Skip to contentSkip to conversation

tracing: Updates for 3.12

PATCH In review Steven Rostedt 1 participant 10 messages Nov 6, 2013

Subsystems (5)
x86 Tracing ABI/API ftrace +1 more subsystems
lore
Share & tools
lore

Steven Rostedt

author maintainer
#0 lore raw

cover tracing: Updates for 3.12

This includes a merge with Paul McKenney's tree to include the rcu_is_watching() change that is required to fix a ftrace/perf bug.

git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace.git
for-next

Head SHA1: d562aff93bfb530b0992141500a402d17081189d

Cody P Schafer (1):
trace/trace_stat: use rbtree postorder iteration helper instead of opencoding

Jamie Iles (1):
recordmcount.pl: Add support for __fentry__

Kevin Hao (1):
ftrace/x86: skip over the breakpoint for ftrace caller

Paul E. McKenney (6):
rcu: Remove redundant code from rcu_cleanup_after_idle()
rcu: Throttle rcu_try_advance_all_cbs() execution
rcu: Throttle invoke_rcu_core() invocations due to non-lazy callbacks
rcu: Is it safe to enter an RCU read-side critical section?
rcu: Change EXPORT_SYMBOL() to EXPORT_SYMBOL_GPL()
rcu: Consistent rcu_is_watching() naming

Steven Rostedt (1):
rcu: Do not trace rcu_is_watching() functions

Steven Rostedt (Red Hat) (2):
Merge branch 'idle.2013.09.25a' of git://git.kernel.org/.../paulmck/linux-rcu into HEAD
ftrace: Have control op function callback only trace when RCU is watching

Tom Zanussi (3):
tracing: Update event filters for multibuffer
tracing: Make register/unregister_ftrace_command __init
tracing: Add support for SOFT_DISABLE to syscall events

----

 arch/x86/kernel/ftrace.c             |  14 ++-
 include/linux/ftrace.h               |   4 +-
 include/linux/ftrace_event.h         |  25 +++-
 include/linux/rcupdate.h             |  24 ++--
 include/linux/rcutiny.h              |  17 +++
 include/linux/rcutree.h              |   2 +
 include/linux/syscalls.h             |   4 +-
 include/trace/ftrace.h               |   7 +-
 kernel/lockdep.c                     |   4 +-
 kernel/rcupdate.c                    |   2 +-
 kernel/rcutiny.c                     |  10 +-
 kernel/rcutree.c                     |  23 +++-
 kernel/rcutree.h                     |   2 +
 kernel/rcutree_plugin.h              |  24 ++--
 kernel/trace/ftrace.c                |  21 +++-
 kernel/trace/trace.c                 |  44 +++++--
 kernel/trace/trace.h                 |  22 +---
 kernel/trace/trace_branch.c          |   2 +-
 kernel/trace/trace_events.c          |  23 ++--
 kernel/trace/trace_events_filter.c   | 218 ++++++++++++++++++++++++++++-------
 kernel/trace/trace_export.c          |   2 +-
 kernel/trace/trace_functions_graph.c |   4 +-
 kernel/trace/trace_kprobe.c          |   4 +-
 kernel/trace/trace_mmiotrace.c       |   4 +-
 kernel/trace/trace_sched_switch.c    |   4 +-
 kernel/trace/trace_stat.c            |  41 +------
 kernel/trace/trace_syscalls.c        |  42 +++++--
 kernel/trace/trace_uprobe.c          |   3 +-
 scripts/recordmcount.pl              |   4 +-
 29 files changed, 409 insertions(+), 191 deletions(-)
author maintainer
#1 lore raw

PATCH 1/8 trace/trace_stat: use rbtree postorder iteration helper instead of opencoding

From: Cody P Schafer <[email protected]>

Use rbtree_postorder_for_each_entry_safe() to destroy the rbtree instead of opencoding an alternate postorder iteration that modifies the tree

---
 kernel/trace/trace_stat.c | 41 +++++------------------------------------
 1 file changed, 5 insertions(+), 36 deletions(-)
kernel/trace/trace_stat.c +5 -36
index 847f88a..7af6736 100644
@@ -43,46 +43,15 @@ static DEFINE_MUTEX(all_stat_sessions_mutex);
 /* The root directory for all stat files */
 static struct dentry		*stat_dir;
 
-/*
- * Iterate through the rbtree using a post order traversal path
- * to release the next node.
- * It won't necessary release one at each iteration
- * but it will at least advance closer to the next one
- * to be released.
- */
-static struct rb_node *release_next(struct tracer_stat *ts,
-				    struct rb_node *node)
+static void __reset_stat_session(struct stat_session *session)
 {
-	struct stat_node *snode;
-	struct rb_node *parent = rb_parent(node);
-
-	if (node->rb_left)
-		return node->rb_left;
-	else if (node->rb_right)
-		return node->rb_right;
-	else {
-		if (!parent)
-			;
-		else if (parent->rb_left == node)
-			parent->rb_left = NULL;
-		else
-			parent->rb_right = NULL;
+	struct stat_node *snode, *n;
 
-		snode = container_of(node, struct stat_node, node);
-		if (ts->stat_release)
-			ts->stat_release(snode->stat);
+	rbtree_postorder_for_each_entry_safe(snode, n, &session->stat_root, node) {
+		if (session->ts->stat_release)
+			session->ts->stat_release(snode->stat);
 		kfree(snode);
-
-		return parent;
 	}
-}
-
-static void __reset_stat_session(struct stat_session *session)
-{
-	struct rb_node *node = session->stat_root.rb_node;
-
-	while (node)
-		node = release_next(session->ts, node);
 
 	session->stat_root = RB_ROOT;
 }
Signature · 1 line
1.8.4.rc3
author maintainer
#2 lore raw

PATCH 2/8 ftrace/x86: skip over the breakpoint for ftrace caller

In reply to Steven Rostedt

From: Kevin Hao <[email protected]>

In commit 8a4d0a687a59 "ftrace: Use breakpoint method to update ftrace caller", we choose to use breakpoint method to update the ftrace caller. But we also need to skip over the breakpoint in function ftrace_int3_handler() for them. Otherwise weird things would happen.

