diff options
author | Jason Baron <jbaron@redhat.com> | 2011-03-16 17:29:47 -0400 |
---|---|---|
committer | Steven Rostedt <rostedt@goodmis.org> | 2011-04-04 12:48:08 -0400 |
commit | d430d3d7e646eb1eac2bb4aa244a644312e67c76 (patch) | |
tree | 0f52534f54d89e41042536ff2f1b2ce74c45c033 /include/linux/perf_event.h | |
parent | ee5e51f51be755830f57445e268ba50e88ccbdbb (diff) |
jump label: Introduce static_branch() interface
Introduce:
static __always_inline bool static_branch(struct jump_label_key *key);
instead of the old JUMP_LABEL(key, label) macro.
In this way, jump labels become really easy to use:
Define:
struct jump_label_key jump_key;
Can be used as:
if (static_branch(&jump_key))
do unlikely code
enable/disale via:
jump_label_inc(&jump_key);
jump_label_dec(&jump_key);
that's it!
For the jump labels disabled case, the static_branch() becomes an
atomic_read(), and jump_label_inc()/dec() are simply atomic_inc(),
atomic_dec() operations. We show testing results for this change below.
Thanks to H. Peter Anvin for suggesting the 'static_branch()' construct.
Since we now require a 'struct jump_label_key *key', we can store a pointer into
the jump table addresses. In this way, we can enable/disable jump labels, in
basically constant time. This change allows us to completely remove the previous
hashtable scheme. Thanks to Peter Zijlstra for this re-write.
Testing:
I ran a series of 'tbench 20' runs 5 times (with reboots) for 3
configurations, where tracepoints were disabled.
jump label configured in
avg: 815.6
jump label *not* configured in (using atomic reads)
avg: 800.1
jump label *not* configured in (regular reads)
avg: 803.4
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <20110316212947.GA8792@redhat.com>
Signed-off-by: Jason Baron <jbaron@redhat.com>
Suggested-by: H. Peter Anvin <hpa@linux.intel.com>
Tested-by: David Daney <ddaney@caviumnetworks.com>
Acked-by: Ralf Baechle <ralf@linux-mips.org>
Acked-by: David S. Miller <davem@davemloft.net>
Acked-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Diffstat (limited to 'include/linux/perf_event.h')
-rw-r--r-- | include/linux/perf_event.h | 26 |
1 files changed, 13 insertions, 13 deletions
diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h index 311b4dc785a1..730b7821690f 100644 --- a/include/linux/perf_event.h +++ b/include/linux/perf_event.h @@ -505,7 +505,7 @@ struct perf_guest_info_callbacks { #include <linux/ftrace.h> #include <linux/cpu.h> #include <linux/irq_work.h> -#include <linux/jump_label_ref.h> +#include <linux/jump_label.h> #include <asm/atomic.h> #include <asm/local.h> @@ -1034,7 +1034,7 @@ static inline int is_software_event(struct perf_event *event) return event->pmu->task_ctx_nr == perf_sw_context; } -extern atomic_t perf_swevent_enabled[PERF_COUNT_SW_MAX]; +extern struct jump_label_key perf_swevent_enabled[PERF_COUNT_SW_MAX]; extern void __perf_sw_event(u32, u64, int, struct pt_regs *, u64); @@ -1063,22 +1063,21 @@ perf_sw_event(u32 event_id, u64 nr, int nmi, struct pt_regs *regs, u64 addr) { struct pt_regs hot_regs; - JUMP_LABEL(&perf_swevent_enabled[event_id], have_event); - return; - -have_event: - if (!regs) { - perf_fetch_caller_regs(&hot_regs); - regs = &hot_regs; + if (static_branch(&perf_swevent_enabled[event_id])) { + if (!regs) { + perf_fetch_caller_regs(&hot_regs); + regs = &hot_regs; + } + __perf_sw_event(event_id, nr, nmi, regs, addr); } - __perf_sw_event(event_id, nr, nmi, regs, addr); } -extern atomic_t perf_sched_events; +extern struct jump_label_key perf_sched_events; static inline void perf_event_task_sched_in(struct task_struct *task) { - COND_STMT(&perf_sched_events, __perf_event_task_sched_in(task)); + if (static_branch(&perf_sched_events)) + __perf_event_task_sched_in(task); } static inline @@ -1086,7 +1085,8 @@ void perf_event_task_sched_out(struct task_struct *task, struct task_struct *nex { perf_sw_event(PERF_COUNT_SW_CONTEXT_SWITCHES, 1, 1, NULL, 0); - COND_STMT(&perf_sched_events, __perf_event_task_sched_out(task, next)); + if (static_branch(&perf_sched_events)) + __perf_event_task_sched_out(task, next); } extern void perf_event_mmap(struct vm_area_struct *vma); |