[PATCH v4 0/5] tracing: Introduce configurable ring sub buffer size

[PATCH v4 0/5] tracing: Introduce configurable ring sub buffer size

[Tzvetomir Stoyanov (VMware)]

From: "Tzvetomir Stoyanov (VMware)" <tstoyanov@vmware.com>

Currently the size of one sub buffer page is global for all buffers and it
is hard coded to one system page. The patch set introduces configurable
ring buffer sub page size, per ring buffer. A new user space interface is
introduced, which allows to change the sub page size of the ftrace buffer, per
ftrace instance.

[Resend to LKML]

v4 changes:
 - Do not change the sub buffer size of the temp event buffer per CPU,
   trace_buffered_event. That size if fixed to one system page.
v3 changes:
 - Fixed ring buffer readers and writers to work with custom sub buffer page
   size.
v2 changes:
 - Split code refactoring in a separate patch.
 - Changed the logic to work with order of system pages, instead of page
   size.
 - Fix freeing of the allocated pages, to take into account the actual order.
 - Code clean ups.
 - Bug fixes.

Tzvetomir Stoyanov (VMware) (5):
  [RFC] tracing: Refactor ring buffer implementation
  [RFC] tracing: Page size per ring buffer
  [RFC] tracing: Add interface for configuring trace sub buffer size
  [RFC] tracing: Set new size of the ring buffer sub page
  [RFC] tracing: Read and write to ring buffers with custom sub buffer
    size

 include/linux/ring_buffer.h          |  17 +-
 kernel/trace/ring_buffer.c           | 345 ++++++++++++++++++++-------
 kernel/trace/ring_buffer_benchmark.c |  10 +-
 kernel/trace/trace.c                 |  84 +++++--
 kernel/trace/trace.h                 |   1 +
 kernel/trace/trace_events.c          |  50 +++-
 6 files changed, 392 insertions(+), 115 deletions(-)

-- 
2.31.1

[PATCH v4 1/5] [RFC] tracing: Refactor ring buffer implementation

[Tzvetomir Stoyanov (VMware)]

In order to introduce sub-buffer size per ring buffer, some internal
refactoring is needed. As ring_buffer_print_page_header() will depend on
the trace_buffer structure, it is moved after the structure definition.

Signed-off-by: Tzvetomir Stoyanov (VMware) <tz.stoyanov@gmail.com>
---
 kernel/trace/ring_buffer.c | 59 +++++++++++++++++++-------------------
 1 file changed, 30 insertions(+), 29 deletions(-)

diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c
index 2699e9e562b1..cc34dbfdd29b 100644
--- a/kernel/trace/ring_buffer.c
+++ b/kernel/trace/ring_buffer.c
@@ -371,35 +371,6 @@ static inline int test_time_stamp(u64 delta)
 /* Max payload is BUF_PAGE_SIZE - header (8bytes) */
 #define BUF_MAX_DATA_SIZE (BUF_PAGE_SIZE - (sizeof(u32) * 2))
 
-int ring_buffer_print_page_header(struct trace_seq *s)
-{
-	struct buffer_data_page field;
-
-	trace_seq_printf(s, "\tfield: u64 timestamp;\t"
-			 "offset:0;\tsize:%u;\tsigned:%u;\n",
-			 (unsigned int)sizeof(field.time_stamp),
-			 (unsigned int)is_signed_type(u64));
-
-	trace_seq_printf(s, "\tfield: local_t commit;\t"
-			 "offset:%u;\tsize:%u;\tsigned:%u;\n",
-			 (unsigned int)offsetof(typeof(field), commit),
-			 (unsigned int)sizeof(field.commit),
-			 (unsigned int)is_signed_type(long));
-
-	trace_seq_printf(s, "\tfield: int overwrite;\t"
-			 "offset:%u;\tsize:%u;\tsigned:%u;\n",
-			 (unsigned int)offsetof(typeof(field), commit),
-			 1,
-			 (unsigned int)is_signed_type(long));
-
-	trace_seq_printf(s, "\tfield: char data;\t"
-			 "offset:%u;\tsize:%u;\tsigned:%u;\n",
-			 (unsigned int)offsetof(typeof(field), data),
-			 (unsigned int)BUF_PAGE_SIZE,
-			 (unsigned int)is_signed_type(char));
-
-	return !trace_seq_has_overflowed(s);
-}
 
 struct rb_irq_work {
 	struct irq_work			work;
@@ -559,6 +530,36 @@ struct ring_buffer_iter {
 	int				missed_events;
 };
 
+int ring_buffer_print_page_header(struct trace_seq *s)
+{
+	struct buffer_data_page field;
+
+	trace_seq_printf(s, "\tfield: u64 timestamp;\t"
+			 "offset:0;\tsize:%u;\tsigned:%u;\n",
+			 (unsigned int)sizeof(field.time_stamp),
+			 (unsigned int)is_signed_type(u64));
+
+	trace_seq_printf(s, "\tfield: local_t commit;\t"
+			 "offset:%u;\tsize:%u;\tsigned:%u;\n",
+			 (unsigned int)offsetof(typeof(field), commit),
+			 (unsigned int)sizeof(field.commit),
+			 (unsigned int)is_signed_type(long));
+
+	trace_seq_printf(s, "\tfield: int overwrite;\t"
+			 "offset:%u;\tsize:%u;\tsigned:%u;\n",
+			 (unsigned int)offsetof(typeof(field), commit),
+			 1,
+			 (unsigned int)is_signed_type(long));
+
+	trace_seq_printf(s, "\tfield: char data;\t"
+			 "offset:%u;\tsize:%u;\tsigned:%u;\n",
+			 (unsigned int)offsetof(typeof(field), data),
+			 (unsigned int)BUF_PAGE_SIZE,
+			 (unsigned int)is_signed_type(char));
+
+	return !trace_seq_has_overflowed(s);
+}
+
 #ifdef RB_TIME_32
 
 /*
-- 
2.31.1

[PATCH v4 2/5] [RFC] tracing: Page size per ring buffer

[Tzvetomir Stoyanov (VMware)]

Currently the size of one sub buffer page is global for all buffers and
it is hard coded to one system page. In order to introduce configurable
ring buffer sub page size, the internal logic should be refactored to
work with sub page size per ring buffer.

Signed-off-by: Tzvetomir Stoyanov (VMware) <tz.stoyanov@gmail.com>
---
 include/linux/ring_buffer.h |  2 +-
 kernel/trace/ring_buffer.c  | 66 ++++++++++++++++++++-----------------
 kernel/trace/trace.c        |  2 +-
 kernel/trace/trace.h        |  1 +
 kernel/trace/trace_events.c | 50 ++++++++++++++++++++++------
 5 files changed, 79 insertions(+), 42 deletions(-)

diff --git a/include/linux/ring_buffer.h b/include/linux/ring_buffer.h
index dac53fd3afea..d9a2e6e8fb79 100644
--- a/include/linux/ring_buffer.h
+++ b/include/linux/ring_buffer.h
@@ -200,7 +200,7 @@ int ring_buffer_read_page(struct trace_buffer *buffer, void **data_page,
 struct trace_seq;
 
 int ring_buffer_print_entry_header(struct trace_seq *s);
-int ring_buffer_print_page_header(struct trace_seq *s);
+int ring_buffer_print_page_header(struct trace_buffer *buffer, struct trace_seq *s);
 
 enum ring_buffer_flags {
 	RB_FL_OVERWRITE		= 1 << 0,
diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c
index cc34dbfdd29b..68fdeff449c3 100644
--- a/kernel/trace/ring_buffer.c
+++ b/kernel/trace/ring_buffer.c
@@ -366,12 +366,6 @@ static inline int test_time_stamp(u64 delta)
 	return 0;
 }
 