---
 arch/x86/kernel/ftrace.c | 14 +++++++++++++-
 1 file changed, 13 insertions(+), 1 deletion(-)
arch/x86/kernel/ftrace.c +13 -1
index 42a392a..d4bdd25 100644
@@ -248,6 +248,15 @@ int ftrace_update_ftrace_func(ftrace_func_t func)
 	return ret;
 }
 
+static int is_ftrace_caller(unsigned long ip)
+{
+	if (ip == (unsigned long)(&ftrace_call) ||
+		ip == (unsigned long)(&ftrace_regs_call))
+		return 1;
+
+	return 0;
+}
+
 /*
  * A breakpoint was added to the code address we are about to
  * modify, and this is the handle that will just skip over it.
@@ -257,10 +266,13 @@ int ftrace_update_ftrace_func(ftrace_func_t func)
  */
 int ftrace_int3_handler(struct pt_regs *regs)
 {
+	unsigned long ip;
+
 	if (WARN_ON_ONCE(!regs))
 		return 0;
 
-	if (!ftrace_location(regs->ip - 1))
+	ip = regs->ip - 1;
+	if (!ftrace_location(ip) && !is_ftrace_caller(ip))
 		return 0;
 
 	regs->ip += MCOUNT_INSN_SIZE - 1;
Signature · 1 line
1.8.4.rc3
author maintainer
#3 lore raw

PATCH 3/8 rcu: Do not trace rcu_is_watching() functions

In reply to Steven Rostedt

From: Steven Rostedt <[email protected]>

As perf uses the rcu_read_lock() primitives for recording into its ring buffer, perf tracing can not be called when RCU in inactive. With the perf function tracing, there are functions that can be traced when RCU is not active, and perf must not have its function callback called when this is the case.

Luckily, Paul McKenney has created a way to detect when RCU is active or not with the rcu_is_watching() function. Unfortunately, this function can also be traced, and if that happens it can cause a bit of overhead for the perf function calls that do the check. Recursion protection prevents anything bad from happening, but there is a bit of added overhead for every function being traced that must detect that the rcu_is_watching() is also being traced.

As rcu_is_watching() is a helper routine and not part of the critical logic in RCU, it does not need to be traced in order to debug RCU itself. Add the "notrace" annotation to all the rcu_is_watching() calls such that we never trace it.

---
 kernel/rcutiny.c | 2 +-
 kernel/rcutree.c | 4 ++--
 2 files changed, 3 insertions(+), 3 deletions(-)
kernel/rcutiny.c +1 -1
index 0fa061d..590c8a8 100644
@@ -179,7 +179,7 @@ EXPORT_SYMBOL_GPL(rcu_irq_enter);
 /*
  * Test whether RCU thinks that the current CPU is idle.
  */
-bool __rcu_is_watching(void)
+bool notrace __rcu_is_watching(void)
 {
 	return rcu_dynticks_nesting;
 }
kernel/rcutree.c +2 -2
index 981d0c1..499bb2b 100644
@@ -662,7 +662,7 @@ void rcu_nmi_exit(void)
  * rcu_is_watching(), the caller of __rcu_is_watching() must have at
  * least disabled preemption.
  */
-bool __rcu_is_watching(void)
+bool notrace __rcu_is_watching(void)
 {
 	return atomic_read(this_cpu_ptr(&rcu_dynticks.dynticks)) & 0x1;
 }
@@ -673,7 +673,7 @@ bool __rcu_is_watching(void)
  * If the current CPU is in its idle loop and is neither in an interrupt
  * or NMI handler, return true.
  */
-bool rcu_is_watching(void)
+bool notrace rcu_is_watching(void)
 {
 	int ret;
 
Signature · 1 line
1.8.4.rc3
author maintainer
#4 lore raw

PATCH 4/8 ftrace: Have control op function callback only trace when RCU is watching

In reply to Steven Rostedt

From: "Steven Rostedt (Red Hat)" <[email protected]>

Dave Jones reported that trinity would be able to trigger the following back trace:

===============================
[ INFO: suspicious RCU usage. ]
3.10.0-rc2+ #38 Not tainted
-------------------------------
include/linux/rcupdate.h:771 rcu_read_lock() used illegally while idle! other info that might help us debug this:

RCU used illegally from idle CPU! rcu_scheduler_active = 1, debug_locks = 0 RCU used illegally from extended quiescent state!
1 lock held by trinity-child1/18786:
#0: (rcu_read_lock){.+.+..}, at: [<ffffffff8113dd48>] __perf_event_overflow+0x108/0x310

 stack backtrace:
 CPU: 3 PID: 18786 Comm: trinity-child1 Not tainted 3.10.0-rc2+ #38
  0000000000000000 ffff88020767bac8 ffffffff816e2f6b ffff88020767baf8
  ffffffff810b5897 ffff88021de92520 0000000000000000 ffff88020767bbf8
  0000000000000000 ffff88020767bb78 ffffffff8113ded4 ffffffff8113dd48
 Call Trace:
  [<ffffffff816e2f6b>] dump_stack+0x19/0x1b
  [<ffffffff810b5897>] lockdep_rcu_suspicious+0xe7/0x120
  [<ffffffff8113ded4>] __perf_event_overflow+0x294/0x310
  [<ffffffff8113dd48>] ? __perf_event_overflow+0x108/0x310
  [<ffffffff81309289>] ? __const_udelay+0x29/0x30
  [<ffffffff81076054>] ? __rcu_read_unlock+0x54/0xa0
  [<ffffffff816f4000>] ? ftrace_call+0x5/0x2f
  [<ffffffff8113dfa1>] perf_swevent_overflow+0x51/0xe0
  [<ffffffff8113e08f>] perf_swevent_event+0x5f/0x90
  [<ffffffff8113e1c9>] perf_tp_event+0x109/0x4f0
  [<ffffffff8113e36f>] ? perf_tp_event+0x2af/0x4f0
  [<ffffffff81074630>] ? __rcu_read_lock+0x20/0x20
  [<ffffffff8112d79f>] perf_ftrace_function_call+0xbf/0xd0
  [<ffffffff8110e1e1>] ? ftrace_ops_control_func+0x181/0x210
  [<ffffffff81074630>] ? __rcu_read_lock+0x20/0x20
  [<ffffffff81100cae>] ? rcu_eqs_enter_common+0x5e/0x470
  [<ffffffff8110e1e1>] ftrace_ops_control_func+0x181/0x210
  [<ffffffff816f4000>] ftrace_call+0x5/0x2f
  [<ffffffff8110e229>] ? ftrace_ops_control_func+0x1c9/0x210
  [<ffffffff816f4000>] ? ftrace_call+0x5/0x2f
  [<ffffffff81074635>] ? debug_lockdep_rcu_enabled+0x5/0x40
  [<ffffffff81074635>] ? debug_lockdep_rcu_enabled+0x5/0x40
  [<ffffffff81100cae>] ? rcu_eqs_enter_common+0x5e/0x470
  [<ffffffff8110112a>] rcu_eqs_enter+0x6a/0xb0
  [<ffffffff81103673>] rcu_user_enter+0x13/0x20
  [<ffffffff8114541a>] user_enter+0x6a/0xd0
  [<ffffffff8100f6d8>] syscall_trace_leave+0x78/0x140
  [<ffffffff816f46af>] int_check_syscall_exit_work+0x34/0x3d
 ------------[ cut here ]------------

