Skip to contentSkip to conversation

tracing fixes

GIT PULL Ingo Molnar 1 participant 1 message Jun 20, 2009

Subsystems (6)
x86 PowerPC Tracing s390 +2 more subsystems
lore
Share & tools
lore
#0 lore raw

Linus,

Please pull the latest tracing-fixes-for-linus git tree from:

git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip.git tracing-fixes-for-linus

Thanks,

Ingo

------------------>
Frederic Weisbecker (2):
tracing/urgent: fix unbalanced ftrace_start_up tracing/urgent: warn in case of ftrace_start_up inbalance

Li Zefan (6):
tracing: fix a typo in tracing_cpumask_write()
tracing: replace a GFP_ATOMIC with GFP_KERNEL allocation tracing/filters: operand can be negative tracing/filters: strloc should be unsigned short tracing/filters: free filter_string in destroy_preds() tracing/filters: fix race between filter setting and module unload

Rusty Russell (1):
cpumask: use new operators in kernel/trace

Steven Rostedt (14):
ring-buffer: prevent adding write in discarded area
ring-buffer: have benchmark test handle discarded events
ring-buffer: remove unused variable
ring-buffer: use commit counters for commit pointer accounting
tracing: update sample event documentation
ring-buffer: use BUF_PAGE_HDR_SIZE in calculating index
ring-buffer: remove useless warn on check
ring-buffer: remove useless compile check for buffer_page size
ring-buffer: check for less than two in size allocation
ring-buffer: add locks around rb_per_cpu_empty
ring-buffer: do not grab locks in nmi
ring-buffer: have benchmark test print to trace buffer
function-graph: disable when both x86_32 and optimize for size are configured
function-graph: add stack frame test

Wu Zhangjin (1):
tracing: fix undeclared 'PAGE_SIZE' in include/linux/trace_seq.h

 arch/powerpc/kernel/ftrace.c               |    2 +-
 arch/s390/kernel/ftrace.c                  |    2 +-
 arch/x86/Kconfig                           |    1 +
 arch/x86/kernel/entry_32.S                 |    2 +
 arch/x86/kernel/entry_64.S                 |    2 +
 arch/x86/kernel/ftrace.c                   |    6 +-
 include/linux/ftrace.h                     |    4 +-
 include/linux/trace_seq.h                  |    2 +
 kernel/trace/Kconfig                       |    8 +
 kernel/trace/ftrace.c                      |    7 +
 kernel/trace/kmemtrace.c                   |    2 +-
 kernel/trace/ring_buffer.c                 |  309 +++++++++++++++++-----------
 kernel/trace/ring_buffer_benchmark.c       |   45 +++--
 kernel/trace/trace.c                       |    8 +-
 kernel/trace/trace_events_filter.c         |   37 ++--
 kernel/trace/trace_functions.c             |    8 +-
 kernel/trace/trace_functions_graph.c       |   36 +++-
 samples/trace_events/Makefile              |    8 +
 samples/trace_events/trace-events-sample.h |   27 ++-
 19 files changed, 331 insertions(+), 185 deletions(-)
arch/powerpc/kernel/ftrace.c +1 -1
index 2d182f1..58d6a61 100644
@@ -605,7 +605,7 @@ void prepare_ftrace_return(unsigned long *parent, unsigned long self_addr)
 		return;
 	}
 
