diff options
author | Linus Torvalds <torvalds@linux-foundation.org> | 2016-03-24 20:52:25 +0300 |
---|---|---|
committer | Linus Torvalds <torvalds@linux-foundation.org> | 2016-03-24 20:52:25 +0300 |
commit | e46b4e2b46e173889b19999b8bd033d5e8b3acf0 (patch) | |
tree | 43f4770868d372c76ce5f084cc8e23d2942e07d5 /kernel/trace/trace_irqsoff.c | |
parent | faea72dd0f155a9d349e191d74830716c702651d (diff) | |
parent | 7e6867bf831c71fe0e47438831ae3a94d4c7ab3c (diff) | |
download | linux-e46b4e2b46e173889b19999b8bd033d5e8b3acf0.tar.xz |
Merge tag 'trace-v4.6' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace
Pull tracing updates from Steven Rostedt:
"Nothing major this round. Mostly small clean ups and fixes.
Some visible changes:
- A new flag was added to distinguish traces done in NMI context.
- Preempt tracer now shows functions where preemption is disabled but
interrupts are still enabled.
Other notes:
- Updates were done to function tracing to allow better performance
with perf.
- Infrastructure code has been added to allow for a new histogram
feature for recording live trace event histograms that can be
configured by simple user commands. The feature itself was just
finished, but needs a round in linux-next before being pulled.
This only includes some infrastructure changes that will be needed"
* tag 'trace-v4.6' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace: (22 commits)
tracing: Record and show NMI state
tracing: Fix trace_printk() to print when not using bprintk()
tracing: Remove redundant reset per-CPU buff in irqsoff tracer
x86: ftrace: Fix the misleading comment for arch/x86/kernel/ftrace.c
tracing: Fix crash from reading trace_pipe with sendfile
tracing: Have preempt(irqs)off trace preempt disabled functions
tracing: Fix return while holding a lock in register_tracer()
ftrace: Use kasprintf() in ftrace_profile_tracefs()
ftrace: Update dynamic ftrace calls only if necessary
ftrace: Make ftrace_hash_rec_enable return update bool
tracing: Fix typoes in code comment and printk in trace_nop.c
tracing, writeback: Replace cgroup path to cgroup ino
tracing: Use flags instead of bool in trigger structure
tracing: Add an unreg_all() callback to trigger commands
tracing: Add needs_rec flag to event triggers
tracing: Add a per-event-trigger 'paused' field
tracing: Add get_syscall_name()
tracing: Add event record param to trigger_ops.func()
tracing: Make event trigger functions available
tracing: Make ftrace_event_field checking functions available
...
Diffstat (limited to 'kernel/trace/trace_irqsoff.c')
-rw-r--r-- | kernel/trace/trace_irqsoff.c | 9 |
1 files changed, 6 insertions, 3 deletions
diff --git a/kernel/trace/trace_irqsoff.c b/kernel/trace/trace_irqsoff.c index e4e56589ec1d..03cdff84d026 100644 --- a/kernel/trace/trace_irqsoff.c +++ b/kernel/trace/trace_irqsoff.c @@ -109,8 +109,12 @@ static int func_prolog_dec(struct trace_array *tr, return 0; local_save_flags(*flags); - /* slight chance to get a false positive on tracing_cpu */ - if (!irqs_disabled_flags(*flags)) + /* + * Slight chance to get a false positive on tracing_cpu, + * although I'm starting to think there isn't a chance. + * Leave this for now just to be paranoid. + */ + if (!irqs_disabled_flags(*flags) && !preempt_count()) return 0; *data = per_cpu_ptr(tr->trace_buffer.data, cpu); @@ -622,7 +626,6 @@ static int __irqsoff_tracer_init(struct trace_array *tr) irqsoff_trace = tr; /* make sure that the tracer is visible */ smp_wmb(); - tracing_reset_online_cpus(&tr->trace_buffer); ftrace_init_array_ops(tr, irqsoff_tracer_call); |