-#define BUF_PAGE_SIZE (PAGE_SIZE - BUF_PAGE_HDR_SIZE)
-
-/* Max payload is BUF_PAGE_SIZE - header (8bytes) */
-#define BUF_MAX_DATA_SIZE (BUF_PAGE_SIZE - (sizeof(u32) * 2))
-
-
 struct rb_irq_work {
 	struct irq_work			work;
 	wait_queue_head_t		waiters;
@@ -515,6 +509,9 @@ struct trace_buffer {
 
 	struct rb_irq_work		irq_work;
 	bool				time_stamp_abs;
+
+	unsigned int			subbuf_size;
+	unsigned int			max_data_size;
 };
 
 struct ring_buffer_iter {
@@ -530,7 +527,7 @@ struct ring_buffer_iter {
 	int				missed_events;
 };
 
-int ring_buffer_print_page_header(struct trace_seq *s)
+int ring_buffer_print_page_header(struct trace_buffer *buffer, struct trace_seq *s)
 {
 	struct buffer_data_page field;
 
@@ -554,7 +551,7 @@ int ring_buffer_print_page_header(struct trace_seq *s)
 	trace_seq_printf(s, "\tfield: char data;\t"
 			 "offset:%u;\tsize:%u;\tsigned:%u;\n",
 			 (unsigned int)offsetof(typeof(field), data),
-			 (unsigned int)BUF_PAGE_SIZE,
+			 (unsigned int)buffer->subbuf_size,
 			 (unsigned int)is_signed_type(char));
 
 	return !trace_seq_has_overflowed(s);
@@ -1726,7 +1723,13 @@ struct trace_buffer *__ring_buffer_alloc(unsigned long size, unsigned flags,
 	if (!zalloc_cpumask_var(&buffer->cpumask, GFP_KERNEL))
 		goto fail_free_buffer;
 
-	nr_pages = DIV_ROUND_UP(size, BUF_PAGE_SIZE);
+	/* Default buffer page size - one system page */
+	buffer->subbuf_size = PAGE_SIZE - BUF_PAGE_HDR_SIZE;
+
+	/* Max payload is buffer page size - header (8bytes) */
+	buffer->max_data_size = buffer->subbuf_size - (sizeof(u32) * 2);
+
+	nr_pages = DIV_ROUND_UP(size, buffer->subbuf_size);
 	buffer->flags = flags;
 	buffer->clock = trace_clock_local;
 	buffer->reader_lock_key = key;
@@ -1920,7 +1923,8 @@ rb_remove_pages(struct ring_buffer_per_cpu *cpu_buffer, unsigned long nr_pages)
 			 * Increment overrun to account for the lost events.
 			 */
 			local_add(page_entries, &cpu_buffer->overrun);
-			local_sub(BUF_PAGE_SIZE, &cpu_buffer->entries_bytes);
+			local_sub(cpu_buffer->buffer->subbuf_size,
+				  &cpu_buffer->entries_bytes);
 		}
 
 		/*
@@ -2042,7 +2046,7 @@ static void update_pages_handler(struct work_struct *work)
  * @size: the new size.
  * @cpu_id: the cpu buffer to resize
  *
- * Minimum size is 2 * BUF_PAGE_SIZE.
+ * Minimum size is 2 * buffer->subbuf_size.
  *
  * Returns 0 on success and < 0 on failure.
  */
@@ -2064,7 +2068,7 @@ int ring_buffer_resize(struct trace_buffer *buffer, unsigned long size,
 	    !cpumask_test_cpu(cpu_id, buffer->cpumask))
 		return 0;
 
-	nr_pages = DIV_ROUND_UP(size, BUF_PAGE_SIZE);
+	nr_pages = DIV_ROUND_UP(size, buffer->subbuf_size);
 
 	/* we need a minimum of two pages */
 	if (nr_pages < 2)
@@ -2291,7 +2295,7 @@ rb_iter_head_event(struct ring_buffer_iter *iter)
 	 */
 	barrier();
 
-	if ((iter->head + length) > commit || length > BUF_MAX_DATA_SIZE)
+	if ((iter->head + length) > commit || length > iter->cpu_buffer->buffer->max_data_size)
 		/* Writer corrupted the read? */
 		goto reset;
 
@@ -2404,7 +2408,8 @@ rb_handle_head_page(struct ring_buffer_per_cpu *cpu_buffer,
 		 * the counters.
 		 */
 		local_add(entries, &cpu_buffer->overrun);
-		local_sub(BUF_PAGE_SIZE, &cpu_buffer->entries_bytes);
+		local_sub(cpu_buffer->buffer->subbuf_size,
+			  &cpu_buffer->entries_bytes);
 
 		/*
 		 * The entries will be zeroed out when we move the
@@ -2523,6 +2528,7 @@ static inline void
 rb_reset_tail(struct ring_buffer_per_cpu *cpu_buffer,
 	      unsigned long tail, struct rb_event_info *info)
 {
+	unsigned long bsize = READ_ONCE(cpu_buffer->buffer->subbuf_size);
 	struct buffer_page *tail_page = info->tail_page;
 	struct ring_buffer_event *event;
 	unsigned long length = info->length;
@@ -2531,13 +2537,13 @@ rb_reset_tail(struct ring_buffer_per_cpu *cpu_buffer,
 	 * Only the event that crossed the page boundary
 	 * must fill the old tail_page with padding.
 	 */
-	if (tail >= BUF_PAGE_SIZE) {
+	if (tail >= bsize) {
 		/*
 		 * If the page was filled, then we still need
 		 * to update the real_end. Reset it to zero
 		 * and the reader will ignore it.
 		 */
-		if (tail == BUF_PAGE_SIZE)
+		if (tail == bsize)
 			tail_page->real_end = 0;
 
 		local_sub(length, &tail_page->write);
@@ -2547,7 +2553,7 @@ rb_reset_tail(struct ring_buffer_per_cpu *cpu_buffer,
 	event = __rb_page_index(tail_page, tail);
 
 	/* account for padding bytes */
-	local_add(BUF_PAGE_SIZE - tail, &cpu_buffer->entries_bytes);
+	local_add(bsize - tail, &cpu_buffer->entries_bytes);
 
 	/*
 	 * Save the original length to the meta data.
@@ -2567,7 +2573,7 @@ rb_reset_tail(struct ring_buffer_per_cpu *cpu_buffer,
 	 * If we are less than the minimum size, we don't need to
 	 * worry about it.
 	 */
-	if (tail > (BUF_PAGE_SIZE - RB_EVNT_MIN_SIZE)) {
+	if (tail > (bsize - RB_EVNT_MIN_SIZE)) {
 		/* No room for any events */
 
 		/* Mark the rest of the page with padding */
@@ -2579,13 +2585,13 @@ rb_reset_tail(struct ring_buffer_per_cpu *cpu_buffer,
 	}
 
