summaryrefslogtreecommitdiffstats
path: root/kernel/trace/trace_irqsoff.c
diff options
context:
space:
mode:
authorSebastian Andrzej Siewior <bigeasy@linutronix.de>2021-01-25 20:45:08 +0100
committerSteven Rostedt (VMware) <rostedt@goodmis.org>2021-02-02 23:02:06 +0100
commit36590c50b2d0729952511129916beeea30d31d81 (patch)
tree6f6a7dec6834401b5c6648c8f053c06b1c499cd9 /kernel/trace/trace_irqsoff.c
parentring-buffer: Drop unneeded check in ring_buffer_resize() (diff)
downloadlinux-36590c50b2d0729952511129916beeea30d31d81.tar.xz
linux-36590c50b2d0729952511129916beeea30d31d81.zip
tracing: Merge irqflags + preempt counter.
The state of the interrupts (irqflags) and the preemption counter are both passed down to tracing_generic_entry_update(). Only one bit of irqflags is actually required: The on/off state. The complete 32bit of the preemption counter isn't needed. Just whether of the upper bits (softirq, hardirq and NMI) are set and the preemption depth is needed. The irqflags and the preemption counter could be evaluated early and the information stored in an integer `trace_ctx'. tracing_generic_entry_update() would use the upper bits as the TRACE_FLAG_* and the lower 8bit as the disabled-preemption depth (considering that one must be substracted from the counter in one special cases). The actual preemption value is not used except for the tracing record. The `irqflags' variable is mostly used only for the tracing record. An exception here is for instance wakeup_tracer_call() or probe_wakeup_sched_switch() which explicilty disable interrupts and use that `irqflags' to save (and restore) the IRQ state and to record the state. Struct trace_event_buffer has also the `pc' and flags' members which can be replaced with `trace_ctx' since their actual value is not used outside of trace recording. This will reduce tracing_generic_entry_update() to simply assign values to struct trace_entry. The evaluation of the TRACE_FLAG_* bits is moved to _tracing_gen_ctx_flags() which replaces preempt_count() and local_save_flags() invocations. As an example, ftrace_syscall_enter() may invoke: - trace_buffer_lock_reserve() -> … -> tracing_generic_entry_update() - event_trigger_unlock_commit() -> ftrace_trace_stack() -> … -> tracing_generic_entry_update() -> ftrace_trace_userstack() -> … -> tracing_generic_entry_update() In this case the TRACE_FLAG_* bits were evaluated three times. By using the `trace_ctx' they are evaluated once and assigned three times. A build with all tracers enabled on x86-64 with and without the patch: text data bss dec hex filename 21970669 17084168 7639260 46694097 2c87ed1 vmlinux.old 21970293 17084168 7639260 46693721 2c87d59 vmlinux.new text shrank by 379 bytes, data remained constant. Link: https://lkml.kernel.org/r/20210125194511.3924915-2-bigeasy@linutronix.de Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
Diffstat (limited to 'kernel/trace/trace_irqsoff.c')
-rw-r--r--kernel/trace/trace_irqsoff.c86
1 files changed, 36 insertions, 50 deletions
diff --git a/kernel/trace/trace_irqsoff.c b/kernel/trace/trace_irqsoff.c
index 6756379b661f..590b3d51afae 100644
--- a/kernel/trace/trace_irqsoff.c
+++ b/kernel/trace/trace_irqsoff.c
@@ -143,11 +143,14 @@ irqsoff_tracer_call(unsigned long ip, unsigned long parent_ip,
struct trace_array *tr = irqsoff_trace;