Perf uses rcu_read_lock() but as the function tracer can trace functions even when RCU is not currently active, this makes the rcu_read_lock() used by perf ineffective.

As perf is currently the only user of the ftrace_ops_control_func() and perf is also the only function callback that actively uses rcu_read_lock(), the quick fix is to prevent the ftrace_ops_control_func() from calling its callbacks if RCU is not active.

With Paul's new "rcu_is_watching()" we can tell if RCU is active or not.

Cc Paul E. McKenney <[email protected]>
Cc Peter Zijlstra <[email protected]>
Cc Jiri Olsa <[email protected]>
---
 kernel/trace/ftrace.c | 9 +++++++++
 1 file changed, 9 insertions(+)
kernel/trace/ftrace.c +9 -0
index 44e826a..080b7d4 100644
@@ -4388,12 +4388,21 @@ ftrace_ops_control_func(unsigned long ip, unsigned long parent_ip,
 	 */
 	preempt_disable_notrace();
 	trace_recursion_set(TRACE_CONTROL_BIT);
+
+	/*
+	 * Control funcs (perf) uses RCU. Only trace if
+	 * RCU is currently active.
+	 */
+	if (!rcu_is_watching())
+		goto out;
+
 	do_for_each_ftrace_op(op, ftrace_control_list) {
 		if (!(op->flags & FTRACE_OPS_FL_STUB) &&
 		    !ftrace_function_local_disabled(op) &&
 		    ftrace_ops_test(op, ip, regs))
 			op->func(ip, parent_ip, op, regs);
 	} while_for_each_ftrace_op(op);
+ out:
 	trace_recursion_clear(TRACE_CONTROL_BIT);
 	preempt_enable_notrace();
 }
Signature · 1 line
1.8.4.rc3
author maintainer
#5 lore raw

PATCH 5/8 recordmcount.pl: Add support for __fentry__

In reply to Steven Rostedt

From: Jamie Iles <[email protected]>

With gcc 4.6.0 the -mfentry feature places the function profiling call at the start of the function. When this is used, the call is to __fentry__ and not mcount. This is required for Ksplice as the C version of recordmcount doesn't insert section symbols for the __mcount_loc section so we fall back to the perl version.

Based on 48bb5dc6cd9d30fe0d594947563da1f8bd9abada (ftrace: Make recordmcount.c handle __fentry__).

---
 scripts/recordmcount.pl | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
scripts/recordmcount.pl +2 -2
index a674fd5..d0da663 100755
@@ -214,13 +214,13 @@ $local_regex = "^[0-9a-fA-F]+\\s+t\\s+(\\S+)";
 $weak_regex = "^[0-9a-fA-F]+\\s+([wW])\\s+(\\S+)";
 $section_regex = "Disassembly of section\\s+(\\S+):";
 $function_regex = "^([0-9a-fA-F]+)\\s+<(.*?)>:";
-$mcount_regex = "^\\s*([0-9a-fA-F]+):.*\\smcount\$";
+$mcount_regex = "^\\s*([0-9a-fA-F]+):.*\\s(mcount|__fentry__)\$";
 $section_type = '@progbits';
 $mcount_adjust = 0;
 $type = ".long";
 
 if ($arch eq "x86_64") {
-    $mcount_regex = "^\\s*([0-9a-fA-F]+):.*\\smcount([+-]0x[0-9a-zA-Z]+)?\$";
+    $mcount_regex = "^\\s*([0-9a-fA-F]+):.*\\s(mcount|__fentry__)([+-]0x[0-9a-zA-Z]+)?\$";
     $type = ".quad";
     $alignment = 8;
     $mcount_adjust = -1;
Signature · 1 line
1.8.4.rc3
author maintainer
#6 lore raw

PATCH 6/8 tracing: Update event filters for multibuffer

In reply to Steven Rostedt

From: Tom Zanussi <[email protected]>

The trace event filters are still tied to event calls rather than event files, which means you don't get what you'd expect when using filters in the multibuffer case:

Before:

  # echo 'bytes_alloc > 8192' > /sys/kernel/debug/tracing/events/kmem/kmalloc/filter
  # cat /sys/kernel/debug/tracing/events/kmem/kmalloc/filter
  bytes_alloc > 8192
  # mkdir /sys/kernel/debug/tracing/instances/test1
  # echo 'bytes_alloc > 2048' > /sys/kernel/debug/tracing/instances/test1/events/kmem/kmalloc/filter
  # cat /sys/kernel/debug/tracing/events/kmem/kmalloc/filter
  bytes_alloc > 2048
  # cat /sys/kernel/debug/tracing/instances/test1/events/kmem/kmalloc/filter
  bytes_alloc > 2048

Setting the filter in tracing/instances/test1/events shouldn't affect the same event in tracing/events as it does above.