 	/* Put in a discarded event */
-	event->array[0] = (BUF_PAGE_SIZE - tail) - RB_EVNT_HDR_SIZE;
+	event->array[0] = (bsize - tail) - RB_EVNT_HDR_SIZE;
 	event->type_len = RINGBUF_TYPE_PADDING;
 	/* time delta must be non zero */
 	event->time_delta = 1;
 
 	/* Set write to end of buffer */
-	length = (tail + length) - BUF_PAGE_SIZE;
+	length = (tail + length) - bsize;
 	local_sub(length, &tail_page->write);
 }
 
@@ -3477,7 +3483,7 @@ __rb_reserve_next(struct ring_buffer_per_cpu *cpu_buffer,
 	tail = write - info->length;
 
 	/* See if we shot pass the end of this buffer page */
-	if (unlikely(write > BUF_PAGE_SIZE)) {
+	if (unlikely(write > cpu_buffer->buffer->subbuf_size)) {
 		/* before and after may now different, fix it up*/
 		b_ok = rb_time_read(&cpu_buffer->before_stamp, &info->before);
 		a_ok = rb_time_read(&cpu_buffer->write_stamp, &info->after);
@@ -3686,7 +3692,7 @@ ring_buffer_lock_reserve(struct trace_buffer *buffer, unsigned long length)
 	if (unlikely(atomic_read(&cpu_buffer->record_disabled)))
 		goto out;
 
-	if (unlikely(length > BUF_MAX_DATA_SIZE))
+	if (unlikely(length > buffer->max_data_size))
 		goto out;
 
 	if (unlikely(trace_recursive_lock(cpu_buffer)))
@@ -3836,7 +3842,7 @@ int ring_buffer_write(struct trace_buffer *buffer,
 	if (atomic_read(&cpu_buffer->record_disabled))
 		goto out;
 
-	if (length > BUF_MAX_DATA_SIZE)
+	if (length > buffer->max_data_size)
 		goto out;
 
 	if (unlikely(trace_recursive_lock(cpu_buffer)))
@@ -4958,7 +4964,7 @@ ring_buffer_read_prepare(struct trace_buffer *buffer, int cpu, gfp_t flags)
 	if (!iter)
 		return NULL;
 
-	iter->event = kmalloc(BUF_MAX_DATA_SIZE, flags);
+	iter->event = kmalloc(buffer->max_data_size, flags);
 	if (!iter->event) {
 		kfree(iter);
 		return NULL;
@@ -5076,14 +5082,14 @@ unsigned long ring_buffer_size(struct trace_buffer *buffer, int cpu)
 {
 	/*
 	 * Earlier, this method returned
-	 *	BUF_PAGE_SIZE * buffer->nr_pages
+	 *	buffer->subbuf_size * buffer->nr_pages
 	 * Since the nr_pages field is now removed, we have converted this to
 	 * return the per cpu buffer value.
 	 */
 	if (!cpumask_test_cpu(cpu, buffer->cpumask))
 		return 0;
 
-	return BUF_PAGE_SIZE * buffer->buffers[cpu]->nr_pages;
+	return buffer->subbuf_size * buffer->buffers[cpu]->nr_pages;
 }
 EXPORT_SYMBOL_GPL(ring_buffer_size);
 
@@ -5619,7 +5625,7 @@ int ring_buffer_read_page(struct trace_buffer *buffer,
 	} else {
 		/* update the entry counter */
 		cpu_buffer->read += rb_page_entries(reader);
-		cpu_buffer->read_bytes += BUF_PAGE_SIZE;
+		cpu_buffer->read_bytes += buffer->subbuf_size;
 
 		/* swap the pages */
 		rb_init_page(bpage);
@@ -5650,7 +5656,7 @@ int ring_buffer_read_page(struct trace_buffer *buffer,
 		/* If there is room at the end of the page to save the
 		 * missed events, then record it there.
 		 */
-		if (BUF_PAGE_SIZE - commit >= sizeof(missed_events)) {
+		if (buffer->subbuf_size - commit >= sizeof(missed_events)) {
 			memcpy(&bpage->data[commit], &missed_events,
 			       sizeof(missed_events));
 			local_add(RB_MISSED_STORED, &bpage->commit);
@@ -5662,8 +5668,8 @@ int ring_buffer_read_page(struct trace_buffer *buffer,
 	/*
 	 * This page may be off to user land. Zero it out here.
 	 */
-	if (commit < BUF_PAGE_SIZE)
-		memset(&bpage->data[commit], 0, BUF_PAGE_SIZE - commit);
+	if (commit < buffer->subbuf_size)
+		memset(&bpage->data[commit], 0, buffer->subbuf_size - commit);
 
  out_unlock:
 	raw_spin_unlock_irqrestore(&cpu_buffer->reader_lock, flags);
diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c
index 88de94da596b..0eb8af875184 100644
--- a/kernel/trace/trace.c
+++ b/kernel/trace/trace.c
@@ -4877,7 +4877,7 @@ static int tracing_release(struct inode *inode, struct file *file)
 	return 0;
 }
 
-static int tracing_release_generic_tr(struct inode *inode, struct file *file)
+int tracing_release_generic_tr(struct inode *inode, struct file *file)
 {
 	struct trace_array *tr = inode->i_private;
 
diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h
index 38715aa6cfdf..101be5d43117 100644
--- a/kernel/trace/trace.h
+++ b/kernel/trace/trace.h
@@ -580,6 +580,7 @@ void tracing_reset_current(int cpu);
 void tracing_reset_all_online_cpus(void);
 int tracing_open_generic(struct inode *inode, struct file *filp);
 int tracing_open_generic_tr(struct inode *inode, struct file *filp);
+int tracing_release_generic_tr(struct inode *inode, struct file *file);
 bool tracing_is_disabled(void);
 bool tracer_tracing_is_on(struct trace_array *tr);
 void tracer_tracing_on(struct trace_array *tr);
diff --git a/kernel/trace/trace_events.c b/kernel/trace/trace_events.c
index 92be9cb1d7d4..7424c20514ec 100644
--- a/kernel/trace/trace_events.c
+++ b/kernel/trace/trace_events.c
@@ -1847,9 +1847,9 @@ subsystem_filter_write(struct file *filp, const char __user *ubuf, size_t cnt,
 }
 
 static ssize_t
-show_header(struct file *filp, char __user *ubuf, size_t cnt, loff_t *ppos)
+show_header_page_file(struct file *filp, char __user *ubuf, size_t cnt, loff_t *ppos)
 {
-	int (*func)(struct trace_seq *s) = filp->private_data;
+	struct trace_array *tr = filp->private_data;
 	struct trace_seq *s;
 	int r;
 
@@ -1862,7 +1862,31 @@ show_header(struct file *filp, char __user *ubuf, size_t cnt, loff_t *ppos)
 
 	trace_seq_init(s);
 
-	func(s);
+	ring_buffer_print_page_header(tr->array_buffer.buffer, s);
+	r = simple_read_from_buffer(ubuf, cnt, ppos,
+				    s->buffer, trace_seq_used(s));
+
+	kfree(s);
+
+	return r;
+}
+
+static ssize_t
+show_header_event_file(struct file *filp, char __user *ubuf, size_t cnt, loff_t *ppos)
+{
+	struct trace_seq *s;
+	int r;
+
+	if (*ppos)
+		return 0;
+
+	s = kmalloc(sizeof(*s), GFP_KERNEL);
+	if (!s)
+		return -ENOMEM;
+
+	trace_seq_init(s);
+
+	ring_buffer_print_entry_header(s);
 	r = simple_read_from_buffer(ubuf, cnt, ppos,
 				    s->buffer, trace_seq_used(s));
 
