Merge branch 'perf-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

Pull perf changes from Ingo Molnar:
 "Core kernel changes:

   - One of the more interesting features in this cycle is the ability
     to attach eBPF programs (user-defined, sandboxed bytecode executed
     by the kernel) to kprobes.

     This allows user-defined instrumentation on a live kernel image
     that can never crash, hang or interfere with the kernel negatively.
     (Right now it's limited to root-only, but in the future we might
     allow unprivileged use as well.)

     (Alexei Starovoitov)

   - Another non-trivial feature is per event clockid support: this
     allows, amongst other things, the selection of different clock
     sources for event timestamps traced via perf.

     This feature is sought by people who'd like to merge perf generated
     events with external events that were measured with different
     clocks:

       - cluster wide profiling

       - for system wide tracing with user-space events,

       - JIT profiling events

     etc.  Matching perf tooling support is added as well, available via
     the -k, --clockid <clockid> parameter to perf record et al.

     (Peter Zijlstra)

  Hardware enablement kernel changes:

   - x86 Intel Processor Trace (PT) support: which is a hardware tracer
     on steroids, available on Broadwell CPUs.

     The hardware trace stream is directly output into the user-space
     ring-buffer, using the 'AUX' data format extension that was added
     to the perf core to support hardware constraints such as the
     necessity to have the tracing buffer physically contiguous.

     This patch-set was developed for two years and this is the result.
     A simple way to make use of this is to use BTS tracing, the PT
     driver emulates BTS output - available via the 'intel_bts' PMU.
     More explicit PT specific tooling support is in the works as well -
     will probably be ready by 4.2.

     (Alexander Shishkin, Peter Zijlstra)

   - x86 Intel Cache QoS Monitoring (CQM) support: this is a hardware
     feature of Intel Xeon CPUs that allows the measurement and
     allocation/partitioning of caches to individual workloads.

     These kernel changes expose the measurement side as a new PMU
     driver, which exposes various QoS related PMU events.  (The
     partitioning change is work in progress and is planned to be merged
     as a cgroup extension.)

     (Matt Fleming, Peter Zijlstra; CPU feature detection by Peter P
     Waskiewicz Jr)

   - x86 Intel Haswell LBR call stack support: this is a new Haswell
     feature that allows the hardware recording of call chains, plus
     tooling support.  To activate this feature you have to enable it
     via the new 'lbr' call-graph recording option:

        perf record --call-graph lbr
        perf report

     or:

        perf top --call-graph lbr

     This hardware feature is a lot faster than stack walk or dwarf
     based unwinding, but has some limitations:

       - It reuses the current LBR facility, so LBR call stack and
         branch record can not be enabled at the same time.

       - It is only available for user-space callchains.

     (Yan, Zheng)

   - x86 Intel Broadwell CPU support and various event constraints and
     event table fixes for earlier models.

     (Andi Kleen)

   - x86 Intel HT CPUs event scheduling workarounds.  This is a complex
     CPU bug affecting the SNB,IVB,HSW families that results in counter
     value corruption.  The mitigation code is automatically enabled and
     is transparent.

     (Maria Dimakopoulou, Stephane Eranian)

  The perf tooling side had a ton of changes in this cycle as well, so
  I'm only able to list the user visible changes here, in addition to
  the tooling changes outlined above:

  User visible changes affecting all tools:

      - Improve support of compressed kernel modules (Jiri Olsa)
      - Save DSO loading errno to better report errors (Arnaldo Carvalho de Melo)
      - Bash completion for subcommands (Yunlong Song)
      - Add 'I' event modifier for perf_event_attr.exclude_idle bit (Jiri Olsa)
      - Support missing -f to override perf.data file ownership. (Yunlong Song)
      - Show the first event with an invalid filter (David Ahern, Arnaldo Carvalho de Melo)

  User visible changes in individual tools:

    'perf data':

        New tool for converting perf.data to other formats, initially
        for the CTF (Common Trace Format) from LTTng (Jiri Olsa,
        Sebastian Siewior)

    'perf diff':

        Add --kallsyms option (David Ahern)

    'perf list':

        Allow listing events with 'tracepoint' prefix (Yunlong Song)

        Sort the output of the command (Yunlong Song)

    'perf kmem':

        Respect -i option (Jiri Olsa)

        Print big numbers using thousands' group (Namhyung Kim)

        Allow -v option (Namhyung Kim)

        Fix alignment of slab result table (Namhyung Kim)

    'perf probe':