-	if (ftrace_push_return_trace(old, self_addr, &trace.depth) == -EBUSY) {
+	if (ftrace_push_return_trace(old, self_addr, &trace.depth, 0) == -EBUSY) {
 		*parent = old;
 		return;
 	}
arch/s390/kernel/ftrace.c +1 -1
index 82ddfd3..3e298e6 100644
@@ -190,7 +190,7 @@ unsigned long prepare_ftrace_return(unsigned long ip, unsigned long parent)
 		goto out;
 	if (unlikely(atomic_read(&current->tracing_graph_pause)))
 		goto out;
-	if (ftrace_push_return_trace(parent, ip, &trace.depth) == -EBUSY)
+	if (ftrace_push_return_trace(parent, ip, &trace.depth, 0) == -EBUSY)
 		goto out;
 	trace.func = ftrace_mcount_call_adjust(ip) & PSW_ADDR_INSN;
 	/* Only trace if the calling function expects to. */
arch/x86/Kconfig +1 -0
index 356d2ec..fcf12af 100644
@@ -33,6 +33,7 @@ config X86
 	select HAVE_DYNAMIC_FTRACE
 	select HAVE_FUNCTION_TRACER
 	select HAVE_FUNCTION_GRAPH_TRACER
+	select HAVE_FUNCTION_GRAPH_FP_TEST
 	select HAVE_FUNCTION_TRACE_MCOUNT_TEST
 	select HAVE_FTRACE_NMI_ENTER if DYNAMIC_FTRACE
 	select HAVE_FTRACE_SYSCALLS
arch/x86/kernel/entry_32.S +2 -0
index c929add..0d4b285 100644
@@ -1154,6 +1154,7 @@ ENTRY(ftrace_graph_caller)
 	pushl %edx
 	movl 0xc(%esp), %edx
 	lea 0x4(%ebp), %eax
+	movl (%ebp), %ecx
 	subl $MCOUNT_INSN_SIZE, %edx
 	call prepare_ftrace_return
 	popl %edx
@@ -1168,6 +1169,7 @@ return_to_handler:
 	pushl %eax
 	pushl %ecx
 	pushl %edx
+	movl %ebp, %eax
 	call ftrace_return_to_handler
 	movl %eax, 0xc(%esp)
 	popl %edx
arch/x86/kernel/entry_64.S +2 -0
index de74f0a..c251be7 100644
@@ -135,6 +135,7 @@ ENTRY(ftrace_graph_caller)
 
 	leaq 8(%rbp), %rdi
 	movq 0x38(%rsp), %rsi
+	movq (%rbp), %rdx
 	subq $MCOUNT_INSN_SIZE, %rsi
 
 	call	prepare_ftrace_return
@@ -150,6 +151,7 @@ GLOBAL(return_to_handler)
 	/* Save the return values */
 	movq %rax, (%rsp)
 	movq %rdx, 8(%rsp)
+	movq %rbp, %rdi
 
 	call ftrace_return_to_handler
 
arch/x86/kernel/ftrace.c +4 -2
index b79c553..d94e1ea 100644
@@ -408,7 +408,8 @@ int ftrace_disable_ftrace_graph_caller(void)
  * Hook the return address and push it in the stack of return addrs
  * in current thread info.
  */
-void prepare_ftrace_return(unsigned long *parent, unsigned long self_addr)
+void prepare_ftrace_return(unsigned long *parent, unsigned long self_addr,
+			   unsigned long frame_pointer)
 {
 	unsigned long old;
 	int faulted;
@@ -453,7 +454,8 @@ void prepare_ftrace_return(unsigned long *parent, unsigned long self_addr)
 		return;
 	}
 
-	if (ftrace_push_return_trace(old, self_addr, &trace.depth) == -EBUSY) {
+	if (ftrace_push_return_trace(old, self_addr, &trace.depth,
+		    frame_pointer) == -EBUSY) {
 		*parent = old;
 		return;
 	}
include/linux/ftrace.h +3 -1
index 39b95c5..dc3b132 100644
@@ -362,6 +362,7 @@ struct ftrace_ret_stack {
 	unsigned long func;
 	unsigned long long calltime;
 	unsigned long long subtime;
+	unsigned long fp;
 };
 
 /*
@@ -372,7 +373,8 @@ struct ftrace_ret_stack {
 extern void return_to_handler(void);
 
 extern int
-ftrace_push_return_trace(unsigned long ret, unsigned long func, int *depth);
+ftrace_push_return_trace(unsigned long ret, unsigned long func, int *depth,
+			 unsigned long frame_pointer);
 
 /*
  * Sometimes we don't want to trace a function with the function
include/linux/trace_seq.h +2 -0
index c68bccb..c134dd1 100644
@@ -3,6 +3,8 @@
 
 #include <linux/fs.h>
 
+#include <asm/page.h>
+
 /*
  * Trace sequences are used to allow a function to call several other functions
  * to create a string of data to use (up to a max of PAGE_SIZE.
kernel/trace/Kconfig +8 -0
index 4a13e5a..b17ed87 100644
@@ -18,6 +18,13 @@ config HAVE_FUNCTION_TRACER
 config HAVE_FUNCTION_GRAPH_TRACER
 	bool
 