@@ -2117,10 +2141,18 @@ static const struct file_operations ftrace_tr_enable_fops = {
 	.release = subsystem_release,
 };
 
-static const struct file_operations ftrace_show_header_fops = {
-	.open = tracing_open_generic,
-	.read = show_header,
+static const struct file_operations ftrace_show_header_page_fops = {
+	.open = tracing_open_generic_tr,
+	.read = show_header_page_file,
+	.llseek = default_llseek,
+	.release = tracing_release_generic_tr,
+};
+
+static const struct file_operations ftrace_show_header_event_fops = {
+	.open = tracing_open_generic_tr,
+	.read = show_header_event_file,
 	.llseek = default_llseek,
+	.release = tracing_release_generic_tr,
 };
 
 static int
@@ -3481,14 +3513,12 @@ create_event_toplevel_files(struct dentry *parent, struct trace_array *tr)
 
 	/* ring buffer internal formats */
 	entry = trace_create_file("header_page", TRACE_MODE_READ, d_events,
-				  ring_buffer_print_page_header,
-				  &ftrace_show_header_fops);
+				  tr, &ftrace_show_header_page_fops);
 	if (!entry)
 		pr_warn("Could not create tracefs 'header_page' entry\n");
 
 	entry = trace_create_file("header_event", TRACE_MODE_READ, d_events,
-				  ring_buffer_print_entry_header,
-				  &ftrace_show_header_fops);
+				  tr, &ftrace_show_header_event_fops);
 	if (!entry)
 		pr_warn("Could not create tracefs 'header_event' entry\n");
 
-- 
2.31.1

[PATCH v4 3/5] [RFC] tracing: Add interface for configuring trace sub buffer size

[Tzvetomir Stoyanov (VMware)]

The trace ring buffer sub page size can be configured, per trace
instance. A new ftrace file "buffer_subbuf_order" is added to get and
set the size of the ring buffer sub page for current trace instance.
The size must be an order of system page size, that's why the new
interface works with system page order, instead of absolute page size:
0 means the ring buffer sub page is equal to 1 system page and so
forth:
0 - 1 system page
1 - 2 system pages
2 - 4 system pages
...
The ring buffer sub page size is limited between 1 and 128 system
pages. The default value is 1 system page.
New ring buffer APIs are introduced:
 ring_buffer_subbuf_order_set()
 ring_buffer_subbuf_order_get()
 ring_buffer_subbuf_size_get()

Signed-off-by: Tzvetomir Stoyanov (VMware) <tz.stoyanov@gmail.com>
---
 include/linux/ring_buffer.h |  4 ++
 kernel/trace/ring_buffer.c  | 73 +++++++++++++++++++++++++++++++++++++
 kernel/trace/trace.c        | 48 ++++++++++++++++++++++++
 3 files changed, 125 insertions(+)

diff --git a/include/linux/ring_buffer.h b/include/linux/ring_buffer.h
index d9a2e6e8fb79..9103462f6e85 100644
--- a/include/linux/ring_buffer.h
+++ b/include/linux/ring_buffer.h
@@ -202,6 +202,10 @@ struct trace_seq;
 int ring_buffer_print_entry_header(struct trace_seq *s);
 int ring_buffer_print_page_header(struct trace_buffer *buffer, struct trace_seq *s);
 
+int ring_buffer_subbuf_order_get(struct trace_buffer *buffer);
+int ring_buffer_subbuf_order_set(struct trace_buffer *buffer, int order);
+int ring_buffer_subbuf_size_get(struct trace_buffer *buffer);
+
 enum ring_buffer_flags {
 	RB_FL_OVERWRITE		= 1 << 0,
 };
diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c
index 68fdeff449c3..4aa5361a8f4c 100644
--- a/kernel/trace/ring_buffer.c
+++ b/kernel/trace/ring_buffer.c
@@ -511,6 +511,7 @@ struct trace_buffer {
 	bool				time_stamp_abs;
 
 	unsigned int			subbuf_size;
+	unsigned int			subbuf_order;
 	unsigned int			max_data_size;
 };
 
@@ -5679,6 +5680,78 @@ int ring_buffer_read_page(struct trace_buffer *buffer,
 }
 EXPORT_SYMBOL_GPL(ring_buffer_read_page);
 
+/**
+ * ring_buffer_subbuf_size_get - get size of the sub buffer.
+ * @buffer: the buffer to get the sub buffer size from
+ *
+ * Returns size of the sub buffer, in bytes.
+ */
+int ring_buffer_subbuf_size_get(struct trace_buffer *buffer)
+{
+	return buffer->subbuf_size + BUF_PAGE_HDR_SIZE;
+}
+EXPORT_SYMBOL_GPL(ring_buffer_subbuf_size_get);
+
+/**
+ * ring_buffer_subbuf_order_get - get order of system sub pages in one buffer page.
+ * @buffer: The ring_buffer to get the system sub page order from
+ *
+ * By default, one ring buffer sub page equals to one system page. This parameter
+ * is configurable, per ring buffer. The size of the ring buffer sub page can be
+ * extended, but must be an order of system page size.
+ *
+ * Returns the order of buffer sub page size, in system pages:
+ * 0 means the sub buffer size is 1 system page and so forth.
+ * In case of an error < 0 is returned.
+ */
+int ring_buffer_subbuf_order_get(struct trace_buffer *buffer)
+{
+	if (!buffer)
+		return -EINVAL;
+
+	return buffer->subbuf_order;
+}
+EXPORT_SYMBOL_GPL(ring_buffer_subbuf_order_get);
+
+/**
+ * ring_buffer_subbuf_order_set - set the size of ring buffer sub page.
+ * @buffer: The ring_buffer to set the new page size.
+ * @order: Order of the system pages in one sub buffer page
+ *
+ * By default, one ring buffer pages equals to one system page. This API can be
+ * used to set new size of the ring buffer page. The size must be order of
+ * system page size, that's why the input parameter @order is the order of
+ * system pages that are allocated for one ring buffer page:
+ *  0 - 1 system page
+ *  1 - 2 system pages
+ *  3 - 4 system pages
+ *  ...
+ *
+ * Returns 0 on success or < 0 in case of an error.
+ */
+int ring_buffer_subbuf_order_set(struct trace_buffer *buffer, int order)
+{
+	int psize;
+
+	if (!buffer || order < 0)
+		return -EINVAL;
+
+	if (buffer->subbuf_order == order)
+		return 0;
+
+	psize = (1 << order) * PAGE_SIZE;
+	if (psize <= BUF_PAGE_HDR_SIZE)
+		return -EINVAL;
+
+	buffer->subbuf_order = order;
+	buffer->subbuf_size = psize - BUF_PAGE_HDR_SIZE;
+
+	/* Todo: reset the buffer with the new page size */
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(ring_buffer_subbuf_order_set);
+
 /*
  * We only allocate new buffers, never free them if the CPU goes down.
  * If we were to free the buffer, then the user would lose any trace that was in
diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c
index 0eb8af875184..867a220b4ef2 100644
--- a/kernel/trace/trace.c
+++ b/kernel/trace/trace.c
@@ -9015,6 +9015,51 @@ static const struct file_operations buffer_percent_fops = {
 	.llseek		= default_llseek,
 };
 