        Support multiple probes on different binaries on the same command line (Masami Hiramatsu)

        Support unnamed union/structure members data collection. (Masami Hiramatsu)

        Check kprobes blacklist when adding new events. (Masami Hiramatsu)

    'perf record':

        Teach 'perf record' about perf_event_attr.clockid (Peter Zijlstra)

        Support recording running/enabled time (Andi Kleen)

    'perf sched':

        Improve the performance of 'perf sched replay' on high CPU core count machines (Yunlong Song)

    'perf report' and 'perf top':

        Allow annotating entries in callchains in the hists browser (Arnaldo Carvalho de Melo)

        Indicate which callchain entries are annotated in the
        TUI hists browser (Arnaldo Carvalho de Melo)

        Add pid/tid filtering to 'report' and 'script' commands (David Ahern)

        Consider PERF_RECORD_ events with cpumode == 0 in 'perf top', removing one
        cause of long term memory usage buildup, i.e. not processing PERF_RECORD_EXIT
        events (Arnaldo Carvalho de Melo)

    'perf stat':

        Report unsupported events properly (Suzuki K. Poulose)

        Output running time and run/enabled ratio in CSV mode (Andi Kleen)

    'perf trace':

        Handle legacy syscalls tracepoints (David Ahern, Arnaldo Carvalho de Melo)

        Only insert blank duration bracket when tracing syscalls (Arnaldo Carvalho de Melo)

        Filter out the trace pid when no threads are specified (Arnaldo Carvalho de Melo)

        Dump stack on segfaults (Arnaldo Carvalho de Melo)

        No need to explicitely enable evsels for workload started from perf, let it
        be enabled via perf_event_attr.enable_on_exec, removing some events that take
        place in the 'perf trace' before a workload is really started by it.
        (Arnaldo Carvalho de Melo)

        Allow mixing with tracepoints and suppressing plain syscalls. (Arnaldo Carvalho de Melo)

  There's also been a ton of infrastructure work done, such as the
  split-out of perf's build system into tools/build/ and other changes -
  see the shortlog and changelog for details"

* 'perf-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (358 commits)
  perf/x86/intel/pt: Clean up the control flow in pt_pmu_hw_init()
  perf evlist: Fix type for references to data_head/tail
  perf probe: Check the orphaned -x option
  perf probe: Support multiple probes on different binaries
  perf buildid-list: Fix segfault when show DSOs with hits
  perf tools: Fix cross-endian analysis
  perf tools: Fix error path to do closedir() when synthesizing threads
  perf tools: Fix synthesizing fork_event.ppid for non-main thread
  perf tools: Add 'I' event modifier for exclude_idle bit
  perf report: Don't call map__kmap if map is NULL.
  perf tests: Fix attr tests
  perf probe: Fix ARM 32 building error
  perf tools: Merge all perf_event_attr print functions
  perf record: Add clockid parameter
  perf sched replay: Use replay_repeat to calculate the runavg of cpu usage instead of the default value 10
  perf sched replay: Support using -f to override perf.data file ownership
  perf sched replay: Fix the EMFILE error caused by the limitation of the maximum open files
  perf sched replay: Handle the dead halt of sem_wait when create_tasks() fails for any task
  perf sched replay: Fix the segmentation fault problem caused by pr_err in threads
  perf sched replay: Realloc the memory of pid_to_task stepwise to adapt to the different pid_max configurations
  ...

1 files changed, 262 insertions, 43 deletions

diff --git a/tools/lib/traceevent/event-parse.c b/tools/lib/traceevent/event-parse.c
index 2c0bd8f2aad0..e0917c0f5d9f 100644
--- a/tools/lib/traceevent/event-parse.c
+++ b/tools/lib/traceevent/event-parse.c
@@ -304,7 +304,10 @@ int pevent_register_comm(struct pevent *pevent, const char *comm, int pid)
 	if (!item)
 		return -1;
 
-	item->comm = strdup(comm);
+	if (comm)
+		item->comm = strdup(comm);
+	else
+		item->comm = strdup("<...>");
 	if (!item->comm) {
 		free(item);
 		return -1;
@@ -318,9 +321,14 @@ int pevent_register_comm(struct pevent *pevent, const char *comm, int pid)
 	return 0;
 }
 