After:

  # echo 'bytes_alloc > 8192' > /sys/kernel/debug/tracing/events/kmem/kmalloc/filter
  # cat /sys/kernel/debug/tracing/events/kmem/kmalloc/filter
  bytes_alloc > 8192
  # mkdir /sys/kernel/debug/tracing/instances/test1
  # echo 'bytes_alloc > 2048' > /sys/kernel/debug/tracing/instances/test1/events/kmem/kmalloc/filter
  # cat /sys/kernel/debug/tracing/events/kmem/kmalloc/filter
  bytes_alloc > 8192
  # cat /sys/kernel/debug/tracing/instances/test1/events/kmem/kmalloc/filter
  bytes_alloc > 2048

We'd like to just move the filter directly from ftrace_event_call to ftrace_event_file, but there are a couple cases that don't yet have multibuffer support and therefore have to continue using the current event_call-based filters. For those cases, a new USE_CALL_FILTER bit is added to the event_call flags, whose main purpose is to keep the old behavior for those cases until they can be updated with multibuffer support; at that point, the USE_CALL_FILTER flag (and the new associated call_filter_check_discard() function) can go away.

The multibuffer support also made filter_current_check_discard() redundant, so this change removes that function as well and replaces it with filter_check_discard() (or call_filter_check_discard() as appropriate).

---
 include/linux/ftrace_event.h         |  25 +++-
 include/linux/syscalls.h             |   4 +-
 include/trace/ftrace.h               |   7 +-
 kernel/trace/trace.c                 |  40 +++++--
 kernel/trace/trace.h                 |  18 +--
 kernel/trace/trace_branch.c          |   2 +-
 kernel/trace/trace_events.c          |  23 ++--
 kernel/trace/trace_events_filter.c   | 218 ++++++++++++++++++++++++++++-------
 kernel/trace/trace_export.c          |   2 +-
 kernel/trace/trace_functions_graph.c |   4 +-
 kernel/trace/trace_kprobe.c          |   4 +-
 kernel/trace/trace_mmiotrace.c       |   4 +-
 kernel/trace/trace_sched_switch.c    |   4 +-
 kernel/trace/trace_syscalls.c        |   8 +-
 kernel/trace/trace_uprobe.c          |   3 +-
 15 files changed, 264 insertions(+), 102 deletions(-)
include/linux/ftrace_event.h +20 -5
index 5eaa746..9abbe63 100644
@@ -202,6 +202,7 @@ enum {
 	TRACE_EVENT_FL_NO_SET_FILTER_BIT,
 	TRACE_EVENT_FL_IGNORE_ENABLE_BIT,
 	TRACE_EVENT_FL_WAS_ENABLED_BIT,
+	TRACE_EVENT_FL_USE_CALL_FILTER_BIT,
 };
 
 /*
@@ -213,6 +214,7 @@ enum {
  *  WAS_ENABLED   - Set and stays set when an event was ever enabled
  *                    (used for module unloading, if a module event is enabled,
  *                     it is best to clear the buffers that used it).
+ *  USE_CALL_FILTER - For ftrace internal events, don't use file filter
  */
 enum {
 	TRACE_EVENT_FL_FILTERED		= (1 << TRACE_EVENT_FL_FILTERED_BIT),
@@ -220,6 +222,7 @@ enum {
 	TRACE_EVENT_FL_NO_SET_FILTER	= (1 << TRACE_EVENT_FL_NO_SET_FILTER_BIT),
 	TRACE_EVENT_FL_IGNORE_ENABLE	= (1 << TRACE_EVENT_FL_IGNORE_ENABLE_BIT),
 	TRACE_EVENT_FL_WAS_ENABLED	= (1 << TRACE_EVENT_FL_WAS_ENABLED_BIT),
+	TRACE_EVENT_FL_USE_CALL_FILTER	= (1 << TRACE_EVENT_FL_USE_CALL_FILTER_BIT),
 };
 
 struct ftrace_event_call {
@@ -238,6 +241,7 @@ struct ftrace_event_call {
 	 *   bit 2:		failed to apply filter
 	 *   bit 3:		ftrace internal event (do not enable)
 	 *   bit 4:		Event was enabled by module
+	 *   bit 5:		use call filter rather than file filter
 	 */
 	int			flags; /* static flags of different events */
 
@@ -253,6 +257,8 @@ struct ftrace_subsystem_dir;
 enum {
 	FTRACE_EVENT_FL_ENABLED_BIT,
 	FTRACE_EVENT_FL_RECORDED_CMD_BIT,
+	FTRACE_EVENT_FL_FILTERED_BIT,
+	FTRACE_EVENT_FL_NO_SET_FILTER_BIT,
 	FTRACE_EVENT_FL_SOFT_MODE_BIT,
 	FTRACE_EVENT_FL_SOFT_DISABLED_BIT,
 };
@@ -261,6 +267,8 @@ enum {
  * Ftrace event file flags:
  *  ENABLED	  - The event is enabled
  *  RECORDED_CMD  - The comms should be recorded at sched_switch
+ *  FILTERED	  - The event has a filter attached
+ *  NO_SET_FILTER - Set when filter has error and is to be ignored
  *  SOFT_MODE     - The event is enabled/disabled by SOFT_DISABLED
  *  SOFT_DISABLED - When set, do not trace the event (even though its
  *                   tracepoint may be enabled)
@@ -268,6 +276,8 @@ enum {
 enum {
 	FTRACE_EVENT_FL_ENABLED		= (1 << FTRACE_EVENT_FL_ENABLED_BIT),
 	FTRACE_EVENT_FL_RECORDED_CMD	= (1 << FTRACE_EVENT_FL_RECORDED_CMD_BIT),
+	FTRACE_EVENT_FL_FILTERED	= (1 << FTRACE_EVENT_FL_FILTERED_BIT),
+	FTRACE_EVENT_FL_NO_SET_FILTER	= (1 << FTRACE_EVENT_FL_NO_SET_FILTER_BIT),
 	FTRACE_EVENT_FL_SOFT_MODE	= (1 << FTRACE_EVENT_FL_SOFT_MODE_BIT),
 	FTRACE_EVENT_FL_SOFT_DISABLED	= (1 << FTRACE_EVENT_FL_SOFT_DISABLED_BIT),
 };
@@ -275,6 +285,7 @@ enum {
 struct ftrace_event_file {
 	struct list_head		list;
 	struct ftrace_event_call	*event_call;
+	struct event_filter		*filter;
 	struct dentry			*dir;
 	struct trace_array		*tr;
 	struct ftrace_subsystem_dir	*system;
@@ -310,12 +321,16 @@ struct ftrace_event_file {
 