+static ssize_t
+buffer_order_read(struct file *filp, char __user *ubuf, size_t cnt, loff_t *ppos)
+{
+	struct trace_array *tr = filp->private_data;
+	char buf[64];
+	int r;
+
+	r = sprintf(buf, "%d\n", ring_buffer_subbuf_order_get(tr->array_buffer.buffer));
+
+	return simple_read_from_buffer(ubuf, cnt, ppos, buf, r);
+}
+
+static ssize_t
+buffer_order_write(struct file *filp, const char __user *ubuf,
+		   size_t cnt, loff_t *ppos)
+{
+	struct trace_array *tr = filp->private_data;
+	unsigned long val;
+	int ret;
+
+	ret = kstrtoul_from_user(ubuf, cnt, 10, &val);
+	if (ret)
+		return ret;
+
+	/* limit between 1 and 128 system pages */
+	if (val < 0 || val > 7)
+		return -EINVAL;
+
+	ret = ring_buffer_subbuf_order_set(tr->array_buffer.buffer, val);
+	if (ret)
+		return ret;
+
+	(*ppos)++;
+
+	return cnt;
+}
+
+static const struct file_operations buffer_order_fops = {
+	.open		= tracing_open_generic_tr,
+	.read		= buffer_order_read,
+	.write		= buffer_order_write,
+	.release	= tracing_release_generic_tr,
+	.llseek		= default_llseek,
+};
+
 static struct dentry *trace_instance_dir;
 
 static void
@@ -9468,6 +9513,9 @@ init_tracer_tracefs(struct trace_array *tr, struct dentry *d_tracer)
 	trace_create_file("buffer_percent", TRACE_MODE_READ, d_tracer,
 			tr, &buffer_percent_fops);
 
+	trace_create_file("buffer_subbuf_order", TRACE_MODE_WRITE, d_tracer,
+			  tr, &buffer_order_fops);
+
 	create_trace_options_dir(tr);
 
 	trace_create_maxlat_file(tr, d_tracer);
-- 
2.31.1

[PATCH v4 4/5] [RFC] tracing: Set new size of the ring buffer sub page

[Tzvetomir Stoyanov (VMware)]

There are two approaches when changing the size of the ring buffer
sub page:
 1. Destroying all pages and allocating new pages with the new size.
 2. Allocating new pages, copying the content of the old pages before
    destroying them.
The first approach is easier, it is selected in the proposed
implementation. Changing the ring buffer sub page size is supposed to
not happen frequently. Usually, that size should be set only once,
when the buffer is not in use yet and is supposed to be empty.

Signed-off-by: Tzvetomir Stoyanov (VMware) <tz.stoyanov@gmail.com>
---
 kernel/trace/ring_buffer.c | 80 ++++++++++++++++++++++++++++++++++----
 1 file changed, 73 insertions(+), 7 deletions(-)

diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c
index 4aa5361a8f4c..a40fcb1cb299 100644
--- a/kernel/trace/ring_buffer.c
+++ b/kernel/trace/ring_buffer.c
@@ -323,6 +323,7 @@ struct buffer_page {
 	unsigned	 read;		/* index for next read */
 	local_t		 entries;	/* entries on this page */
 	unsigned long	 real_end;	/* real end of data */
+	unsigned	 order;		/* order of the page */
 	struct buffer_data_page *page;	/* Actual data page */
 };
 
@@ -352,7 +353,7 @@ static void rb_init_page(struct buffer_data_page *bpage)
  */
 static void free_buffer_page(struct buffer_page *bpage)
 {
-	free_page((unsigned long)bpage->page);
+	free_pages((unsigned long)bpage->page, bpage->order);
 	kfree(bpage);
 }
 
@@ -1563,10 +1564,12 @@ static int __rb_allocate_pages(struct ring_buffer_per_cpu *cpu_buffer,
 
 		list_add(&bpage->list, pages);
 
-		page = alloc_pages_node(cpu_to_node(cpu_buffer->cpu), mflags, 0);
+		page = alloc_pages_node(cpu_to_node(cpu_buffer->cpu), mflags,
+					cpu_buffer->buffer->subbuf_order);
 		if (!page)
 			goto free_pages;
 		bpage->page = page_address(page);
+		bpage->order = cpu_buffer->buffer->subbuf_order;
 		rb_init_page(bpage->page);
 
 		if (user_thread && fatal_signal_pending(current))
@@ -1645,7 +1648,8 @@ rb_allocate_cpu_buffer(struct trace_buffer *buffer, long nr_pages, int cpu)
 	rb_check_bpage(cpu_buffer, bpage);
 
 	cpu_buffer->reader_page = bpage;
-	page = alloc_pages_node(cpu_to_node(cpu), GFP_KERNEL, 0);
+
+	page = alloc_pages_node(cpu_to_node(cpu), GFP_KERNEL, cpu_buffer->buffer->subbuf_order);
 	if (!page)
 		goto fail_free_reader;
 	bpage->page = page_address(page);
@@ -1725,6 +1729,7 @@ struct trace_buffer *__ring_buffer_alloc(unsigned long size, unsigned flags,
 		goto fail_free_buffer;
 
 	/* Default buffer page size - one system page */
+	buffer->subbuf_order = 0;
 	buffer->subbuf_size = PAGE_SIZE - BUF_PAGE_HDR_SIZE;
 
 	/* Max payload is buffer page size - header (8bytes) */
@@ -5434,8 +5439,8 @@ void *ring_buffer_alloc_read_page(struct trace_buffer *buffer, int cpu)
 	if (bpage)
 		goto out;
 
-	page = alloc_pages_node(cpu_to_node(cpu),
-				GFP_KERNEL | __GFP_NORETRY, 0);
+	page = alloc_pages_node(cpu_to_node(cpu), GFP_KERNEL | __GFP_NORETRY,
+				cpu_buffer->buffer->subbuf_order);
 	if (!page)
 		return ERR_PTR(-ENOMEM);
 
@@ -5479,7 +5484,7 @@ void ring_buffer_free_read_page(struct trace_buffer *buffer, int cpu, void *data
 	local_irq_restore(flags);
 
  out:
-	free_page((unsigned long)bpage);
+	free_pages((unsigned long)bpage, buffer->subbuf_order);
 }
 EXPORT_SYMBOL_GPL(ring_buffer_free_read_page);
 
@@ -5731,7 +5736,13 @@ EXPORT_SYMBOL_GPL(ring_buffer_subbuf_order_get);
  */
 int ring_buffer_subbuf_order_set(struct trace_buffer *buffer, int order)
 {
+	struct ring_buffer_per_cpu **cpu_buffers;
+	int old_order, old_size;
+	int nr_pages;
 	int psize;
+	int bsize;
+	int err;
+	int cpu;
 
 	if (!buffer || order < 0)
 		return -EINVAL;
@@ -5743,12 +5754,67 @@ int ring_buffer_subbuf_order_set(struct trace_buffer *buffer, int order)
 	if (psize <= BUF_PAGE_HDR_SIZE)
 		return -EINVAL;
 
