Skip to contentSkip to conversation

ring-buffer: Some minor fixes for the latest changes

GIT PULL New Steven Rostedt 2 participants 4 messages May 25, 2010 – May 28, 2010

Subsystems (2)
lore
Share & tools
lore

Steven Rostedt

author maintainer
#0 lore raw

Ingo,

Please pull the latest tip/perf/core tree, which can be found at:

git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-2.6-trace.git
tip/perf/core

Steven Rostedt (2):
ring-buffer: Reset "real_end" when page is filled
ring-buffer: Move zeroing out excess in page to ring buffer code

----

 kernel/trace/ring_buffer.c |   19 +++++++++++++++++--
 kernel/trace/trace.c       |    6 ------
 2 files changed, 17 insertions(+), 8 deletions(-)
author maintainer
#2 lore raw

PATCH 2/2 ring-buffer: Move zeroing out excess in page to ring buffer code

From: Steven Rostedt <[email protected]>

Currently the trace splice code zeros out the excess bytes in the page before sending it off to userspace.

This is to make sure userspace is not getting anything it should not be when reading the pages, because the excess data was never initialized to zero before writing (for perfomance reasons).

But the splice code has no business in doing this work, it should be done by the ring buffer. With the latest changes for recording lost events, the splice code gets it wrong anyway.

Move the zeroing out of excess bytes into the ring buffer code.

---
 kernel/trace/ring_buffer.c |   11 +++++++++--
 kernel/trace/trace.c       |    6 ------
 2 files changed, 9 insertions(+), 8 deletions(-)
kernel/trace/ring_buffer.c +9 -2
index b0702ff..1da7b6e 100644
@@ -3902,12 +3902,12 @@ int ring_buffer_read_page(struct ring_buffer *buffer,
 	ret = read;
 
 	cpu_buffer->lost_events = 0;
+
+	commit = local_read(&bpage->commit);
 	/*
 	 * Set a flag in the commit field if we lost events
 	 */
 	if (missed_events) {
-		commit = local_read(&bpage->commit);
-
 		/* If there is room at the end of the page to save the
 		 * missed events, then record it there.
 		 */
@@ -3915,10 +3915,17 @@ int ring_buffer_read_page(struct ring_buffer *buffer,
 			memcpy(&bpage->data[commit], &missed_events,
 			       sizeof(missed_events));
 			local_add(RB_MISSED_STORED, &bpage->commit);
+			commit += sizeof(missed_events);
 		}
 		local_add(RB_MISSED_EVENTS, &bpage->commit);
 	}
 
+	/*
+	 * 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);
+
  out_unlock:
 	spin_unlock_irqrestore(&cpu_buffer->reader_lock, flags);
 
kernel/trace/trace.c +0 -6
index ba0ec81..95d0b1a 100644
@@ -3661,7 +3661,6 @@ tracing_buffers_read(struct file *filp, char __user *ubuf,
 		     size_t count, loff_t *ppos)
 {
 	struct ftrace_buffer_info *info = filp->private_data;
-	unsigned int pos;
 	ssize_t ret;
 	size_t size;
 
@@ -3688,11 +3687,6 @@ tracing_buffers_read(struct file *filp, char __user *ubuf,
 	if (ret < 0)
 		return 0;
 
-	pos = ring_buffer_page_len(info->spare);
-
-	if (pos < PAGE_SIZE)
-		memset(info->spare + pos, 0, PAGE_SIZE - pos);
-
 read:
 	size = PAGE_SIZE - info->read;
 	if (size > count)
Signature · 1 line
1.7.0
author maintainer
#1 lore raw

PATCH 1/2 ring-buffer: Reset "real_end" when page is filled

In reply to Steven Rostedt

From: Steven Rostedt <[email protected]>

The code to store the "lost events" requires knowing the real end of the page. Since the 'commit' includes the padding at the end of a page a "real_end" variable was used to keep track of the end not including the padding.

If events were lost, the reader can place the count of events in the padded area if there is enough room.

The bug this patch fixes is that when we fill the page we do not reset the real_end variable, and if the writer had wrapped a few times, the real_end would be incorrect.

This patch simply resets the real_end if the page was filled.

---
 kernel/trace/ring_buffer.c |    8 ++++++++
 1 files changed, 8 insertions(+), 0 deletions(-)
kernel/trace/ring_buffer.c +8 -0
index 7f6059c..b0702ff 100644
@@ -1768,6 +1768,14 @@ rb_reset_tail(struct ring_buffer_per_cpu *cpu_buffer,
 	 * must fill the old tail_page with padding.
 	 */
 	if (tail >= BUF_PAGE_SIZE) {
+		/*
+		 * 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)
+			tail_page->real_end = 0;
+
 		local_sub(length, &tail_page->write);
 		return;
 	}
Signature · 1 line
1.7.0
#3 lore raw
In reply to Steven Rostedt
Quoted 8 lines · Steven Rostedt

Ingo,

Please pull the latest tip/perf/core tree, which can be found at:

git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-2.6-trace.git
tip/perf/core

Quoted 8 lines

Steven Rostedt (2):
ring-buffer: Reset "real_end" when page is filled
ring-buffer: Move zeroing out excess in page to ring buffer code

----

 kernel/trace/ring_buffer.c |   19 +++++++++++++++++--
 kernel/trace/trace.c       |    6 ------
 2 files changed, 17 insertions(+), 8 deletions(-)

Pulled, thanks a lot Steve!

Ingo