-void pevent_register_trace_clock(struct pevent *pevent, char *trace_clock)
+int pevent_register_trace_clock(struct pevent *pevent, const char *trace_clock)
 {
-	pevent->trace_clock = trace_clock;
+	pevent->trace_clock = strdup(trace_clock);
+	if (!pevent->trace_clock) {
+		errno = ENOMEM;
+		return -1;
+	}
+	return 0;
 }
 
 struct func_map {
@@ -758,6 +766,11 @@ static void free_arg(struct print_arg *arg)
 		free_arg(arg->hex.field);
 		free_arg(arg->hex.size);
 		break;
+	case PRINT_INT_ARRAY:
+		free_arg(arg->int_array.field);
+		free_arg(arg->int_array.count);
+		free_arg(arg->int_array.el_size);
+		break;
 	case PRINT_TYPE:
 		free(arg->typecast.type);
 		free_arg(arg->typecast.item);
@@ -1926,7 +1939,22 @@ process_op(struct event_format *event, struct print_arg *arg, char **tok)
 			goto out_warn_free;
 
 		type = process_arg_token(event, right, tok, type);
-		arg->op.right = right;
+
+		if (right->type == PRINT_OP &&
+		    get_op_prio(arg->op.op) < get_op_prio(right->op.op)) {
+			struct print_arg tmp;
+
+			/* rotate ops according to the priority */
+			arg->op.right = right->op.left;
+
+			tmp = *arg;
+			*arg = *right;
+			*right = tmp;
+
+			arg->op.left = right;
+		} else {
+			arg->op.right = right;
+		}
 
 	} else if (strcmp(token, "[") == 0) {
 
@@ -2014,6 +2042,38 @@ process_entry(struct event_format *event __maybe_unused, struct print_arg *arg,
 	return EVENT_ERROR;
 }
 
+static int alloc_and_process_delim(struct event_format *event, char *next_token,
+				   struct print_arg **print_arg)
+{
+	struct print_arg *field;
+	enum event_type type;
+	char *token;
+	int ret = 0;
+
+	field = alloc_arg();
+	if (!field) {
+		do_warning_event(event, "%s: not enough memory!", __func__);
+		errno = ENOMEM;
+		return -1;
+	}
+
+	type = process_arg(event, field, &token);
+
+	if (test_type_token(type, token, EVENT_DELIM, next_token)) {
+		errno = EINVAL;
+		ret = -1;
+		free_arg(field);
+		goto out_free_token;
+	}
+
+	*print_arg = field;
+
+out_free_token:
+	free_token(token);
+
+	return ret;
+}
+
 static char *arg_eval (struct print_arg *arg);
 
 static unsigned long long
@@ -2486,49 +2546,46 @@ out_free:
 static enum event_type
 process_hex(struct event_format *event, struct print_arg *arg, char **tok)
 {
-	struct print_arg *field;
-	enum event_type type;
-	char *token = NULL;
-
 	memset(arg, 0, sizeof(*arg));
 	arg->type = PRINT_HEX;
 
-	field = alloc_arg();
-	if (!field) {
-		do_warning_event(event, "%s: not enough memory!", __func__);
-		goto out_free;
-	}
+	if (alloc_and_process_delim(event, ",", &arg->hex.field))
+		goto out;
 
-	type = process_arg(event, field, &token);
+	if (alloc_and_process_delim(event, ")", &arg->hex.size))
+		goto free_field;
 
-	if (test_type_token(type, token, EVENT_DELIM, ","))
-		goto out_free;
-
-	arg->hex.field = field;
+	return read_token_item(tok);
 
-	free_token(token);
+free_field:
+	free_arg(arg->hex.field);
+out:
+	*tok = NULL;
+	return EVENT_ERROR;
+}
 
-	field = alloc_arg();
-	if (!field) {
-		do_warning_event(event, "%s: not enough memory!", __func__);
-		*tok = NULL;
-		return EVENT_ERROR;
-	}
+static enum event_type
+process_int_array(struct event_format *event, struct print_arg *arg, char **tok)
+{
+	memset(arg, 0, sizeof(*arg));
+	arg->type = PRINT_INT_ARRAY;
 
-	type = process_arg(event, field, &token);
+	if (alloc_and_process_delim(event, ",", &arg->int_array.field))
+		goto out;
 
-	if (test_type_token(type, token, EVENT_DELIM, ")"))
-		goto out_free;
+	if (alloc_and_process_delim(event, ",", &arg->int_array.count))
+		goto free_field;
 
-	arg->hex.size = field;
+	if (alloc_and_process_delim(event, ")", &arg->int_array.el_size))
+		goto free_size;
 
-	free_token(token);
-	type = read_token_item(tok);
-	return type;
+	return read_token_item(tok);
 