+	bsize = sizeof(void *) * buffer->cpus;
+	cpu_buffers = kzalloc(bsize, GFP_KERNEL);
+	if (!cpu_buffers)
+		return -ENOMEM;
+
+	old_order = buffer->subbuf_order;
+	old_size = buffer->subbuf_size;
+
+	/* prevent another thread from changing buffer sizes */
+	mutex_lock(&buffer->mutex);
+	atomic_inc(&buffer->record_disabled);
+
+	/* Make sure all commits have finished */
+	synchronize_rcu();
+
 	buffer->subbuf_order = order;
 	buffer->subbuf_size = psize - BUF_PAGE_HDR_SIZE;
 
-	/* Todo: reset the buffer with the new page size */
+	/* Make sure all new buffers are allocated, before deleting the old ones */
+	for_each_buffer_cpu(buffer, cpu) {
+		if (!cpumask_test_cpu(cpu, buffer->cpumask))
+			continue;
+
+		nr_pages = buffer->buffers[cpu]->nr_pages;
+		cpu_buffers[cpu] = rb_allocate_cpu_buffer(buffer, nr_pages, cpu);
+		if (!cpu_buffers[cpu]) {
+			err = -ENOMEM;
+			goto error;
+		}
+	}
+
+	for_each_buffer_cpu(buffer, cpu) {
+		if (!cpumask_test_cpu(cpu, buffer->cpumask))
+			continue;
+
+		rb_free_cpu_buffer(buffer->buffers[cpu]);
+		buffer->buffers[cpu] = cpu_buffers[cpu];
+	}
+
+	atomic_dec(&buffer->record_disabled);
+	mutex_unlock(&buffer->mutex);
+
+	kfree(cpu_buffers);
 
 	return 0;
+
+error:
+	buffer->subbuf_order = old_order;
+	buffer->subbuf_size = old_size;
+
+	atomic_dec(&buffer->record_disabled);
+	mutex_unlock(&buffer->mutex);
+
+	for_each_buffer_cpu(buffer, cpu) {
+		if (!cpu_buffers[cpu])
+			continue;
+		kfree(cpu_buffers[cpu]);
+	}
+	kfree(cpu_buffers);
+
+	return err;
 }
 EXPORT_SYMBOL_GPL(ring_buffer_subbuf_order_set);
 
-- 
2.31.1

[PATCH v4 5/5] [RFC] tracing: Read and write to ring buffers with custom sub buffer size

[Tzvetomir Stoyanov (VMware)]

As the size of the ring sub buffer page can be changed dynamically,
the logic that reads and writes to the buffer should be fixed to take
that into account. Some internal ring buffer APIs are changed:
 ring_buffer_alloc_read_page()
 ring_buffer_free_read_page()
 ring_buffer_read_page()
A new API is introduced:
 ring_buffer_read_page_data()

Signed-off-by: Tzvetomir Stoyanov (VMware) <tz.stoyanov@gmail.com>
---
 include/linux/ring_buffer.h          | 11 ++--
 kernel/trace/ring_buffer.c           | 75 ++++++++++++++++++++--------
 kernel/trace/ring_buffer_benchmark.c | 10 ++--
 kernel/trace/trace.c                 | 34 +++++++------
 4 files changed, 89 insertions(+), 41 deletions(-)

diff --git a/include/linux/ring_buffer.h b/include/linux/ring_buffer.h
index 9103462f6e85..8d6807e3865d 100644
--- a/include/linux/ring_buffer.h
+++ b/include/linux/ring_buffer.h
@@ -192,10 +192,15 @@ bool ring_buffer_time_stamp_abs(struct trace_buffer *buffer);
 size_t ring_buffer_nr_pages(struct trace_buffer *buffer, int cpu);
 size_t ring_buffer_nr_dirty_pages(struct trace_buffer *buffer, int cpu);
 
-void *ring_buffer_alloc_read_page(struct trace_buffer *buffer, int cpu);
-void ring_buffer_free_read_page(struct trace_buffer *buffer, int cpu, void *data);
-int ring_buffer_read_page(struct trace_buffer *buffer, void **data_page,
+struct buffer_data_read_page;
+struct buffer_data_read_page *
+ring_buffer_alloc_read_page(struct trace_buffer *buffer, int cpu);
+void ring_buffer_free_read_page(struct trace_buffer *buffer, int cpu,
+				struct buffer_data_read_page *page);
+int ring_buffer_read_page(struct trace_buffer *buffer,
+			  struct buffer_data_read_page *data_page,
 			  size_t len, int cpu, int full);
+void *ring_buffer_read_page_data(struct buffer_data_read_page *page);
 
 struct trace_seq;
 
diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c
index a40fcb1cb299..fd22e0fc7195 100644
--- a/kernel/trace/ring_buffer.c
+++ b/kernel/trace/ring_buffer.c
@@ -309,6 +309,11 @@ struct buffer_data_page {
 	unsigned char	 data[] RB_ALIGN_DATA;	/* data of buffer page */
 };
 
+struct buffer_data_read_page {
+	unsigned		order;	/* order of the page */
+	struct buffer_data_page	*data;	/* actual data, stored in this page */
+};
+
 /*
  * Note, the buffer_page list must be first. The buffer pages
  * are allocated in cache lines, which means that each buffer
@@ -5414,40 +5419,48 @@ EXPORT_SYMBOL_GPL(ring_buffer_swap_cpu);
  * Returns:
  *  The page allocated, or ERR_PTR
  */
-void *ring_buffer_alloc_read_page(struct trace_buffer *buffer, int cpu)
+struct buffer_data_read_page *
+ring_buffer_alloc_read_page(struct trace_buffer *buffer, int cpu)
 {
 	struct ring_buffer_per_cpu *cpu_buffer;
-	struct buffer_data_page *bpage = NULL;
+	struct buffer_data_read_page *bpage = NULL;
 	unsigned long flags;
 	struct page *page;
 
 	if (!cpumask_test_cpu(cpu, buffer->cpumask))
 		return ERR_PTR(-ENODEV);
 
+	bpage = kzalloc(sizeof(*bpage), GFP_KERNEL);
+	if (!bpage)
+		return ERR_PTR(-ENOMEM);
+
+	bpage->order = buffer->subbuf_order;
 	cpu_buffer = buffer->buffers[cpu];
 	local_irq_save(flags);
 	arch_spin_lock(&cpu_buffer->lock);
 
 	if (cpu_buffer->free_page) {
-		bpage = cpu_buffer->free_page;
+		bpage->data = cpu_buffer->free_page;
 		cpu_buffer->free_page = NULL;
 	}
 
 	arch_spin_unlock(&cpu_buffer->lock);
 	local_irq_restore(flags);
 
-	if (bpage)
+	if (bpage->data)
 		goto out;
 
 	page = alloc_pages_node(cpu_to_node(cpu), GFP_KERNEL | __GFP_NORETRY,
 				cpu_buffer->buffer->subbuf_order);
-	if (!page)
+	if (!page) {
+		kfree(bpage);
 		return ERR_PTR(-ENOMEM);
+	}
 
-	bpage = page_address(page);
+	bpage->data = page_address(page);
 
  out:
-	rb_init_page(bpage);
+	rb_init_page(bpage->data);
 
 	return bpage;
 }
@@ -5457,19 +5470,24 @@ EXPORT_SYMBOL_GPL(ring_buffer_alloc_read_page);
  * ring_buffer_free_read_page - free an allocated read page
  * @buffer: the buffer the page was allocate for
  * @cpu: the cpu buffer the page came from
- * @data: the page to free
+ * @page: the page to free
  *
  * Free a page allocated from ring_buffer_alloc_read_page.
  */
-void ring_buffer_free_read_page(struct trace_buffer *buffer, int cpu, void *data)
+void ring_buffer_free_read_page(struct trace_buffer *buffer, int cpu,
+				struct buffer_data_read_page *data_page)
 {
 	struct ring_buffer_per_cpu *cpu_buffer = buffer->buffers[cpu];
-	struct buffer_data_page *bpage = data;
+	struct buffer_data_page *bpage = data_page->data;
 	struct page *page = virt_to_page(bpage);
 	unsigned long flags;
 