 #define MAX_FILTER_STR_VAL	256	/* Should handle KSYM_SYMBOL_LEN */
 
-extern void destroy_preds(struct ftrace_event_call *call);
+extern void destroy_preds(struct ftrace_event_file *file);
+extern void destroy_call_preds(struct ftrace_event_call *call);
 extern int filter_match_preds(struct event_filter *filter, void *rec);
-extern int filter_current_check_discard(struct ring_buffer *buffer,
-					struct ftrace_event_call *call,
-					void *rec,
-					struct ring_buffer_event *event);
+
+extern int filter_check_discard(struct ftrace_event_file *file, void *rec,
+				struct ring_buffer *buffer,
+				struct ring_buffer_event *event);
+extern int call_filter_check_discard(struct ftrace_event_call *call, void *rec,
+				     struct ring_buffer *buffer,
+				     struct ring_buffer_event *event);
 
 enum {
 	FILTER_OTHER = 0,
include/linux/syscalls.h +2 -2
index 7fac04e..10bafa9 100644
@@ -120,7 +120,7 @@ extern struct trace_event_functions exit_syscall_print_funcs;
 		.class			= &event_class_syscall_enter,	\
 		.event.funcs            = &enter_syscall_print_funcs,	\
 		.data			= (void *)&__syscall_meta_##sname,\
-		.flags			= TRACE_EVENT_FL_CAP_ANY,	\
+		.flags			= TRACE_EVENT_FL_CAP_ANY | TRACE_EVENT_FL_USE_CALL_FILTER,\
 	};								\
 	static struct ftrace_event_call __used				\
 	  __attribute__((section("_ftrace_events")))			\
@@ -134,7 +134,7 @@ extern struct trace_event_functions exit_syscall_print_funcs;
 		.class			= &event_class_syscall_exit,	\
 		.event.funcs		= &exit_syscall_print_funcs,	\
 		.data			= (void *)&__syscall_meta_##sname,\
-		.flags			= TRACE_EVENT_FL_CAP_ANY,	\
+		.flags			= TRACE_EVENT_FL_CAP_ANY | TRACE_EVENT_FL_USE_CALL_FILTER,\
 	};								\
 	static struct ftrace_event_call __used				\
 	  __attribute__((section("_ftrace_events")))			\
include/trace/ftrace.h +3 -4
index 5c7ab17..52594b2 100644
@@ -437,9 +437,8 @@ static inline notrace int ftrace_get_offsets_##call(			\
  *	{ <assign>; }  <-- Here we assign the entries by the __field and
  *			   __array macros.
  *
- *	if (!filter_current_check_discard(buffer, event_call, entry, event))
- *		trace_nowake_buffer_unlock_commit(buffer,
- *						   event, irq_flags, pc);
+ *	if (!filter_check_discard(ftrace_file, entry, buffer, event))
+ *		trace_buffer_unlock_commit(buffer, event, irq_flags, pc);
  * }
  *
  * static struct trace_event ftrace_event_type_<call> = {
@@ -553,7 +552,7 @@ ftrace_raw_event_##call(void *__data, proto)				\
 									\
 	{ assign; }							\
 									\
-	if (!filter_current_check_discard(buffer, event_call, entry, event)) \
+	if (!filter_check_discard(ftrace_file, entry, buffer, event))	\
 		trace_buffer_unlock_commit(buffer, event, irq_flags, pc); \
 }
 /*
kernel/trace/trace.c +30 -10
index 063a92b..489da8b 100644
@@ -235,13 +235,33 @@ void trace_array_put(struct trace_array *this_tr)
 	mutex_unlock(&trace_types_lock);
 }
 
-int filter_current_check_discard(struct ring_buffer *buffer,
-				 struct ftrace_event_call *call, void *rec,
-				 struct ring_buffer_event *event)
+int filter_check_discard(struct ftrace_event_file *file, void *rec,
+			 struct ring_buffer *buffer,
+			 struct ring_buffer_event *event)
 {
-	return filter_check_discard(call, rec, buffer, event);
+	if (unlikely(file->flags & FTRACE_EVENT_FL_FILTERED) &&
+	    !filter_match_preds(file->filter, rec)) {
+		ring_buffer_discard_commit(buffer, event);
+		return 1;
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(filter_check_discard);
+
+int call_filter_check_discard(struct ftrace_event_call *call, void *rec,
+			      struct ring_buffer *buffer,
+			      struct ring_buffer_event *event)
+{
+	if (unlikely(call->flags & TRACE_EVENT_FL_FILTERED) &&
+	    !filter_match_preds(call->filter, rec)) {
+		ring_buffer_discard_commit(buffer, event);
+		return 1;
+	}
+
+	return 0;
 }
-EXPORT_SYMBOL_GPL(filter_current_check_discard);
+EXPORT_SYMBOL_GPL(call_filter_check_discard);
 
 cycle_t buffer_ftrace_now(struct trace_buffer *buf, int cpu)
 {
@@ -1633,7 +1653,7 @@ trace_function(struct trace_array *tr,
 	entry->ip			= ip;
 	entry->parent_ip		= parent_ip;
 
-	if (!filter_check_discard(call, entry, buffer, event))
+	if (!call_filter_check_discard(call, entry, buffer, event))
 		__buffer_unlock_commit(buffer, event);
 }
 
@@ -1717,7 +1737,7 @@ static void __ftrace_trace_stack(struct ring_buffer *buffer,
 
 	entry->size = trace.nr_entries;
 