+config HAVE_FUNCTION_GRAPH_FP_TEST
+	bool
+	help
+	 An arch may pass in a unique value (frame pointer) to both the
+	 entering and exiting of a function. On exit, the value is compared
+	 and if it does not match, then it will panic the kernel.
+
 config HAVE_FUNCTION_TRACE_MCOUNT_TEST
 	bool
 	help
@@ -121,6 +128,7 @@ config FUNCTION_GRAPH_TRACER
 	bool "Kernel Function Graph Tracer"
 	depends on HAVE_FUNCTION_GRAPH_TRACER
 	depends on FUNCTION_TRACER
+	depends on !X86_32 || !CC_OPTIMIZE_FOR_SIZE
 	default y
 	help
 	  Enable the kernel to trace a function at both its return
kernel/trace/ftrace.c +7 -0
index bb60732..3718d55 100644
@@ -1224,6 +1224,13 @@ static void ftrace_shutdown(int command)
 		return;
 
 	ftrace_start_up--;
+	/*
+	 * Just warn in case of unbalance, no need to kill ftrace, it's not
+	 * critical but the ftrace_call callers may be never nopped again after
+	 * further ftrace uses.
+	 */
+	WARN_ON_ONCE(ftrace_start_up < 0);
+
 	if (!ftrace_start_up)
 		command |= FTRACE_DISABLE_CALLS;
 
kernel/trace/kmemtrace.c +1 -1
index 86cdf67..1edaa95 100644
@@ -186,7 +186,7 @@ static int kmem_trace_init(struct trace_array *tr)
 	int cpu;
 	kmemtrace_array = tr;
 
-	for_each_cpu_mask(cpu, cpu_possible_map)
+	for_each_cpu(cpu, cpu_possible_mask)
 		tracing_reset(tr, cpu);
 
 	kmemtrace_start_probes();
kernel/trace/ring_buffer.c +192 -117

541 lines not on this page · raw message

kernel/trace/ring_buffer_benchmark.c +24 -21

109 lines not on this page · raw message

kernel/trace/trace.c +4 -4
index 8acd9b8..0f57f1b 100644
@@ -2191,11 +2191,12 @@ tracing_cpumask_write(struct file *filp, const char __user *ubuf,
 	if (!alloc_cpumask_var(&tracing_cpumask_new, GFP_KERNEL))
 		return -ENOMEM;
 
-	mutex_lock(&tracing_cpumask_update_lock);
 	err = cpumask_parse_user(ubuf, count, tracing_cpumask_new);
 	if (err)
 		goto err_unlock;
 
+	mutex_lock(&tracing_cpumask_update_lock);
+
 	local_irq_disable();
 	__raw_spin_lock(&ftrace_max_lock);
 	for_each_tracing_cpu(cpu) {
@@ -2223,8 +2224,7 @@ tracing_cpumask_write(struct file *filp, const char __user *ubuf,
 	return count;
 
 err_unlock:
-	mutex_unlock(&tracing_cpumask_update_lock);
-	free_cpumask_var(tracing_cpumask);
+	free_cpumask_var(tracing_cpumask_new);
 
 	return err;
 }
@@ -3627,7 +3627,7 @@ tracing_stats_read(struct file *filp, char __user *ubuf,
 	struct trace_seq *s;
 	unsigned long cnt;
 
-	s = kmalloc(sizeof(*s), GFP_ATOMIC);
+	s = kmalloc(sizeof(*s), GFP_KERNEL);
 	if (!s)
 		return ENOMEM;
 
kernel/trace/trace_events_filter.c +18 -19
index db6e54b..936c621 100644
@@ -27,8 +27,6 @@
 #include "trace.h"
 #include "trace_output.h"
 
-static DEFINE_MUTEX(filter_mutex);
-
 enum filter_op_ids
 {
 	OP_OR,
@@ -178,7 +176,7 @@ static int filter_pred_string(struct filter_pred *pred, void *event,
 static int filter_pred_strloc(struct filter_pred *pred, void *event,
 			      int val1, int val2)
 {
-	int str_loc = *(int *)(event + pred->offset);
+	unsigned short str_loc = *(unsigned short *)(event + pred->offset);
 	char *addr = (char *)(event + str_loc);
 	int cmp, match;
 