-	/* If the page is still in use someplace else, we can't reuse it */
-	if (page_ref_count(page) > 1)
+	/*
+	 * If the page is still in use someplace else, or order of the page
+	 * is different from the subbuffer order of the buffer -
+	 * we can't reuse it
+	 */
+	if (page_ref_count(page) > 1 || data_page->order != buffer->subbuf_order)
 		goto out;
 
 	local_irq_save(flags);
@@ -5484,7 +5502,8 @@ void ring_buffer_free_read_page(struct trace_buffer *buffer, int cpu, void *data
 	local_irq_restore(flags);
 
  out:
-	free_pages((unsigned long)bpage, buffer->subbuf_order);
+	free_pages((unsigned long)bpage, data_page->order);
+	kfree(data_page);
 }
 EXPORT_SYMBOL_GPL(ring_buffer_free_read_page);
 
@@ -5505,9 +5524,10 @@ EXPORT_SYMBOL_GPL(ring_buffer_free_read_page);
  *	rpage = ring_buffer_alloc_read_page(buffer, cpu);
  *	if (IS_ERR(rpage))
  *		return PTR_ERR(rpage);
- *	ret = ring_buffer_read_page(buffer, &rpage, len, cpu, 0);
+ *	ret = ring_buffer_read_page(buffer, rpage, len, cpu, 0);
  *	if (ret >= 0)
- *		process_page(rpage, ret);
+ *		process_page(ring_buffer_read_page_data(rpage), ret);
+ *	ring_buffer_free_read_page(buffer, cpu, rpage);
  *
  * When @full is set, the function will not return true unless
  * the writer is off the reader page.
@@ -5522,7 +5542,8 @@ EXPORT_SYMBOL_GPL(ring_buffer_free_read_page);
  *  <0 if no data has been transferred.
  */
 int ring_buffer_read_page(struct trace_buffer *buffer,
-			  void **data_page, size_t len, int cpu, int full)
+			  struct buffer_data_read_page *data_page,
+			  size_t len, int cpu, int full)
 {
 	struct ring_buffer_per_cpu *cpu_buffer = buffer->buffers[cpu];
 	struct ring_buffer_event *event;
@@ -5547,10 +5568,12 @@ int ring_buffer_read_page(struct trace_buffer *buffer,
 
 	len -= BUF_PAGE_HDR_SIZE;
 
-	if (!data_page)
+	if (!data_page || !data_page->data)
+		goto out;
+	if (data_page->order != buffer->subbuf_order)
 		goto out;
 
-	bpage = *data_page;
+	bpage = data_page->data;
 	if (!bpage)
 		goto out;
 
@@ -5636,11 +5659,11 @@ int ring_buffer_read_page(struct trace_buffer *buffer,
 		/* swap the pages */
 		rb_init_page(bpage);
 		bpage = reader->page;
-		reader->page = *data_page;
+		reader->page = data_page->data;
 		local_set(&reader->write, 0);
 		local_set(&reader->entries, 0);
 		reader->read = 0;
-		*data_page = bpage;
+		data_page->data = bpage;
 
 		/*
 		 * Use the real_end for the data size,
@@ -5685,6 +5708,18 @@ int ring_buffer_read_page(struct trace_buffer *buffer,
 }
 EXPORT_SYMBOL_GPL(ring_buffer_read_page);
 
+/**
+ * ring_buffer_read_page_data - get pointer to the data in the page.
+ * @page:  the page to get the data from
+ *
+ * Returns pointer to the actual data in this page.
+ */
+void *ring_buffer_read_page_data(struct buffer_data_read_page *page)
+{
+	return page->data;
+}
+EXPORT_SYMBOL_GPL(ring_buffer_read_page_data);
+
 /**
  * ring_buffer_subbuf_size_get - get size of the sub buffer.
  * @buffer: the buffer to get the sub buffer size from
diff --git a/kernel/trace/ring_buffer_benchmark.c b/kernel/trace/ring_buffer_benchmark.c
index 78e576575b79..7202d6d650e6 100644
--- a/kernel/trace/ring_buffer_benchmark.c
+++ b/kernel/trace/ring_buffer_benchmark.c
@@ -104,10 +104,11 @@ static enum event_status read_event(int cpu)
 
 static enum event_status read_page(int cpu)
 {
+	struct buffer_data_read_page *bpage;
 	struct ring_buffer_event *event;
 	struct rb_page *rpage;
 	unsigned long commit;
-	void *bpage;
+	int page_size;
 	int *entry;
 	int ret;
 	int inc;
@@ -117,14 +118,15 @@ static enum event_status read_page(int cpu)
 	if (IS_ERR(bpage))
 		return EVENT_DROPPED;
 
-	ret = ring_buffer_read_page(buffer, &bpage, PAGE_SIZE, cpu, 1);
+	page_size = ring_buffer_subbuf_size_get(buffer);
+	ret = ring_buffer_read_page(buffer, bpage, page_size, cpu, 1);
 	if (ret >= 0) {
-		rpage = bpage;
+		rpage = ring_buffer_read_page_data(bpage);
 		/* The commit may have missed event flags set, clear them */
 		commit = local_read(&rpage->commit) & 0xfffff;
 		for (i = 0; i < commit && !test_error ; i += inc) {
 
-			if (i >= (PAGE_SIZE - offsetof(struct rb_page, data))) {
+			if (i >= (page_size - offsetof(struct rb_page, data))) {
 				TEST_ERROR();
 				break;
 			}
diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c
index 867a220b4ef2..edcf30ea1d25 100644
--- a/kernel/trace/trace.c
+++ b/kernel/trace/trace.c
@@ -8004,6 +8004,8 @@ tracing_buffers_read(struct file *filp, char __user *ubuf,
 {
 	struct ftrace_buffer_info *info = filp->private_data;
 	struct trace_iterator *iter = &info->iter;
+	void *trace_data;
+	int page_size;
 	ssize_t ret = 0;
 	ssize_t size;
 
@@ -8015,6 +8017,8 @@ tracing_buffers_read(struct file *filp, char __user *ubuf,
 		return -EBUSY;
 #endif
 
+	page_size = ring_buffer_subbuf_size_get(iter->array_buffer->buffer);
+
 	if (!info->spare) {
 		info->spare = ring_buffer_alloc_read_page(iter->array_buffer->buffer,
 							  iter->cpu_file);
@@ -8029,13 +8033,13 @@ tracing_buffers_read(struct file *filp, char __user *ubuf,
 		return ret;
 