-	if (!filter_check_discard(call, entry, buffer, event))
+	if (!call_filter_check_discard(call, entry, buffer, event))
 		__buffer_unlock_commit(buffer, event);
 
  out:
@@ -1819,7 +1839,7 @@ ftrace_trace_userstack(struct ring_buffer *buffer, unsigned long flags, int pc)
 	trace.entries		= entry->caller;
 
 	save_stack_trace_user(&trace);
-	if (!filter_check_discard(call, entry, buffer, event))
+	if (!call_filter_check_discard(call, entry, buffer, event))
 		__buffer_unlock_commit(buffer, event);
 
  out_drop_count:
@@ -2011,7 +2031,7 @@ int trace_vbprintk(unsigned long ip, const char *fmt, va_list args)
 	entry->fmt			= fmt;
 
 	memcpy(entry->buf, tbuffer, sizeof(u32) * len);
-	if (!filter_check_discard(call, entry, buffer, event)) {
+	if (!call_filter_check_discard(call, entry, buffer, event)) {
 		__buffer_unlock_commit(buffer, event);
 		ftrace_trace_stack(buffer, flags, 6, pc);
 	}
@@ -2066,7 +2086,7 @@ __trace_array_vprintk(struct ring_buffer *buffer,
 
 	memcpy(&entry->buf, tbuffer, len);
 	entry->buf[len] = '\0';
-	if (!filter_check_discard(call, entry, buffer, event)) {
+	if (!call_filter_check_discard(call, entry, buffer, event)) {
 		__buffer_unlock_commit(buffer, event);
 		ftrace_trace_stack(buffer, flags, 6, pc);
 	}
kernel/trace/trace.h +2 -16
index d1cf515..12d1a61 100644
@@ -1007,9 +1007,9 @@ struct filter_pred {
 
 extern enum regex_type
 filter_parse_regex(char *buff, int len, char **search, int *not);
-extern void print_event_filter(struct ftrace_event_call *call,
+extern void print_event_filter(struct ftrace_event_file *file,
 			       struct trace_seq *s);
-extern int apply_event_filter(struct ftrace_event_call *call,
+extern int apply_event_filter(struct ftrace_event_file *file,
 			      char *filter_string);
 extern int apply_subsystem_event_filter(struct ftrace_subsystem_dir *dir,
 					char *filter_string);
@@ -1020,20 +1020,6 @@ extern int filter_assign_type(const char *type);
 struct ftrace_event_field *
 trace_find_event_field(struct ftrace_event_call *call, char *name);
 
-static inline int
-filter_check_discard(struct ftrace_event_call *call, void *rec,
-		     struct ring_buffer *buffer,
-		     struct ring_buffer_event *event)
-{
-	if (unlikely(call->flags & TRACE_EVENT_FL_FILTERED) &&
-	    !filter_match_preds(call->filter, rec)) {
-		ring_buffer_discard_commit(buffer, event);
-		return 1;
-	}
-
-	return 0;
-}
-
 extern void trace_event_enable_cmd_record(bool enable);
 extern int event_trace_add_tracer(struct dentry *parent, struct trace_array *tr);
 extern int event_trace_del_tracer(struct trace_array *tr);
kernel/trace/trace_branch.c +1 -1
index d594da0..697fb9b 100644
@@ -78,7 +78,7 @@ probe_likely_condition(struct ftrace_branch_data *f, int val, int expect)
 	entry->line = f->line;
 	entry->correct = val == expect;
 
-	if (!filter_check_discard(call, entry, buffer, event))
+	if (!call_filter_check_discard(call, entry, buffer, event))
 		__buffer_unlock_commit(buffer, event);
 
  out:
kernel/trace/trace_events.c +12 -11
index 368a4d5..043f833 100644
@@ -989,7 +989,7 @@ static ssize_t
 event_filter_read(struct file *filp, char __user *ubuf, size_t cnt,
 		  loff_t *ppos)
 {
-	struct ftrace_event_call *call;
+	struct ftrace_event_file *file;
 	struct trace_seq *s;
 	int r = -ENODEV;
 
@@ -1004,12 +1004,12 @@ event_filter_read(struct file *filp, char __user *ubuf, size_t cnt,
 	trace_seq_init(s);
 
 	mutex_lock(&event_mutex);
-	call = event_file_data(filp);
-	if (call)
-		print_event_filter(call, s);
+	file = event_file_data(filp);
+	if (file)
+		print_event_filter(file, s);
 	mutex_unlock(&event_mutex);
 
-	if (call)
+	if (file)
 		r = simple_read_from_buffer(ubuf, cnt, ppos, s->buffer, s->len);
 
 	kfree(s);
@@ -1021,7 +1021,7 @@ static ssize_t
 event_filter_write(struct file *filp, const char __user *ubuf, size_t cnt,
 		   loff_t *ppos)
 {
-	struct ftrace_event_call *call;
+	struct ftrace_event_file *file;
 	char *buf;
 	int err = -ENODEV;
 
@@ -1039,9 +1039,9 @@ event_filter_write(struct file *filp, const char __user *ubuf, size_t cnt,
 	buf[cnt] = '\0';
 
 	mutex_lock(&event_mutex);
-	call = event_file_data(filp);
-	if (call)
-		err = apply_event_filter(call, buf);
+	file = event_file_data(filp);
+	if (file)
+		err = apply_event_filter(file, buf);
 	mutex_unlock(&event_mutex);
 
 	free_page((unsigned long) buf);
@@ -1539,7 +1539,7 @@ event_create_dir(struct dentry *parent, struct ftrace_event_file *file)
 			return -1;
 		}
 	}
-	trace_create_file("filter", 0644, file->dir, call,
+	trace_create_file("filter", 0644, file->dir, file,
 			  &ftrace_event_filter_fops);
 
 	trace_create_file("format", 0444, file->dir, call,
@@ -1577,6 +1577,7 @@ static void event_remove(struct ftrace_event_call *call)
 		if (file->event_call != call)
 			continue;
 		ftrace_event_enable_disable(file, 0);
+		destroy_preds(file);
 		/*
 		 * The do_for_each_event_file() is
 		 * a double loop. After finding the call for this
@@ -1700,7 +1701,7 @@ static void __trace_remove_event_call(struct ftrace_event_call *call)
 {
 	event_remove(call);
 	trace_destroy_fields(call);
-	destroy_preds(call);
+	destroy_call_preds(call);
 }
 
 static int probe_remove_event_call(struct ftrace_event_call *call)
kernel/trace/trace_events_filter.c +179 -39