- out_free:
-	free_arg(field);
-	free_token(token);
+free_size:
+	free_arg(arg->int_array.count);
+free_field:
+	free_arg(arg->int_array.field);
+out:
 	*tok = NULL;
 	return EVENT_ERROR;
 }
@@ -2828,6 +2885,10 @@ process_function(struct event_format *event, struct print_arg *arg,
 		free_token(token);
 		return process_hex(event, arg, tok);
 	}
+	if (strcmp(token, "__print_array") == 0) {
+		free_token(token);
+		return process_int_array(event, arg, tok);
+	}
 	if (strcmp(token, "__get_str") == 0) {
 		free_token(token);
 		return process_str(event, arg, tok);
@@ -3356,6 +3417,7 @@ eval_num_arg(void *data, int size, struct event_format *event, struct print_arg
 		break;
 	case PRINT_FLAGS:
 	case PRINT_SYMBOL:
+	case PRINT_INT_ARRAY:
 	case PRINT_HEX:
 		break;
 	case PRINT_TYPE:
@@ -3568,7 +3630,7 @@ static const struct flag flags[] = {
 	{ "HRTIMER_RESTART", 1 },
 };
 
-static unsigned long long eval_flag(const char *flag)
+static long long eval_flag(const char *flag)
 {
 	int i;
 
@@ -3584,7 +3646,7 @@ static unsigned long long eval_flag(const char *flag)
 		if (strcmp(flags[i].name, flag) == 0)
 			return flags[i].value;
 
-	return 0;
+	return -1LL;
 }
 
 static void print_str_to_seq(struct trace_seq *s, const char *format,
@@ -3658,7 +3720,7 @@ static void print_str_arg(struct trace_seq *s, void *data, int size,
 	struct print_flag_sym *flag;
 	struct format_field *field;
 	struct printk_map *printk;
-	unsigned long long val, fval;
+	long long val, fval;
 	unsigned long addr;
 	char *str;
 	unsigned char *hex;
@@ -3717,11 +3779,11 @@ static void print_str_arg(struct trace_seq *s, void *data, int size,
 		print = 0;
 		for (flag = arg->flags.flags; flag; flag = flag->next) {
 			fval = eval_flag(flag->value);
-			if (!val && !fval) {
+			if (!val && fval < 0) {
 				print_str_to_seq(s, format, len_arg, flag->str);
 				break;
 			}
-			if (fval && (val & fval) == fval) {
+			if (fval > 0 && (val & fval) == fval) {
 				if (print && arg->flags.delim)
 					trace_seq_puts(s, arg->flags.delim);
 				print_str_to_seq(s, format, len_arg, flag->str);
@@ -3766,6 +3828,54 @@ static void print_str_arg(struct trace_seq *s, void *data, int size,
 		}
 		break;
 
+	case PRINT_INT_ARRAY: {
+		void *num;
+		int el_size;
+
+		if (arg->int_array.field->type == PRINT_DYNAMIC_ARRAY) {
+			unsigned long offset;
+			struct format_field *field =
+				arg->int_array.field->dynarray.field;
+			offset = pevent_read_number(pevent,
+						    data + field->offset,
+						    field->size);
+			num = data + (offset & 0xffff);
+		} else {
+			field = arg->int_array.field->field.field;
+			if (!field) {
+				str = arg->int_array.field->field.name;
+				field = pevent_find_any_field(event, str);
+				if (!field)
+					goto out_warning_field;
+				arg->int_array.field->field.field = field;
+			}
+			num = data + field->offset;
+		}
+		len = eval_num_arg(data, size, event, arg->int_array.count);
+		el_size = eval_num_arg(data, size, event,
+				       arg->int_array.el_size);
+		for (i = 0; i < len; i++) {
+			if (i)
+				trace_seq_putc(s, ' ');
+
+			if (el_size == 1) {
+				trace_seq_printf(s, "%u", *(uint8_t *)num);
+			} else if (el_size == 2) {
+				trace_seq_printf(s, "%u", *(uint16_t *)num);
+			} else if (el_size == 4) {
+				trace_seq_printf(s, "%u", *(uint32_t *)num);
+			} else if (el_size == 8) {
+				trace_seq_printf(s, "%lu", *(uint64_t *)num);
+			} else {
+				trace_seq_printf(s, "BAD SIZE:%d 0x%x",
+						 el_size, *(uint8_t *)num);
+				el_size = 1;
+			}
+
+			num += el_size;
+		}
+		break;
+	}
 	case PRINT_TYPE:
 		break;
 	case PRINT_STRING: {
@@ -3997,6 +4107,10 @@ static struct print_arg *make_bprint_args(char *fmt, void *data, int size, struc
 				goto process_again;
 			case '.':
 				goto process_again;
+			case 'z':
+			case 'Z':
+				ls = 1;
+				goto process_again;
 			case 'p':
 				ls = 1;
 				/* fall through */
@@ -4939,6 +5053,96 @@ const char *pevent_data_comm_from_pid(struct pevent *pevent, int pid)
 	return comm;
 }
 
