Re: [PATCH v8 01/10] tracing: Add support for SOFT_DISABLE to syscall events
On Fri, 06 Sep 2013 21:22:15 +0900 Masami Hiramatsu wrote: > > And AFAICS, the tracepoint (on which the syscall tracer based) > call-site uses rcu_read_lock_sched_notrace() instead of rcu_read_lock(), > in that case, I think we should use synchronize_sched(). is that wrong? > Ah, no, you are correct. I was thinking that the tracepoints used rcu_read_lock(), but looking at the code, it's rcu_read_lock_sched(). Thus a synchronize_sched() *is* required. Thanks, and please ignore my other comments about using "synchronize_rcu()". -- Steve -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: [PATCH v8 01/10] tracing: Add support for SOFT_DISABLE to syscall events
(2013/09/06 1:01), Steven Rostedt wrote: > On Mon, 2 Sep 2013 22:52:17 -0500 > >> @@ -415,10 +429,15 @@ static void unreg_event_syscall_enter(struct >> ftrace_event_file *file, >> return; >> mutex_lock(_trace_lock); >> tr->sys_refcount_enter--; >> -clear_bit(num, tr->enabled_enter_syscalls); >> +rcu_assign_pointer(tr->enter_syscall_files[num], NULL); >> if (!tr->sys_refcount_enter) >> unregister_trace_sys_enter(ftrace_syscall_enter, tr); >> mutex_unlock(_trace_lock); >> +/* >> + * Callers expect the event to be completely disabled on >> + * return, so wait for current handlers to finish. >> + */ >> +synchronize_sched(); > > We only have to wait for rcu, not preemption correct? Then we need to > do synchronize_rcu() instead. Hmm, the reason why trace_kprobe.c uses synchronize_sched() in unreg function, is to avoid touching freeing event_file in running kprobe handlers which run under preemption disabled. And AFAICS, the tracepoint (on which the syscall tracer based) call-site uses rcu_read_lock_sched_notrace() instead of rcu_read_lock(), in that case, I think we should use synchronize_sched(). is that wrong? Thank you, -- Masami HIRAMATSU IT Management Research Dept. Linux Technology Center Hitachi, Ltd., Yokohama Research Laboratory E-mail: masami.hiramatsu...@hitachi.com -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: [PATCH v8 01/10] tracing: Add support for SOFT_DISABLE to syscall events
(2013/09/06 1:01), Steven Rostedt wrote: On Mon, 2 Sep 2013 22:52:17 -0500 @@ -415,10 +429,15 @@ static void unreg_event_syscall_enter(struct ftrace_event_file *file, return; mutex_lock(syscall_trace_lock); tr-sys_refcount_enter--; -clear_bit(num, tr-enabled_enter_syscalls); +rcu_assign_pointer(tr-enter_syscall_files[num], NULL); if (!tr-sys_refcount_enter) unregister_trace_sys_enter(ftrace_syscall_enter, tr); mutex_unlock(syscall_trace_lock); +/* + * Callers expect the event to be completely disabled on + * return, so wait for current handlers to finish. + */ +synchronize_sched(); We only have to wait for rcu, not preemption correct? Then we need to do synchronize_rcu() instead. Hmm, the reason why trace_kprobe.c uses synchronize_sched() in unreg function, is to avoid touching freeing event_file in running kprobe handlers which run under preemption disabled. And AFAICS, the tracepoint (on which the syscall tracer based) call-site uses rcu_read_lock_sched_notrace() instead of rcu_read_lock(), in that case, I think we should use synchronize_sched(). is that wrong? Thank you, -- Masami HIRAMATSU IT Management Research Dept. Linux Technology Center Hitachi, Ltd., Yokohama Research Laboratory E-mail: masami.hiramatsu...@hitachi.com -- To unsubscribe from this list: send the line unsubscribe linux-kernel in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: [PATCH v8 01/10] tracing: Add support for SOFT_DISABLE to syscall events
On Fri, 06 Sep 2013 21:22:15 +0900 Masami Hiramatsu masami.hiramatsu...@hitachi.com wrote: And AFAICS, the tracepoint (on which the syscall tracer based) call-site uses rcu_read_lock_sched_notrace() instead of rcu_read_lock(), in that case, I think we should use synchronize_sched(). is that wrong? Ah, no, you are correct. I was thinking that the tracepoints used rcu_read_lock(), but looking at the code, it's rcu_read_lock_sched(). Thus a synchronize_sched() *is* required. Thanks, and please ignore my other comments about using synchronize_rcu(). -- Steve -- To unsubscribe from this list: send the line unsubscribe linux-kernel in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
[PATCH v8 01/10] tracing: Add support for SOFT_DISABLE to syscall events
The original SOFT_DISABLE patches didn't add support for soft disable
of syscall events; this adds it and paves the way for future patches
allowing triggers to be added to syscall events, since triggers are
built on top of SOFT_DISABLE.