@@ -294,12 +292,12 @@ void print_event_filter(struct ftrace_event_call *call, struct trace_seq *s)
 {
 	struct event_filter *filter = call->filter;
 
-	mutex_lock(&filter_mutex);
+	mutex_lock(&event_mutex);
 	if (filter->filter_string)
 		trace_seq_printf(s, "%s\n", filter->filter_string);
 	else
 		trace_seq_printf(s, "none\n");
-	mutex_unlock(&filter_mutex);
+	mutex_unlock(&event_mutex);
 }
 
 void print_subsystem_event_filter(struct event_subsystem *system,
@@ -307,12 +305,12 @@ void print_subsystem_event_filter(struct event_subsystem *system,
 {
 	struct event_filter *filter = system->filter;
 
-	mutex_lock(&filter_mutex);
+	mutex_lock(&event_mutex);
 	if (filter->filter_string)
 		trace_seq_printf(s, "%s\n", filter->filter_string);
 	else
 		trace_seq_printf(s, "none\n");
-	mutex_unlock(&filter_mutex);
+	mutex_unlock(&event_mutex);
 }
 
 static struct ftrace_event_field *
@@ -381,6 +379,7 @@ void destroy_preds(struct ftrace_event_call *call)
 			filter_free_pred(filter->preds[i]);
 	}
 	kfree(filter->preds);
+	kfree(filter->filter_string);
 	kfree(filter);
 	call->filter = NULL;
 }
@@ -433,7 +432,6 @@ static void filter_free_subsystem_preds(struct event_subsystem *system)
 		filter->n_preds = 0;
 	}
 
-	mutex_lock(&event_mutex);
 	list_for_each_entry(call, &ftrace_events, list) {
 		if (!call->define_fields)
 			continue;
@@ -443,7 +441,6 @@ static void filter_free_subsystem_preds(struct event_subsystem *system)
 			remove_filter_string(call->filter);
 		}
 	}
-	mutex_unlock(&event_mutex);
 }
 
 static int filter_add_pred_fn(struct filter_parse_state *ps,
@@ -546,6 +543,7 @@ static int filter_add_pred(struct filter_parse_state *ps,
 	filter_pred_fn_t fn;
 	unsigned long long val;
 	int string_type;
+	int ret;
 
 	pred->fn = filter_pred_none;
 
@@ -581,7 +579,11 @@ static int filter_add_pred(struct filter_parse_state *ps,
 			pred->not = 1;
 		return filter_add_pred_fn(ps, call, pred, fn);
 	} else {
-		if (strict_strtoull(pred->str_val, 0, &val)) {
+		if (field->is_signed)
+			ret = strict_strtoll(pred->str_val, 0, &val);
+		else
+			ret = strict_strtoull(pred->str_val, 0, &val);
+		if (ret) {
 			parse_error(ps, FILT_ERR_ILLEGAL_INTVAL, 0);
 			return -EINVAL;
 		}
@@ -625,7 +627,6 @@ static int filter_add_subsystem_pred(struct filter_parse_state *ps,
 	filter->preds[filter->n_preds] = pred;
 	filter->n_preds++;
 
-	mutex_lock(&event_mutex);
 	list_for_each_entry(call, &ftrace_events, list) {
 
 		if (!call->define_fields)
@@ -636,14 +637,12 @@ static int filter_add_subsystem_pred(struct filter_parse_state *ps,
 
 		err = filter_add_pred(ps, call, pred);
 		if (err) {
-			mutex_unlock(&event_mutex);
 			filter_free_subsystem_preds(system);
 			parse_error(ps, FILT_ERR_BAD_SUBSYS_FILTER, 0);
 			goto out;
 		}
 		replace_filter_string(call->filter, filter_string);
 	}
-	mutex_unlock(&event_mutex);
 out:
 	return err;
 }
@@ -1070,12 +1069,12 @@ int apply_event_filter(struct ftrace_event_call *call, char *filter_string)
 
 	struct filter_parse_state *ps;
 
