Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1532386 > unrolled thread

[for-next][PATCH 8/8] ring-buffer: Force rb_end_commit() and rb_set_commit_to_write() inline

Back to article view | Back to linux.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.

Contents

  [for-next][PATCH 8/8] ring-buffer: Force rb_end_commit() and rb_set_commit_to_write()  inline Steven Rostedt <rostedt@goodmis.org> - 2016-11-29 16:10 +0100

#1532386 — [for-next][PATCH 8/8] ring-buffer: Force rb_end_commit() and rb_set_commit_to_write() inline

From: "Steven Rostedt (Red Hat)" <rostedt@goodmis.org>

Both rb_end_commit() and rb_set_commit_to_write() are in the fast path of
the ring buffer recording. Make sure they are always inlined.

Link: http://lkml.kernel.org/r/20161121183700.GW26852@two.firstfloor.org

Reported-by: Andi Kleen <andi@firstfloor.org>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
---
 kernel/trace/ring_buffer.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c
index 04579a0c7711..f17476f9d7f4 100644
--- a/kernel/trace/ring_buffer.c
+++ b/kernel/trace/ring_buffer.c
@@ -2386,7 +2386,7 @@ static void rb_start_commit(struct ring_buffer_per_cpu *cpu_buffer)
 	local_inc(&cpu_buffer->commits);
 }
 
-static void
+static __always_inline void
 rb_set_commit_to_write(struct ring_buffer_per_cpu *cpu_buffer)
 {
 	unsigned long max_count;
@@ -2441,7 +2441,7 @@ rb_set_commit_to_write(struct ring_buffer_per_cpu *cpu_buffer)
 		goto again;
 }
 
-static inline void rb_end_commit(struct ring_buffer_per_cpu *cpu_buffer)
+static __always_inline void rb_end_commit(struct ring_buffer_per_cpu *cpu_buffer)
 {
 	unsigned long commits;
 
-- 
2.10.2

[toc] | [standalone]

Back to top | Article view | linux.kernel

csiph-web