struct trace_array_cpu *data;
unsigned long flags;
+ unsigned int trace_ctx;
if (!func_prolog_dec(tr, &data, &flags))
return;
- trace_function(tr, ip, parent_ip, flags, preempt_count());
+ trace_ctx = tracing_gen_ctx_flags(flags);
+
+ trace_function(tr, ip, parent_ip, trace_ctx);
atomic_dec(&data->disabled);
}
@@ -177,8 +180,8 @@ static int irqsoff_graph_entry(struct ftrace_graph_ent *trace)
struct trace_array *tr = irqsoff_trace;
struct trace_array_cpu *data;
unsigned long flags;
+ unsigned int trace_ctx;
int ret;
- int pc;
if (ftrace_graph_ignore_func(trace))
return 0;
@@ -195,8 +198,8 @@ static int irqsoff_graph_entry(struct ftrace_graph_ent *trace)
if (!func_prolog_dec(tr, &data, &flags))
return 0;
- pc = preempt_count();
- ret = __trace_graph_entry(tr, trace, flags, pc);
+ trace_ctx = tracing_gen_ctx_flags(flags);
+ ret = __trace_graph_entry(tr, trace, trace_ctx);
atomic_dec(&data->disabled);
return ret;
@@ -207,15 +210,15 @@ static void irqsoff_graph_return(struct ftrace_graph_ret *trace)
struct trace_array *tr = irqsoff_trace;
struct trace_array_cpu *data;
unsigned long flags;
- int pc;
+ unsigned int trace_ctx;
ftrace_graph_addr_finish(trace);
if (!func_prolog_dec(tr, &data, &flags))
return;
- pc = preempt_count();
- __trace_graph_return(tr, trace, flags, pc);
+ trace_ctx = tracing_gen_ctx_flags(flags);
+ __trace_graph_return(tr, trace, trace_ctx);
atomic_dec(&data->disabled);
}
@@ -267,12 +270,12 @@ static void irqsoff_print_header(struct seq_file *s)
static void
__trace_function(struct trace_array *tr,
unsigned long ip, unsigned long parent_ip,
- unsigned long flags, int pc)
+ unsigned int trace_ctx)
{
if (is_graph(tr))
- trace_graph_function(tr, ip, parent_ip, flags, pc);
+ trace_graph_function(tr, ip, parent_ip, trace_ctx);
else
- trace_function(tr, ip, parent_ip, flags, pc);
+ trace_function(tr, ip, parent_ip, trace_ctx);
}
#else
@@ -322,15 +325,13 @@ check_critical_timing(struct trace_array *tr,
{
u64 T0, T1, delta;
unsigned long flags;
- int pc;
+ unsigned int trace_ctx;
T0 = data->preempt_timestamp;
T1 = ftrace_now(cpu);
delta = T1-T0;
- local_save_flags(flags);
-
- pc = preempt_count();
+ trace_ctx = tracing_gen_ctx();
if (!report_latency(tr, delta))
goto out;
@@ -341,9 +342,9 @@ check_critical_timing(struct trace_array *tr,
if (!report_latency(tr, delta))
goto out_unlock;
- __trace_function(tr, CALLER_ADDR0, parent_ip, flags, pc);
+ __trace_function(tr, CALLER_ADDR0, parent_ip, trace_ctx);
/* Skip 5 functions to get to the irq/preempt enable function */
- __trace_stack(tr, flags, 5, pc);
+ __trace_stack(tr, trace_ctx, 5);
if (data->critical_sequence != max_sequence)
goto out_unlock;
@@ -363,16 +364,15 @@ out_unlock:
out:
data->critical_sequence = max_sequence;
data->preempt_timestamp = ftrace_now(cpu);
- __trace_function(tr, CALLER_ADDR0, parent_ip, flags, pc);
+ __trace_function(tr, CALLER_ADDR0, parent_ip, trace_ctx);
}
static nokprobe_inline void
-start_critical_timing(unsigned long ip, unsigned long parent_ip, int pc)
+start_critical_timing(unsigned long ip, unsigned long parent_ip)
{
int cpu;
struct trace_array *tr = irqsoff_trace;
struct trace_array_cpu *data;
- unsigned long flags;
if (!tracer_enabled || !tracing_is_enabled())
return;
@@ -393,9 +393,7 @@ start_critical_timing(unsigned long ip, unsigned long parent_ip, int pc)
data->preempt_timestamp = ftrace_now(cpu);