-	mutex_lock(&filter_mutex);
+	mutex_lock(&event_mutex);
 
 	if (!strcmp(strstrip(filter_string), "0")) {
 		filter_disable_preds(call);
 		remove_filter_string(call->filter);
-		mutex_unlock(&filter_mutex);
+		mutex_unlock(&event_mutex);
 		return 0;
 	}
 
@@ -1103,7 +1102,7 @@ out:
 	postfix_clear(ps);
 	kfree(ps);
 out_unlock:
-	mutex_unlock(&filter_mutex);
+	mutex_unlock(&event_mutex);
 
 	return err;
 }
@@ -1115,12 +1114,12 @@ int apply_subsystem_event_filter(struct event_subsystem *system,
 
 	struct filter_parse_state *ps;
 
-	mutex_lock(&filter_mutex);
+	mutex_lock(&event_mutex);
 
 	if (!strcmp(strstrip(filter_string), "0")) {
 		filter_free_subsystem_preds(system);
 		remove_filter_string(system->filter);
-		mutex_unlock(&filter_mutex);
+		mutex_unlock(&event_mutex);
 		return 0;
 	}
 
@@ -1148,7 +1147,7 @@ out:
 	postfix_clear(ps);
 	kfree(ps);
 out_unlock:
-	mutex_unlock(&filter_mutex);
+	mutex_unlock(&event_mutex);
 
 	return err;
 }
kernel/trace/trace_functions.c +5 -3
index c9a0b7d..90f1347 100644
@@ -193,9 +193,11 @@ static void tracing_start_function_trace(void)
 static void tracing_stop_function_trace(void)
 {
 	ftrace_function_enabled = 0;
-	/* OK if they are not registered */
-	unregister_ftrace_function(&trace_stack_ops);
-	unregister_ftrace_function(&trace_ops);
+
+	if (func_flags.val & TRACE_FUNC_OPT_STACK)
+		unregister_ftrace_function(&trace_stack_ops);
+	else
+		unregister_ftrace_function(&trace_ops);
 }
 
 static int func_set_flag(u32 old_flags, u32 bit, int set)
kernel/trace/trace_functions_graph.c +32 -4
index 8b59241..d2249ab 100644
@@ -57,7 +57,8 @@ static struct tracer_flags tracer_flags = {
 
 /* Add a function return address to the trace stack on thread info.*/
 int
-ftrace_push_return_trace(unsigned long ret, unsigned long func, int *depth)
+ftrace_push_return_trace(unsigned long ret, unsigned long func, int *depth,
+			 unsigned long frame_pointer)
 {
 	unsigned long long calltime;
 	int index;
@@ -85,6 +86,7 @@ ftrace_push_return_trace(unsigned long ret, unsigned long func, int *depth)
 	current->ret_stack[index].func = func;
 	current->ret_stack[index].calltime = calltime;
 	current->ret_stack[index].subtime = 0;
+	current->ret_stack[index].fp = frame_pointer;
 	*depth = index;
 
 	return 0;
@@ -92,7 +94,8 @@ ftrace_push_return_trace(unsigned long ret, unsigned long func, int *depth)
 
 /* Retrieve a function return address to the trace stack on thread info.*/
 static void
-ftrace_pop_return_trace(struct ftrace_graph_ret *trace, unsigned long *ret)
+ftrace_pop_return_trace(struct ftrace_graph_ret *trace, unsigned long *ret,
+			unsigned long frame_pointer)
 {
 	int index;
 
@@ -106,6 +109,31 @@ ftrace_pop_return_trace(struct ftrace_graph_ret *trace, unsigned long *ret)
 		return;
 	}
 