 	/* Do we have previous read data to read? */
-	if (info->read < PAGE_SIZE)
+	if (info->read < page_size)
 		goto read;
 
  again:
 	trace_access_lock(iter->cpu_file);
 	ret = ring_buffer_read_page(iter->array_buffer->buffer,
-				    &info->spare,
+				    info->spare,
 				    count,
 				    iter->cpu_file, 0);
 	trace_access_unlock(iter->cpu_file);
@@ -8056,11 +8060,11 @@ tracing_buffers_read(struct file *filp, char __user *ubuf,
 
 	info->read = 0;
  read:
-	size = PAGE_SIZE - info->read;
+	size = page_size - info->read;
 	if (size > count)
 		size = count;
-
-	ret = copy_to_user(ubuf, info->spare + info->read, size);
+	trace_data = ring_buffer_read_page_data(info->spare);
+	ret = copy_to_user(ubuf, trace_data + info->read, size);
 	if (ret == size)
 		return -EFAULT;
 
@@ -8165,6 +8169,7 @@ tracing_buffers_splice_read(struct file *file, loff_t *ppos,
 		.spd_release	= buffer_spd_release,
 	};
 	struct buffer_ref *ref;
+	int page_size;
 	int entries, i;
 	ssize_t ret = 0;
 
@@ -8173,13 +8178,14 @@ tracing_buffers_splice_read(struct file *file, loff_t *ppos,
 		return -EBUSY;
 #endif
 
-	if (*ppos & (PAGE_SIZE - 1))
+	page_size = ring_buffer_subbuf_size_get(iter->array_buffer->buffer);
+	if (*ppos & (page_size - 1))
 		return -EINVAL;
 
-	if (len & (PAGE_SIZE - 1)) {
-		if (len < PAGE_SIZE)
+	if (len & (page_size - 1)) {
+		if (len < page_size)
 			return -EINVAL;
-		len &= PAGE_MASK;
+		len &= (~(page_size - 1));
 	}
 
 	if (splice_grow_spd(pipe, &spd))
@@ -8189,7 +8195,7 @@ tracing_buffers_splice_read(struct file *file, loff_t *ppos,
 	trace_access_lock(iter->cpu_file);
 	entries = ring_buffer_entries_cpu(iter->array_buffer->buffer, iter->cpu_file);
 
-	for (i = 0; i < spd.nr_pages_max && len && entries; i++, len -= PAGE_SIZE) {
+	for (i = 0; i < spd.nr_pages_max && len && entries; i++, len -= page_size) {
 		struct page *page;
 		int r;
 
@@ -8210,7 +8216,7 @@ tracing_buffers_splice_read(struct file *file, loff_t *ppos,
 		}
 		ref->cpu = iter->cpu_file;
 
-		r = ring_buffer_read_page(ref->buffer, &ref->page,
+		r = ring_buffer_read_page(ref->buffer, ref->page,
 					  len, iter->cpu_file, 1);
 		if (r < 0) {
 			ring_buffer_free_read_page(ref->buffer, ref->cpu,
@@ -8219,14 +8225,14 @@ tracing_buffers_splice_read(struct file *file, loff_t *ppos,
 			break;
 		}
 
-		page = virt_to_page(ref->page);
+		page = virt_to_page(ring_buffer_read_page_data(ref->page));
 
 		spd.pages[i] = page;
-		spd.partial[i].len = PAGE_SIZE;
+		spd.partial[i].len = page_size;
 		spd.partial[i].offset = 0;
 		spd.partial[i].private = (unsigned long)ref;
 		spd.nr_pages++;
-		*ppos += PAGE_SIZE;
+		*ppos += page_size;
 
 		entries = ring_buffer_entries_cpu(iter->array_buffer->buffer, iter->cpu_file);
 	}
-- 
2.31.1

[PATCH v4 1/5] [RFC] tracing: Refactor ring buffer implementation

[Tzvetomir Stoyanov (VMware)]

In order to introduce sub-buffer size per ring buffer, some internal
refactoring is needed. As ring_buffer_print_page_header() will depend on
the trace_buffer structure, it is moved after the structure definition.

Signed-off-by: Tzvetomir Stoyanov (VMware) <tz.stoyanov@gmail.com>
---
 kernel/trace/ring_buffer.c | 59 +++++++++++++++++++-------------------
 1 file changed, 30 insertions(+), 29 deletions(-)

diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c
index 2699e9e562b1..cc34dbfdd29b 100644
--- a/kernel/trace/ring_buffer.c
+++ b/kernel/trace/ring_buffer.c
@@ -371,35 +371,6 @@ static inline int test_time_stamp(u64 delta)
 /* Max payload is BUF_PAGE_SIZE - header (8bytes) */
 #define BUF_MAX_DATA_SIZE (BUF_PAGE_SIZE - (sizeof(u32) * 2))
 
-int ring_buffer_print_page_header(struct trace_seq *s)
-{
-	struct buffer_data_page field;
-
-	trace_seq_printf(s, "\tfield: u64 timestamp;\t"
-			 "offset:0;\tsize:%u;\tsigned:%u;\n",
-			 (unsigned int)sizeof(field.time_stamp),
-			 (unsigned int)is_signed_type(u64));
-
-	trace_seq_printf(s, "\tfield: local_t commit;\t"
-			 "offset:%u;\tsize:%u;\tsigned:%u;\n",
-			 (unsigned int)offsetof(typeof(field), commit),
-			 (unsigned int)sizeof(field.commit),
-			 (unsigned int)is_signed_type(long));
-
-	trace_seq_printf(s, "\tfield: int overwrite;\t"
-			 "offset:%u;\tsize:%u;\tsigned:%u;\n",
-			 (unsigned int)offsetof(typeof(field), commit),
-			 1,
-			 (unsigned int)is_signed_type(long));
-
-	trace_seq_printf(s, "\tfield: char data;\t"
-			 "offset:%u;\tsize:%u;\tsigned:%u;\n",
-			 (unsigned int)offsetof(typeof(field), data),
-			 (unsigned int)BUF_PAGE_SIZE,
-			 (unsigned int)is_signed_type(char));
-
-	return !trace_seq_has_overflowed(s);
-}
 
 struct rb_irq_work {
 	struct irq_work			work;
@@ -559,6 +530,36 @@ struct ring_buffer_iter {
 	int				missed_events;
 };
 
+int ring_buffer_print_page_header(struct trace_seq *s)
+{
+	struct buffer_data_page field;
+
+	trace_seq_printf(s, "\tfield: u64 timestamp;\t"
+			 "offset:0;\tsize:%u;\tsigned:%u;\n",
+			 (unsigned int)sizeof(field.time_stamp),
+			 (unsigned int)is_signed_type(u64));
+
+	trace_seq_printf(s, "\tfield: local_t commit;\t"
+			 "offset:%u;\tsize:%u;\tsigned:%u;\n",
+			 (unsigned int)offsetof(typeof(field), commit),
+			 (unsigned int)sizeof(field.commit),
+			 (unsigned int)is_signed_type(long));
+
+	trace_seq_printf(s, "\tfield: int overwrite;\t"
+			 "offset:%u;\tsize:%u;\tsigned:%u;\n",
+			 (unsigned int)offsetof(typeof(field), commit),
+			 1,
+			 (unsigned int)is_signed_type(long));
+
+	trace_seq_printf(s, "\tfield: char data;\t"
+			 "offset:%u;\tsize:%u;\tsigned:%u;\n",
+			 (unsigned int)offsetof(typeof(field), data),
+			 (unsigned int)BUF_PAGE_SIZE,
+			 (unsigned int)is_signed_type(char));
+
+	return !trace_seq_has_overflowed(s);
+}
+
 #ifdef RB_TIME_32
 
 /*
-- 
2.31.1