Add an array of ftrace_event_file pointers indexed by syscall number
to the trace array and remove the existing enabled bitmaps, which as a
result are now redundant. The ftrace_event_file structs in turn
contain the soft disable flags we need for per-syscall soft disable
accounting; later patches add additional 'trigger' flags and
per-syscall triggers and filters.
Signed-off-by: Tom Zanussi
---
kernel/trace/trace.h | 4 ++--
kernel/trace/trace_syscalls.c | 36 ++--
2 files changed, 32 insertions(+), 8 deletions(-)
diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h
index fe39acd..b1227b9 100644
--- a/kernel/trace/trace.h
+++ b/kernel/trace/trace.h
@@ -192,8 +192,8 @@ struct trace_array {
#ifdef CONFIG_FTRACE_SYSCALLS
int sys_refcount_enter;
int sys_refcount_exit;
- DECLARE_BITMAP(enabled_enter_syscalls, NR_syscalls);
- DECLARE_BITMAP(enabled_exit_syscalls, NR_syscalls);
+ struct ftrace_event_file *enter_syscall_files[NR_syscalls];
+ struct ftrace_event_file *exit_syscall_files[NR_syscalls];
#endif
int stop_count;
int clock_id;
diff --git a/kernel/trace/trace_syscalls.c b/kernel/trace/trace_syscalls.c
index 559329d..af4b71c 100644
--- a/kernel/trace/trace_syscalls.c
+++ b/kernel/trace/trace_syscalls.c
@@ -302,6 +302,7 @@ static int __init syscall_exit_define_fields(struct
ftrace_event_call *call)
static void ftrace_syscall_enter(void *data, struct pt_regs *regs, long id)
{
struct trace_array *tr = data;
+ struct ftrace_event_file *ftrace_file;
struct syscall_trace_enter *entry;
struct syscall_metadata *sys_data;
struct ring_buffer_event *event;
@@ -314,7 +315,13 @@ static void ftrace_syscall_enter(void *data, struct
pt_regs *regs, long id)
syscall_nr = trace_get_syscall_nr(current, regs);
if (syscall_nr < 0)
return;
- if (!test_bit(syscall_nr, tr->enabled_enter_syscalls))
+
+ /* Here we're inside the tp handler's rcu_read_lock (__DO_TRACE()) */
+ ftrace_file = rcu_dereference(tr->enter_syscall_files[syscall_nr]);
+ if (!ftrace_file)
+ return;
+
+ if (test_bit(FTRACE_EVENT_FL_SOFT_DISABLED_BIT, _file->flags))
return;
sys_data = syscall_nr_to_meta(syscall_nr);
@@ -345,6 +352,7 @@ static void ftrace_syscall_enter(void *data, struct pt_regs
*regs, long id)
static void ftrace_syscall_exit(void *data, struct pt_regs *regs, long ret)
{
struct trace_array *tr = data;
+ struct ftrace_event_file *ftrace_file;
struct syscall_trace_exit *entry;