373 lines not on this page · raw message

kernel/trace/trace_export.c +1 -1
index d21a746..7c3e3e7 100644
@@ -180,7 +180,7 @@ struct ftrace_event_call __used event_##call = {			\
 	.event.type		= etype,				\
 	.class			= &event_class_ftrace_##call,		\
 	.print_fmt		= print,				\
-	.flags			= TRACE_EVENT_FL_IGNORE_ENABLE,		\
+	.flags			= TRACE_EVENT_FL_IGNORE_ENABLE | TRACE_EVENT_FL_USE_CALL_FILTER, \
 };									\
 struct ftrace_event_call __used						\
 __attribute__((section("_ftrace_events"))) *__event_##call = &event_##call;
kernel/trace/trace_functions_graph.c +2 -2
index e08c030..80387d1 100644
@@ -270,7 +270,7 @@ int __trace_graph_entry(struct trace_array *tr,
 		return 0;
 	entry	= ring_buffer_event_data(event);
 	entry->graph_ent			= *trace;
-	if (!filter_current_check_discard(buffer, call, entry, event))
+	if (!call_filter_check_discard(call, entry, buffer, event))
 		__buffer_unlock_commit(buffer, event);
 
 	return 1;
@@ -385,7 +385,7 @@ void __trace_graph_return(struct trace_array *tr,
 		return;
 	entry	= ring_buffer_event_data(event);
 	entry->ret				= *trace;
-	if (!filter_current_check_discard(buffer, call, entry, event))
+	if (!call_filter_check_discard(call, entry, buffer, event))
 		__buffer_unlock_commit(buffer, event);
 }
 
kernel/trace/trace_kprobe.c +2 -2
index 243f683..dae9541 100644
@@ -835,7 +835,7 @@ __kprobe_trace_func(struct trace_probe *tp, struct pt_regs *regs,
 	entry->ip = (unsigned long)tp->rp.kp.addr;
 	store_trace_args(sizeof(*entry), tp, regs, (u8 *)&entry[1], dsize);
 
-	if (!filter_current_check_discard(buffer, call, entry, event))
+	if (!filter_check_discard(ftrace_file, entry, buffer, event))
 		trace_buffer_unlock_commit_regs(buffer, event,
 						irq_flags, pc, regs);
 }
@@ -884,7 +884,7 @@ __kretprobe_trace_func(struct trace_probe *tp, struct kretprobe_instance *ri,
 	entry->ret_ip = (unsigned long)ri->ret_addr;
 	store_trace_args(sizeof(*entry), tp, regs, (u8 *)&entry[1], dsize);
 
-	if (!filter_current_check_discard(buffer, call, entry, event))
+	if (!filter_check_discard(ftrace_file, entry, buffer, event))
 		trace_buffer_unlock_commit_regs(buffer, event,
 						irq_flags, pc, regs);
 }
kernel/trace/trace_mmiotrace.c +2 -2
index b3dcfb2..0abd9b8 100644
@@ -323,7 +323,7 @@ static void __trace_mmiotrace_rw(struct trace_array *tr,
 	entry	= ring_buffer_event_data(event);
 	entry->rw			= *rw;
 
-	if (!filter_check_discard(call, entry, buffer, event))
+	if (!call_filter_check_discard(call, entry, buffer, event))
 		trace_buffer_unlock_commit(buffer, event, 0, pc);
 }
 
@@ -353,7 +353,7 @@ static void __trace_mmiotrace_map(struct trace_array *tr,
 	entry	= ring_buffer_event_data(event);
 	entry->map			= *map;
 
-	if (!filter_check_discard(call, entry, buffer, event))
+	if (!call_filter_check_discard(call, entry, buffer, event))
 		trace_buffer_unlock_commit(buffer, event, 0, pc);
 }
 
kernel/trace/trace_sched_switch.c +2 -2
index 4e98e3b..3f34dc9 100644
@@ -45,7 +45,7 @@ tracing_sched_switch_trace(struct trace_array *tr,
 	entry->next_state		= next->state;
 	entry->next_cpu	= task_cpu(next);
 
-	if (!filter_check_discard(call, entry, buffer, event))
+	if (!call_filter_check_discard(call, entry, buffer, event))
 		trace_buffer_unlock_commit(buffer, event, flags, pc);
 }
 
@@ -101,7 +101,7 @@ tracing_sched_wakeup_trace(struct trace_array *tr,
 	entry->next_state		= wakee->state;
 	entry->next_cpu			= task_cpu(wakee);
 
-	if (!filter_check_discard(call, entry, buffer, event))
+	if (!call_filter_check_discard(call, entry, buffer, event))
 		trace_buffer_unlock_commit(buffer, event, flags, pc);
 }
 
kernel/trace/trace_syscalls.c +4 -4
index 559329d..32644ee 100644
@@ -336,8 +336,8 @@ static void ftrace_syscall_enter(void *data, struct pt_regs *regs, long id)
 	entry->nr = syscall_nr;
 	syscall_get_arguments(current, regs, 0, sys_data->nb_args, entry->args);
 
-	if (!filter_current_check_discard(buffer, sys_data->enter_event,
-					  entry, event))
+	if (!call_filter_check_discard(sys_data->enter_event, entry,
+				       buffer, event))
 		trace_current_buffer_unlock_commit(buffer, event,
 						   irq_flags, pc);
 }
@@ -377,8 +377,8 @@ static void ftrace_syscall_exit(void *data, struct pt_regs *regs, long ret)
 	entry->nr = syscall_nr;
 	entry->ret = syscall_get_return_value(current, regs);
 
-	if (!filter_current_check_discard(buffer, sys_data->exit_event,
-					  entry, event))
+	if (!call_filter_check_discard(sys_data->exit_event, entry,
+				       buffer, event))
 		trace_current_buffer_unlock_commit(buffer, event,
 						   irq_flags, pc);
 }
kernel/trace/trace_uprobe.c +2 -1
index 272261b..b6dcc42 100644
@@ -128,6 +128,7 @@ alloc_trace_uprobe(const char *group, const char *event, int nargs, bool is_ret)
 	if (is_ret)
 		tu->consumer.ret_handler = uretprobe_dispatcher;
 	init_trace_uprobe_filter(&tu->filter);
+	tu->call.flags |= TRACE_EVENT_FL_USE_CALL_FILTER;
 	return tu;
 
 error:
@@ -561,7 +562,7 @@ static void uprobe_trace_print(struct trace_uprobe *tu,
 	for (i = 0; i < tu->nr_args; i++)
 		call_fetch(&tu->args[i].fetch, regs, data + tu->args[i].offset);
 
