Messages in this thread | | | Date | Mon, 13 Apr 2015 21:29:11 -0500 | From | "Suresh E. Warrier" <> | Subject | [PATCH 0/2] ring-buffer: More precise timestamps for nested writes |
| |
This patch set provides supports a way to record precise timestamps for nested writes and a tunable to enable the behavior.
Nested writes occur when a trace writer is preempted, usually through an interrupt and the interrupt handler also writes its own trace. It has been noticed on large systems with lots of interrupts, several hundred trace events can be recorded with the same time stamp.
Fixing this requires a small critical section in the trace write code in which interrupts are disabled. This is an expensive operation for many architectures. Hence, the need for a kernel tunable that allows the feature to be enabled by admins only when needed.
Suresh Warrier (2): ring-buffer: Introduce precise nested timestamp sysfs attribute ring-buffer: More precise timestamps for nested writes
Documentation/trace/ftrace.txt | 8 ++++ include/linux/ring_buffer.h | 9 +++++ kernel/trace/ring_buffer.c | 85 ++++++++++++++++++++++++++++++++++++++++-- kernel/trace/trace.c | 51 +++++++++++++++++++++++++ 4 files changed, 149 insertions(+), 4 deletions(-)
-- 1.8.3.4
| |