struct syscall_metadata *sys_data;
struct ring_buffer_event *event;
@@ -356,7 +364,13 @@ static void ftrace_syscall_exit(void *data, struct pt_regs
*regs, long ret)
syscall_nr = trace_get_syscall_nr(current, regs);
if (syscall_nr < 0)
return;
- if (!test_bit(syscall_nr, tr->enabled_exit_syscalls))
+
+ /* Here we're inside the tp handler's rcu_read_lock (__DO_TRACE()) */
+ ftrace_file = rcu_dereference(tr->exit_syscall_files[syscall_nr]);
+ if (!ftrace_file)
+ return;
+
+ if (test_bit(FTRACE_EVENT_FL_SOFT_DISABLED_BIT, _file->flags))
return;
sys_data = syscall_nr_to_meta(syscall_nr);
@@ -397,7 +411,7 @@ static int reg_event_syscall_enter(struct ftrace_event_file
*file,
if (!tr->sys_refcount_enter)
ret = register_trace_sys_enter(ftrace_syscall_enter, tr);
if (!ret) {
- set_bit(num, tr->enabled_enter_syscalls);
+ rcu_assign_pointer(tr->enter_syscall_files[num], file);
tr->sys_refcount_enter++;
}
mutex_unlock(_trace_lock);
@@ -415,10 +429,15 @@ static void unreg_event_syscall_enter(struct
ftrace_event_file *file,
return;
mutex_lock(_trace_lock);
tr->sys_refcount_enter--;
- clear_bit(num, tr->enabled_enter_syscalls);
+ rcu_assign_pointer(tr->enter_syscall_files[num], NULL);
if (!tr->sys_refcount_enter)
unregister_trace_sys_enter(ftrace_syscall_enter, tr);
mutex_unlock(_trace_lock);
+ /*
+* Callers expect the event to be completely disabled on
+* return, so wait for current handlers to finish.
+*/
+ synchronize_sched();
}
static int reg_event_syscall_exit(struct ftrace_event_file *file,
@@ -435,7 +454,7 @@ static int reg_event_syscall_exit(struct ftrace_event_file
*file,
if (!tr->sys_refcount_exit)
ret = register_trace_sys_exit(ftrace_syscall_exit,
[PATCH v8 01/10] tracing: Add support for SOFT_DISABLE to syscall events
The original SOFT_DISABLE patches didn't add support for soft disable
of syscall events; this adds it and paves the way for future patches
allowing triggers to be added to syscall events, since triggers are
built on top of SOFT_DISABLE.
Add an array of ftrace_event_file pointers indexed by syscall number
to the trace array and remove the existing enabled bitmaps, which as a
result are now redundant. The ftrace_event_file structs in turn
contain the soft disable flags we need for per-syscall soft disable
accounting; later patches add additional 'trigger' flags and
per-syscall triggers and filters.