-	if (!filter_current_check_discard(buffer, call, entry, event))
+	if (!call_filter_check_discard(call, entry, buffer, event))
 		trace_buffer_unlock_commit(buffer, event, 0, 0);
 }
 
Signature · 1 line
1.8.4.rc3
author maintainer
#7 lore raw

PATCH 7/8 tracing: Make register/unregister_ftrace_command __init

In reply to Steven Rostedt

From: Tom Zanussi <[email protected]>

register/unregister_ftrace_command() are only ever called from __init functions, so can themselves be made __init.

Also make register_snapshot_cmd() __init for the same reason.

---
 include/linux/ftrace.h |  4 ++--
 kernel/trace/ftrace.c  | 12 ++++++++++--
 kernel/trace/trace.c   |  4 ++--
 3 files changed, 14 insertions(+), 6 deletions(-)
include/linux/ftrace.h +2 -2
index ec85d48..31ea4b4 100644
@@ -533,11 +533,11 @@ static inline int ftrace_force_update(void) { return 0; }
 static inline void ftrace_disable_daemon(void) { }
 static inline void ftrace_enable_daemon(void) { }
 static inline void ftrace_release_mod(struct module *mod) {}
-static inline int register_ftrace_command(struct ftrace_func_command *cmd)
+static inline __init int register_ftrace_command(struct ftrace_func_command *cmd)
 {
 	return -EINVAL;
 }
-static inline int unregister_ftrace_command(char *cmd_name)
+static inline __init int unregister_ftrace_command(char *cmd_name)
 {
 	return -EINVAL;
 }
kernel/trace/ftrace.c +10 -2
index 080b7d4..22fa556 100644
@@ -3307,7 +3307,11 @@ void unregister_ftrace_function_probe_all(char *glob)
 static LIST_HEAD(ftrace_commands);
 static DEFINE_MUTEX(ftrace_cmd_mutex);
 
-int register_ftrace_command(struct ftrace_func_command *cmd)
+/*
+ * Currently we only register ftrace commands from __init, so mark this
+ * __init too.
+ */
+__init int register_ftrace_command(struct ftrace_func_command *cmd)
 {
 	struct ftrace_func_command *p;
 	int ret = 0;
@@ -3326,7 +3330,11 @@ int register_ftrace_command(struct ftrace_func_command *cmd)
 	return ret;
 }
 
-int unregister_ftrace_command(struct ftrace_func_command *cmd)
+/*
+ * Currently we only unregister ftrace commands from __init, so mark
+ * this __init too.
+ */
+__init int unregister_ftrace_command(struct ftrace_func_command *cmd)
 {
 	struct ftrace_func_command *p, *n;
 	int ret = -ENODEV;
kernel/trace/trace.c +2 -2
index 489da8b..f9fa42b 100644
@@ -5477,12 +5477,12 @@ static struct ftrace_func_command ftrace_snapshot_cmd = {
 	.func			= ftrace_trace_snapshot_callback,
 };
 
-static int register_snapshot_cmd(void)
+static __init int register_snapshot_cmd(void)
 {
 	return register_ftrace_command(&ftrace_snapshot_cmd);
 }
 #else
-static inline int register_snapshot_cmd(void) { return 0; }
+static inline __init int register_snapshot_cmd(void) { return 0; }
 #endif /* defined(CONFIG_TRACER_SNAPSHOT) && defined(CONFIG_DYNAMIC_FTRACE) */
 
 struct dentry *tracing_init_dentry_tr(struct trace_array *tr)
Signature · 1 line
1.8.4.rc3
author maintainer
#8 lore raw

PATCH 8/8 tracing: Add support for SOFT_DISABLE to syscall events

In reply to Steven Rostedt

From: Tom Zanussi <[email protected]>

The original SOFT_DISABLE patches didn't add support for soft disable of syscall events; this adds it.

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.

Adding ftrace_event_files also means we can remove the USE_CALL_FILTER bit, thus enabling multibuffer filter support for syscall events.

---
 include/linux/syscalls.h      |  4 ++--
 kernel/trace/trace.h          |  4 ++--
 kernel/trace/trace_syscalls.c | 42 ++++++++++++++++++++++++++++++++----------
 3 files changed, 36 insertions(+), 14 deletions(-)
include/linux/syscalls.h +2 -2
index 10bafa9..2ef31bf 100644
@@ -120,7 +120,7 @@ extern struct trace_event_functions exit_syscall_print_funcs;
 		.class			= &event_class_syscall_enter,	\
 		.event.funcs            = &enter_syscall_print_funcs,	\
 		.data			= (void *)&__syscall_meta_##sname,\
-		.flags			= TRACE_EVENT_FL_CAP_ANY | TRACE_EVENT_FL_USE_CALL_FILTER,\
+		.flags                  = TRACE_EVENT_FL_CAP_ANY,	\
 	};								\
 	static struct ftrace_event_call __used				\
 	  __attribute__((section("_ftrace_events")))			\
@@ -134,7 +134,7 @@ extern struct trace_event_functions exit_syscall_print_funcs;
 		.class			= &event_class_syscall_exit,	\
 		.event.funcs		= &exit_syscall_print_funcs,	\
 		.data			= (void *)&__syscall_meta_##sname,\
-		.flags			= TRACE_EVENT_FL_CAP_ANY | TRACE_EVENT_FL_USE_CALL_FILTER,\
+		.flags                  = TRACE_EVENT_FL_CAP_ANY,	\
 	};								\
 	static struct ftrace_event_call __used				\
 	  __attribute__((section("_ftrace_events")))			\
kernel/trace/trace.h +2 -2
index 12d1a61..9c27cda 100644
@@ -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;
kernel/trace/trace_syscalls.c +32 -10

108 lines not on this page · raw message

Signature · 1 line
1.8.4.rc3