+#ifdef CONFIG_HAVE_FUNCTION_GRAPH_FP_TEST
+	/*
+	 * The arch may choose to record the frame pointer used
+	 * and check it here to make sure that it is what we expect it
+	 * to be. If gcc does not set the place holder of the return
+	 * address in the frame pointer, and does a copy instead, then
+	 * the function graph trace will fail. This test detects this
+	 * case.
+	 *
+	 * Currently, x86_32 with optimize for size (-Os) makes the latest
+	 * gcc do the above.
+	 */
+	if (unlikely(current->ret_stack[index].fp != frame_pointer)) {
+		ftrace_graph_stop();
+		WARN(1, "Bad frame pointer: expected %lx, received %lx\n"
+		     "  from func %pF return to %lx\n",
+		     current->ret_stack[index].fp,
+		     frame_pointer,
+		     (void *)current->ret_stack[index].func,
+		     current->ret_stack[index].ret);
+		*ret = (unsigned long)panic;
+		return;
+	}
+#endif
+
 	*ret = current->ret_stack[index].ret;
 	trace->func = current->ret_stack[index].func;
 	trace->calltime = current->ret_stack[index].calltime;
@@ -117,12 +145,12 @@ ftrace_pop_return_trace(struct ftrace_graph_ret *trace, unsigned long *ret)
  * Send the trace to the ring-buffer.
  * @return the original return address.
  */
-unsigned long ftrace_return_to_handler(void)
+unsigned long ftrace_return_to_handler(unsigned long frame_pointer)
 {
 	struct ftrace_graph_ret trace;
 	unsigned long ret;
 
-	ftrace_pop_return_trace(&trace, &ret);
+	ftrace_pop_return_trace(&trace, &ret, frame_pointer);
 	trace.rettime = trace_clock_local();
 	ftrace_graph_return(&trace);
 	barrier();
samples/trace_events/Makefile +8 -0
index 0d428dc..0f8d921 100644
@@ -1,6 +1,14 @@
 # builds the trace events example kernel modules;
 # then to use one (as root):  insmod <module_name.ko>
 
+# If you include a trace header outside of include/trace/events
+# then the file that does the #define CREATE_TRACE_POINTS must
+# have that tracer file in its main search path. This is because
+# define_trace.h will include it, and must be able to find it from
+# the include/trace directory.
+#
+# Here trace-events-sample.c does the CREATE_TRACE_POINTS.
+#
 CFLAGS_trace-events-sample.o := -I$(src)
 
 obj-$(CONFIG_SAMPLE_TRACE_EVENTS) += trace-events-sample.o
samples/trace_events/trace-events-sample.h +16 -11
index 128a897..9977a75 100644
@@ -19,16 +19,21 @@
  * If TRACE_SYSTEM is defined, that will be the directory created
  * in the ftrace directory under /debugfs/tracing/events/<system>
  *
- * The define_trace.h belowe will also look for a file name of
+ * The define_trace.h below will also look for a file name of
  * TRACE_SYSTEM.h where TRACE_SYSTEM is what is defined here.
+ * In this case, it would look for sample.h
  *
- * If you want a different system than file name, you can override
- * the header name by defining TRACE_INCLUDE_FILE
+ * If the header name will be different than the system name
+ * (as in this case), then you can override the header name that
+ * define_trace.h will look up by defining TRACE_INCLUDE_FILE
  *
- * If this file was called, goofy.h, then we would define:
+ * This file is called trace-events-sample.h but we want the system
+ * to be called "sample". Therefore we must define the name of this
+ * file:
  *
- * #define TRACE_INCLUDE_FILE goofy
+ * #define TRACE_INCLUDE_FILE trace-events-sample
  *
+ * As we do an the bottom of this file.
  */
 #undef TRACE_SYSTEM
 #define TRACE_SYSTEM sample
@@ -99,13 +104,13 @@ TRACE_EVENT(foo_bar,
  *
  * #define TRACE_INCLUDE_PATH ../../samples/trace_events
  *
- * But I chose to simply make it use the current directory and then in
- * the Makefile I added:
+ * But the safest and easiest way to simply make it use the directory
+ * that the file is in is to add in the Makefile:
  *
- * CFLAGS_trace-events-sample.o := -I$(PWD)/samples/trace_events/
+ * CFLAGS_trace-events-sample.o := -I$(src)
  *
  * This will make sure the current path is part of the include
- * structure for our file so that we can find it.
+ * structure for our file so that define_trace.h can find it.
  *
  * I could have made only the top level directory the include:
  *
@@ -115,8 +120,8 @@ TRACE_EVENT(foo_bar,
  *
  * #define TRACE_INCLUDE_PATH samples/trace_events
  *
- * But then if something defines "samples" or "trace_events" then we
- * could risk that being converted too, and give us an unexpected
+ * But then if something defines "samples" or "trace_events" as a macro
+ * then we could risk that being converted too, and give us an unexpected
  * result.
  */
 #undef TRACE_INCLUDE_PATH