Signed-off-by: Tom Zanussi tom.zanu...@linux.intel.com
---
kernel/trace/trace.h | 4 ++--
kernel/trace/trace_syscalls.c | 36 ++--
2 files changed, 32 insertions(+), 8 deletions(-)
diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h
index fe39acd..b1227b9 100644
--- a/kernel/trace/trace.h
+++ b/kernel/trace/trace.h
@@ -192,8 +192,8 @@ struct trace_array {
#ifdef CONFIG_FTRACE_SYSCALLS
int sys_refcount_enter;
int sys_refcount_exit;
- DECLARE_BITMAP(enabled_enter_syscalls, NR_syscalls);
- DECLARE_BITMAP(enabled_exit_syscalls, NR_syscalls);
+ struct ftrace_event_file *enter_syscall_files[NR_syscalls];
+ struct ftrace_event_file *exit_syscall_files[NR_syscalls];
#endif
int stop_count;
int clock_id;
diff --git a/kernel/trace/trace_syscalls.c b/kernel/trace/trace_syscalls.c
index 559329d..af4b71c 100644
--- a/kernel/trace/trace_syscalls.c
+++ b/kernel/trace/trace_syscalls.c
@@ -302,6 +302,7 @@ static int __init syscall_exit_define_fields(struct
ftrace_event_call *call)
static void ftrace_syscall_enter(void *data, struct pt_regs *regs, long id)
{
struct trace_array *tr = data;
+ struct ftrace_event_file *ftrace_file;
struct syscall_trace_enter *entry;
struct syscall_metadata *sys_data;
struct ring_buffer_event *event;
@@ -314,7 +315,13 @@ static void ftrace_syscall_enter(void *data, struct
pt_regs *regs, long id)
syscall_nr = trace_get_syscall_nr(current, regs);
if (syscall_nr 0)
return;
- if (!test_bit(syscall_nr, tr-enabled_enter_syscalls))
+
+ /* Here we're inside the tp handler's rcu_read_lock (__DO_TRACE()) */
+ ftrace_file = rcu_dereference(tr-enter_syscall_files[syscall_nr]);
+ if (!ftrace_file)
+ return;
+
+ if (test_bit(FTRACE_EVENT_FL_SOFT_DISABLED_BIT, ftrace_file-flags))
return;
sys_data = syscall_nr_to_meta(syscall_nr);
@@ -345,6 +352,7 @@ static void ftrace_syscall_enter(void *data, struct pt_regs
*regs, long id)
static void ftrace_syscall_exit(void *data, struct pt_regs *regs, long ret)
{
struct trace_array *tr = data;
+ struct ftrace_event_file *ftrace_file;
struct syscall_trace_exit *entry;
struct syscall_metadata *sys_data;
struct ring_buffer_event *event;
@@ -356,7 +364,13 @@ static void ftrace_syscall_exit(void *data, struct pt_regs
*regs, long ret)
syscall_nr = trace_get_syscall_nr(current, regs);
if (syscall_nr 0)
return;
- if (!test_bit(syscall_nr, tr-enabled_exit_syscalls))
+
+ /* Here we're inside the tp handler's rcu_read_lock (__DO_TRACE()) */
+ ftrace_file = rcu_dereference(tr-exit_syscall_files[syscall_nr]);
+ if (!ftrace_file)
+ return;
+
+ if (test_bit(FTRACE_EVENT_FL_SOFT_DISABLED_BIT, ftrace_file-flags))
return;
sys_data = syscall_nr_to_meta(syscall_nr);
@@ -397,7 +411,7 @@ static int reg_event_syscall_enter(struct ftrace_event_file
*file,
if (!tr-sys_refcount_enter)
ret = register_trace_sys_enter(ftrace_syscall_enter, tr);
if (!ret) {
- set_bit(num, tr-enabled_enter_syscalls);
+ rcu_assign_pointer(tr-enter_syscall_files[num], file);
tr-sys_refcount_enter++;
}
mutex_unlock(syscall_trace_lock);
@@ -415,10 +429,15 @@ static void unreg_event_syscall_enter(struct
ftrace_event_file *file,
return;
mutex_lock(syscall_trace_lock);
tr-sys_refcount_enter--;
- clear_bit(num, tr-enabled_enter_syscalls);
+ rcu_assign_pointer(tr-enter_syscall_files[num], NULL);
if (!tr-sys_refcount_enter)
unregister_trace_sys_enter(ftrace_syscall_enter, tr);
mutex_unlock(syscall_trace_lock);
+ /*
+* Callers expect the event to be completely disabled on
+* return, so wait for current handlers to finish.
+*/
+ synchronize_sched();
}
static int reg_event_syscall_exit(struct ftrace_event_file *file,
@@ -435,7 +454,7 @@ static int reg_event_syscall_exit(struct ftrace_event_file
*file,
if (!tr-sys_refcount_exit)
ret =
