Merge tag 'trace-ringbuffer-v6.10' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace

Pull tracing ring buffer updates from Steven Rostedt:
 "Add ring_buffer memory mappings.

  The tracing ring buffer was created based on being mostly used with
  the splice system call. It is broken up into page ordered sub-buffers
  and the reader swaps a new sub-buffer with an existing sub-buffer
  that's part of the write buffer. It then has total access to the
  swapped out sub-buffer and can do copyless movements of the memory
  into other mediums (file system, network, etc).

  The buffer is great for passing around the ring buffer contents in the
  kernel, but is not so good for when the consumer is the user space
  task itself.

  A new interface is added that allows user space to memory map the ring
  buffer. It will get all the write sub-buffers as well as reader
  sub-buffer (that is not written to). It can send an ioctl to change
  which sub-buffer is the new reader sub-buffer.

  The ring buffer is read only to user space. It only needs to call the
  ioctl when it is finished with a sub-buffer and needs a new sub-buffer
  that the writer will not write over.

  A self test program was also created for testing and can be used as an
  example for the interface to user space. The libtracefs (external to
  the kernel) also has code that interacts with this, although it is
  disabled until the interface is in a official release. It can be
  enabled by compiling the library with a special flag. This was used
  for testing applications that perform better with the buffer being
  mapped.

  Memory mapped buffers have limitations. The main one is that it can
  not be used with the snapshot logic. If the buffer is mapped,
  snapshots will be disabled. If any logic is set to trigger snapshots
  on a buffer, that buffer will not be allowed to be mapped"

* tag 'trace-ringbuffer-v6.10' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace:
  ring-buffer: Add cast to unsigned long addr passed to virt_to_page()
  ring-buffer: Have mmapped ring buffer keep track of missed events
  ring-buffer/selftest: Add ring-buffer mapping test
  Documentation: tracing: Add ring-buffer mapping
  tracing: Allow user-space mapping of the ring-buffer
  ring-buffer: Introducing ring-buffer mapping functions
  ring-buffer: Allocate sub-buffers with __GFP_COMP

