkin@linux.intel.com
Signed-off-by: Ingo Molnar <mingo@...nel.org>
---
 arch/x86/events/intel/pt.c | 54 +++++++++++++++++++++++++++++++++++++-
 1 file changed, 54 insertions(+)

diff --git a/arch/x86/events/intel/pt.c b/arch/x86/events/intel/pt.c
index 170f3b4..2f20d5a 100644
--- a/arch/x86/events/intel/pt.c
+++ b/arch/x86/events/intel/pt.c
@@ -1208,6 +1208,13 @@ pt_buffer_setup_aux(struct perf_event *event, void **pages,
 	if (!nr_pages)
 		return NULL;
 
+	/*
+	 * Only support AUX sampling in snapshot mode, where we don't
+	 * generate NMIs.
+	 */
+	if (event->attr.aux_sample_size && !snapshot)
+		return NULL;
+
 	if (cpu == -1)
 		cpu = raw_smp_processor_id();
 	node = cpu_to_node(cpu);
@@ -1506,6 +1513,52 @@ static void pt_event_stop(struct perf_event *event, int mode)
 	}
 }
 
+static long pt_event_snapshot_aux(struct perf_event *event,
+				  struct perf_output_handle *handle,
+				  unsigned long size)
+{
+	struct pt *pt = this_cpu_ptr(&pt_ctx);
+	struct pt_buffer *buf = perf_get_aux(&pt->handle);
+	unsigned long from = 0, to;
+	long ret;
+
+	if (WARN_ON_ONCE(!buf))
+		return 0;
+
+	/*
+	 * Sampling is only allowed on snapshot events;
+	 * see pt_buffer_setup_aux().
+	 */
+	if (WARN_ON_ONCE(!buf->snapshot))
+		return 0;
+
+	/*
+	 * Here, handle_nmi tells us if the tracing is on
+	 */
+	if (READ_ONCE(pt->handle_nmi))
+		pt_config_stop(event);
+
+	pt_read_offset(buf);
+	pt_update_head(pt);
+
+	to = local_read(&buf->data_size);
+	if (to < size)
+		from = buf->nr_pages << PAGE_SHIFT;
+	from += to - size;
+
+	ret = perf_output_copy_aux(&pt->handle, handle, from, to);
+
+	/*
+	 * If the tracing was on when we turned up, restart it.
+	 * Compiler barrier not needed as we couldn't have been
+	 * preempted by anything that touches pt->handle_nmi.
+	 */
+	if (pt->handle_nmi)
+		pt_config_start(event);
+
+	return ret;
+}
+
 static void pt_event_del(struct perf_event *event, int mode)
 {
 	pt_event_stop(event, PERF_EF_UPDATE);
@@ -1625,6 +1678,7 @@ static __init int pt_init(void)
 	pt_pmu.pmu.del			 = pt_event_del;
 	pt_pmu.pmu.start		 = pt_event_start;
 	pt_pmu.pmu.stop			 = pt_event_stop;
+	pt_pmu.pmu.snapshot_aux		 = pt_event_snapshot_aux;
 	pt_pmu.pmu.read			 = pt_event_read;
 	pt_pmu.pmu.setup_aux		 = pt_buffer_setup_aux;
 	pt_pmu.pmu.free_aux		 = pt_buffer_free_aux;