data->critical_start = parent_ip ? : ip;
- local_save_flags(flags);
-
- __trace_function(tr, ip, parent_ip, flags, pc);
+ __trace_function(tr, ip, parent_ip, tracing_gen_ctx());
per_cpu(tracing_cpu, cpu) = 1;
@@ -403,12 +401,12 @@ start_critical_timing(unsigned long ip, unsigned long parent_ip, int pc)
}
static nokprobe_inline void
-stop_critical_timing(unsigned long ip, unsigned long parent_ip, int pc)
+stop_critical_timing(unsigned long ip, unsigned long parent_ip)
{
int cpu;
struct trace_array *tr = irqsoff_trace;
struct trace_array_cpu *data;
- unsigned long flags;
+ unsigned int trace_ctx;
cpu = raw_smp_processor_id();
/* Always clear the tracing cpu on stopping the trace */
@@ -428,8 +426,8 @@ stop_critical_timing(unsigned long ip, unsigned long parent_ip, int pc)
atomic_inc(&data->disabled);
- local_save_flags(flags);
- __trace_function(tr, ip, parent_ip, flags, pc);
+ trace_ctx = tracing_gen_ctx();
+ __trace_function(tr, ip, parent_ip, trace_ctx);
check_critical_timing(tr, data, parent_ip ? : ip, cpu);
data->critical_start = 0;
atomic_dec(&data->disabled);
@@ -438,20 +436,16 @@ stop_critical_timing(unsigned long ip, unsigned long parent_ip, int pc)
/* start and stop critical timings used to for stoppage (in idle) */
void start_critical_timings(void)
{
- int pc = preempt_count();
-
- if (preempt_trace(pc) || irq_trace())
- start_critical_timing(CALLER_ADDR0, CALLER_ADDR1, pc);
+ if (preempt_trace(preempt_count()) || irq_trace())
+ start_critical_timing(CALLER_ADDR0, CALLER_ADDR1);
}
EXPORT_SYMBOL_GPL(start_critical_timings);
NOKPROBE_SYMBOL(start_critical_timings);
void stop_critical_timings(void)
{
- int pc = preempt_count();
-
- if (preempt_trace(pc) || irq_trace())
- stop_critical_timing(CALLER_ADDR0, CALLER_ADDR1, pc);
+ if (preempt_trace(preempt_count()) || irq_trace())
+ stop_critical_timing(CALLER_ADDR0, CALLER_ADDR1);
}
EXPORT_SYMBOL_GPL(stop_critical_timings);
NOKPROBE_SYMBOL(stop_critical_timings);
@@ -613,19 +607,15 @@ static void irqsoff_tracer_stop(struct trace_array *tr)
*/
void tracer_hardirqs_on(unsigned long a0, unsigned long a1)
{
- unsigned int pc = preempt_count();
-
- if (!preempt_trace(pc) && irq_trace())
- stop_critical_timing(a0, a1, pc);
+ if (!preempt_trace(preempt_count()) && irq_trace())
+ stop_critical_timing(a0, a1);
}
NOKPROBE_SYMBOL(tracer_hardirqs_on);
void tracer_hardirqs_off(unsigned long a0, unsigned long a1)
{
- unsigned int pc = preempt_count();
-
- if (!preempt_trace(pc) && irq_trace())
- start_critical_timing(a0, a1, pc);
+ if (!preempt_trace(preempt_count()) && irq_trace())
+ start_critical_timing(a0, a1);
}
NOKPROBE_SYMBOL(tracer_hardirqs_off);
@@ -665,18 +655,14 @@ static struct tracer irqsoff_tracer __read_mostly =
#ifdef CONFIG_PREEMPT_TRACER
void tracer_preempt_on(unsigned long a0, unsigned long a1)
{
- int pc = preempt_count();
-
- if (preempt_trace(pc) && !irq_trace())
- stop_critical_timing(a0, a1, pc);
+ if (preempt_trace(preempt_count()) && !irq_trace())
+ stop_critical_timing(a0, a1);
}
void tracer_preempt_off(unsigned long a0, unsigned long a1)
{
- int pc = preempt_count();
-
- if (preempt_trace(pc) && !irq_trace())
- start_critical_timing(a0, a1, pc);
+ if (preempt_trace(preempt_count()) && !irq_trace())
+ start_critical_timing(a0, a1);
}
static int preemptoff_tracer_init(struct trace_array *tr)