4 files changed, 305 insertions, 0 deletions

diff --git a/tools/testing/selftests/ring-buffer/.gitignore b/tools/testing/selftests/ring-buffer/.gitignore
new file mode 100644
index 000000000000..3aed1a2a6c67
--- /dev/null
+++ b/tools/testing/selftests/ring-buffer/.gitignore
@@ -0,0 +1 @@
+map_test
diff --git a/tools/testing/selftests/ring-buffer/Makefile b/tools/testing/selftests/ring-buffer/Makefile
new file mode 100644
index 000000000000..627c5fa6d1ab
--- /dev/null
+++ b/tools/testing/selftests/ring-buffer/Makefile
@@ -0,0 +1,8 @@
+# SPDX-License-Identifier: GPL-2.0
+CFLAGS += -Wl,-no-as-needed -Wall
+CFLAGS += $(KHDR_INCLUDES)
+CFLAGS += -D_GNU_SOURCE
+
+TEST_GEN_PROGS = map_test
+
+include ../lib.mk
diff --git a/tools/testing/selftests/ring-buffer/config b/tools/testing/selftests/ring-buffer/config
new file mode 100644
index 000000000000..d936f8f00e78
--- /dev/null
+++ b/tools/testing/selftests/ring-buffer/config
@@ -0,0 +1,2 @@
+CONFIG_FTRACE=y
+CONFIG_TRACER_SNAPSHOT=y
diff --git a/tools/testing/selftests/ring-buffer/map_test.c b/tools/testing/selftests/ring-buffer/map_test.c
new file mode 100644
index 000000000000..a9006fa7097e
--- /dev/null
+++ b/tools/testing/selftests/ring-buffer/map_test.c
@@ -0,0 +1,294 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Ring-buffer memory mapping tests
+ *
+ * Copyright (c) 2024 Vincent Donnefort <vdonnefort@google.com>
+ */
+#include <fcntl.h>
+#include <sched.h>
+#include <stdbool.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <unistd.h>
+
+#include <linux/trace_mmap.h>
+
+#include <sys/mman.h>
+#include <sys/ioctl.h>
+
+#include "../user_events/user_events_selftests.h" /* share tracefs setup */
+#include "../kselftest_harness.h"
+
+#define TRACEFS_ROOT "/sys/kernel/tracing"
+
+static int __tracefs_write(const char *path, const char *value)
+{
+	int fd, ret;
+
+	fd = open(path, O_WRONLY | O_TRUNC);
+	if (fd < 0)
+		return fd;
+
+	ret = write(fd, value, strlen(value));
+
+	close(fd);
+
+	return ret == -1 ? -errno : 0;
+}
+
+static int __tracefs_write_int(const char *path, int value)
+{
+	char *str;
+	int ret;
+
+	if (asprintf(&str, "%d", value) < 0)
+		return -1;
+
+	ret = __tracefs_write(path, str);
+
+	free(str);
+
+	return ret;
+}
+
+#define tracefs_write_int(path, value) \
+	ASSERT_EQ(__tracefs_write_int((path), (value)), 0)
+
+#define tracefs_write(path, value) \
+	ASSERT_EQ(__tracefs_write((path), (value)), 0)
+
+static int tracefs_reset(void)
+{
+	if (__tracefs_write_int(TRACEFS_ROOT"/tracing_on", 0))
+		return -1;
+	if (__tracefs_write(TRACEFS_ROOT"/trace", ""))
+		return -1;
+	if (__tracefs_write(TRACEFS_ROOT"/set_event", ""))
+		return -1;
+	if (__tracefs_write(TRACEFS_ROOT"/current_tracer", "nop"))
+		return -1;
+
+	return 0;
+}
+
+struct tracefs_cpu_map_desc {
+	struct trace_buffer_meta	*meta;
+	int				cpu_fd;
+};
+
+int tracefs_cpu_map(struct tracefs_cpu_map_desc *desc, int cpu)
+{
+	int page_size = getpagesize();
+	char *cpu_path;
+	void *map;
+
+	if (asprintf(&cpu_path,
+		     TRACEFS_ROOT"/per_cpu/cpu%d/trace_pipe_raw",
+		     cpu) < 0)
+		return -ENOMEM;
+
+	desc->cpu_fd = open(cpu_path, O_RDONLY | O_NONBLOCK);
+	free(cpu_path);
+	if (desc->cpu_fd < 0)
+		return -ENODEV;
+
+	map = mmap(NULL, page_size, PROT_READ, MAP_SHARED, desc->cpu_fd, 0);
+	if (map == MAP_FAILED)
+		return -errno;
+
+	desc->meta = (struct trace_buffer_meta *)map;
+
+	return 0;
+}
+
+void tracefs_cpu_unmap(struct tracefs_cpu_map_desc *desc)
+{
+	munmap(desc->meta, desc->meta->meta_page_size);
+	close(desc->cpu_fd);
+}
+
+FIXTURE(map) {
+	struct tracefs_cpu_map_desc	map_desc;
+	bool				umount;
+};
+
+FIXTURE_VARIANT(map) {
+	int	subbuf_size;
+};
+
+FIXTURE_VARIANT_ADD(map, subbuf_size_4k) {
+	.subbuf_size = 4,
+};
+
+FIXTURE_VARIANT_ADD(map, subbuf_size_8k) {
+	.subbuf_size = 8,
+};
+
+FIXTURE_SETUP(map)
+{
+	int cpu = sched_getcpu();
+	cpu_set_t cpu_mask;
+	bool fail, umount;
