2
0
mirror of https://github.com/edk2-porting/linux-next.git synced 2024-12-23 04:34:11 +08:00
linux-next/kernel/trace
Steven Rostedt b2a866f934 ftrace: function tracer with irqs disabled
Impact: disable interrupts during trace entry creation (as opposed to preempt)

To help with performance, I set the ftracer to not disable interrupts,
and only to disable preemption. If an interrupt occurred, it would not
be traced, because the function tracer protects itself from recursion.
This may be faster, but the trace output might miss some traces.

This patch makes the fuction trace disable interrupts, but it also
adds a runtime feature to disable preemption instead. It does this by
having two different tracer functions. When the function tracer is
enabled, it will check to see which version is requested (irqs disabled
or preemption disabled). Then it will use the corresponding function
as the tracer.

Irq disabling is the default behaviour, but if the user wants better
performance, with the chance of missing traces, then they can choose
the preempt disabled version.

Running hackbench 3 times with the irqs disabled and 3 times with
the preempt disabled function tracer yielded:

tracing type       times            entries recorded
------------      --------          ----------------
irq disabled      43.393            166433066
                  43.282            166172618
                  43.298            166256704

preempt disabled  38.969            159871710
                  38.943            159972935
                  39.325            161056510

Average:

   irqs disabled:  43.324           166287462
preempt disabled:  39.079           160300385

 preempt is 10.8 percent faster than irqs disabled.

I wrote a patch to count function trace recursion and reran hackbench.

With irq disabled: 1,150 times the function tracer did not trace due to
  recursion.
with preempt disabled: 5,117,718 times.

The thousand times with irq disabled could be due to NMIs, or simply a case
where it called a function that was not protected by notrace.

But we also see that a large amount of the trace is lost with the
preempt version.

Signed-off-by: Steven Rostedt <srostedt@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Steven Rostedt <srostedt@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-11-04 10:09:50 +01:00
..
ftrace.c ftrace: perform an initialization for ftrace to enable it 2008-10-28 19:15:58 +01:00
Kconfig tracing, alpha: undefined reference to `save_stack_trace' 2008-11-03 10:12:13 +01:00
Makefile ftrace: rename FTRACE to FUNCTION_TRACER 2008-10-20 18:27:03 +02:00
ring_buffer.c ftrace: insert in the ftrace_preempt_disable()/enable() functions 2008-11-04 10:09:49 +01:00
trace_boot.c tracing/fastboot: fix printk format typo in boot tracer 2008-10-14 10:39:23 +02:00
trace_functions.c ftrace: rename the ftrace tracer to function 2008-10-20 18:27:04 +02:00
trace_irqsoff.c ftrace: rename FTRACE to FUNCTION_TRACER 2008-10-20 18:27:03 +02:00
trace_mmiotrace.c ftrace: preempt disable over interrupt disable 2008-10-14 10:39:09 +02:00
trace_nop.c ftrace: make work with new ring buffer 2008-10-14 10:38:57 +02:00
trace_sched_switch.c ftrace: make some tracers reentrant 2008-10-14 10:39:20 +02:00
trace_sched_wakeup.c ftrace: insert in the ftrace_preempt_disable()/enable() functions 2008-11-04 10:09:49 +01:00
trace_selftest_dynamic.c ftrace: fix dynamic ftrace selftest 2008-05-23 21:13:23 +02:00
trace_selftest.c ftrace: remove daemon 2008-10-23 16:00:22 +02:00
trace_stack.c ftrace: insert in the ftrace_preempt_disable()/enable() functions 2008-11-04 10:09:49 +01:00
trace_sysprof.c ftrace: make work with new ring buffer 2008-10-14 10:38:57 +02:00
trace.c ftrace: function tracer with irqs disabled 2008-11-04 10:09:50 +01:00
trace.h ftrace: function tracer with irqs disabled 2008-11-04 10:09:50 +01:00