+	char *message;
+
+	if (getuid() != 0)
+		SKIP(return, "Skipping: %s", "Please run the test as root");
+
+	if (!tracefs_enabled(&message, &fail, &umount)) {
+		if (fail) {
+			TH_LOG("Tracefs setup failed: %s", message);
+			ASSERT_FALSE(fail);
+		}
+		SKIP(return, "Skipping: %s", message);
+	}
+
+	self->umount = umount;
+
+	ASSERT_GE(cpu, 0);
+
+	ASSERT_EQ(tracefs_reset(), 0);
+
+	tracefs_write_int(TRACEFS_ROOT"/buffer_subbuf_size_kb", variant->subbuf_size);
+
+	ASSERT_EQ(tracefs_cpu_map(&self->map_desc, cpu), 0);
+
+	/*
+	 * Ensure generated events will be found on this very same ring-buffer.
+	 */
+	CPU_ZERO(&cpu_mask);
+	CPU_SET(cpu, &cpu_mask);
+	ASSERT_EQ(sched_setaffinity(0, sizeof(cpu_mask), &cpu_mask), 0);
+}
+
+FIXTURE_TEARDOWN(map)
+{
+	tracefs_reset();
+
+	if (self->umount)
+		tracefs_unmount();
+
+	tracefs_cpu_unmap(&self->map_desc);
+}
+
+TEST_F(map, meta_page_check)
+{
+	struct tracefs_cpu_map_desc *desc = &self->map_desc;
+	int cnt = 0;
+
+	ASSERT_EQ(desc->meta->entries, 0);
+	ASSERT_EQ(desc->meta->overrun, 0);
+	ASSERT_EQ(desc->meta->read, 0);
+
+	ASSERT_EQ(desc->meta->reader.id, 0);
+	ASSERT_EQ(desc->meta->reader.read, 0);
+
+	ASSERT_EQ(ioctl(desc->cpu_fd, TRACE_MMAP_IOCTL_GET_READER), 0);
+	ASSERT_EQ(desc->meta->reader.id, 0);
+
+	tracefs_write_int(TRACEFS_ROOT"/tracing_on", 1);
+	for (int i = 0; i < 16; i++)
+		tracefs_write_int(TRACEFS_ROOT"/trace_marker", i);
+again:
+	ASSERT_EQ(ioctl(desc->cpu_fd, TRACE_MMAP_IOCTL_GET_READER), 0);
+
+	ASSERT_EQ(desc->meta->entries, 16);
+	ASSERT_EQ(desc->meta->overrun, 0);
+	ASSERT_EQ(desc->meta->read, 16);
+
+	ASSERT_EQ(desc->meta->reader.id, 1);
+
+	if (!(cnt++))
+		goto again;
+}
+
+TEST_F(map, data_mmap)
+{
+	struct tracefs_cpu_map_desc *desc = &self->map_desc;
+	unsigned long meta_len, data_len;
+	void *data;
+
+	meta_len = desc->meta->meta_page_size;
+	data_len = desc->meta->subbuf_size * desc->meta->nr_subbufs;
+
+	/* Map all the available subbufs */
+	data = mmap(NULL, data_len, PROT_READ, MAP_SHARED,
+		    desc->cpu_fd, meta_len);
+	ASSERT_NE(data, MAP_FAILED);
+	munmap(data, data_len);
+
+	/* Map all the available subbufs - 1 */
+	data_len -= desc->meta->subbuf_size;
+	data = mmap(NULL, data_len, PROT_READ, MAP_SHARED,
+		    desc->cpu_fd, meta_len);
+	ASSERT_NE(data, MAP_FAILED);
+	munmap(data, data_len);
+
+	/* Overflow the available subbufs by 1 */
+	meta_len += desc->meta->subbuf_size * 2;
+	data = mmap(NULL, data_len, PROT_READ, MAP_SHARED,
+		    desc->cpu_fd, meta_len);
+	ASSERT_EQ(data, MAP_FAILED);
+}
+
+FIXTURE(snapshot) {
+	bool	umount;
+};
+
+FIXTURE_SETUP(snapshot)
+{
+	bool fail, umount;
+	struct stat sb;
+	char *message;
+
+	if (getuid() != 0)
+		SKIP(return, "Skipping: %s", "Please run the test as root");
+
+	if (stat(TRACEFS_ROOT"/snapshot", &sb))
+		SKIP(return, "Skipping: %s", "snapshot not available");
+
+	if (!tracefs_enabled(&message, &fail, &umount)) {
+		if (fail) {
+			TH_LOG("Tracefs setup failed: %s", message);
+			ASSERT_FALSE(fail);
+		}
+		SKIP(return, "Skipping: %s", message);
+	}
+
+	self->umount = umount;
+}
+
+FIXTURE_TEARDOWN(snapshot)
+{
+	__tracefs_write(TRACEFS_ROOT"/events/sched/sched_switch/trigger",
+			"!snapshot");
+	tracefs_reset();
+
+	if (self->umount)
+		tracefs_unmount();
+}
+
+TEST_F(snapshot, excludes_map)
+{
+	struct tracefs_cpu_map_desc map_desc;
+	int cpu = sched_getcpu();
+
+	ASSERT_GE(cpu, 0);
+	tracefs_write(TRACEFS_ROOT"/events/sched/sched_switch/trigger",
+		      "snapshot");
+	ASSERT_EQ(tracefs_cpu_map(&map_desc, cpu), -EBUSY);
+}
+
+TEST_F(snapshot, excluded_by_map)
+{
+	struct tracefs_cpu_map_desc map_desc;
+	int cpu = sched_getcpu();
+
+	ASSERT_EQ(tracefs_cpu_map(&map_desc, cpu), 0);
+
+	ASSERT_EQ(__tracefs_write(TRACEFS_ROOT"/events/sched/sched_switch/trigger",
+				  "snapshot"), -EBUSY);
+	ASSERT_EQ(__tracefs_write(TRACEFS_ROOT"/snapshot",
+				  "1"), -EBUSY);
+}
+
+TEST_HARNESS_MAIN