+static struct cmdline *
+pid_from_cmdlist(struct pevent *pevent, const char *comm, struct cmdline *next)
+{
+	struct cmdline_list *cmdlist = (struct cmdline_list *)next;
+
+	if (cmdlist)
+		cmdlist = cmdlist->next;
+	else
+		cmdlist = pevent->cmdlist;
+
+	while (cmdlist && strcmp(cmdlist->comm, comm) != 0)
+		cmdlist = cmdlist->next;
+
+	return (struct cmdline *)cmdlist;
+}
+
+/**
+ * pevent_data_pid_from_comm - return the pid from a given comm
+ * @pevent: a handle to the pevent
+ * @comm: the cmdline to find the pid from
+ * @next: the cmdline structure to find the next comm
+ *
+ * This returns the cmdline structure that holds a pid for a given
+ * comm, or NULL if none found. As there may be more than one pid for
+ * a given comm, the result of this call can be passed back into
+ * a recurring call in the @next paramater, and then it will find the
+ * next pid.
+ * Also, it does a linear seach, so it may be slow.
+ */
+struct cmdline *pevent_data_pid_from_comm(struct pevent *pevent, const char *comm,
+					  struct cmdline *next)
+{
+	struct cmdline *cmdline;
+
+	/*
+	 * If the cmdlines have not been converted yet, then use
+	 * the list.
+	 */
+	if (!pevent->cmdlines)
+		return pid_from_cmdlist(pevent, comm, next);
+
+	if (next) {
+		/*
+		 * The next pointer could have been still from
+		 * a previous call before cmdlines were created
+		 */
+		if (next < pevent->cmdlines ||
+		    next >= pevent->cmdlines + pevent->cmdline_count)
+			next = NULL;
+		else
+			cmdline  = next++;
+	}
+
+	if (!next)
+		cmdline = pevent->cmdlines;
+
+	while (cmdline < pevent->cmdlines + pevent->cmdline_count) {
+		if (strcmp(cmdline->comm, comm) == 0)
+			return cmdline;
+		cmdline++;
+	}
+	return NULL;
+}
+
+/**
+ * pevent_cmdline_pid - return the pid associated to a given cmdline
+ * @cmdline: The cmdline structure to get the pid from
+ *
+ * Returns the pid for a give cmdline. If @cmdline is NULL, then
+ * -1 is returned.
+ */
+int pevent_cmdline_pid(struct pevent *pevent, struct cmdline *cmdline)
+{
+	struct cmdline_list *cmdlist = (struct cmdline_list *)cmdline;
+
+	if (!cmdline)
+		return -1;
+
+	/*
+	 * If cmdlines have not been created yet, or cmdline is
+	 * not part of the array, then treat it as a cmdlist instead.
+	 */
+	if (!pevent->cmdlines ||
+	    cmdline < pevent->cmdlines ||
+	    cmdline >= pevent->cmdlines + pevent->cmdline_count)
+		return cmdlist->pid;
+
+	return cmdline->pid;
+}
+
 /**
  * pevent_data_comm_from_pid - parse the data into the print format
  * @s: the trace_seq to write to
@@ -5256,6 +5460,15 @@ static void print_args(struct print_arg *args)
 		print_args(args->hex.size);
 		printf(")");
 		break;
+	case PRINT_INT_ARRAY:
+		printf("__print_array(");
+		print_args(args->int_array.field);
+		printf(", ");
+		print_args(args->int_array.count);
+		printf(", ");
+		print_args(args->int_array.el_size);
+		printf(")");
+		break;
 	case PRINT_STRING:
 	case PRINT_BSTRING:
 		printf("__get_str(%s)", args->string.string);
@@ -6228,15 +6441,20 @@ void pevent_ref(struct pevent *pevent)
 	pevent->ref_count++;
 }
 
+void pevent_free_format_field(struct format_field *field)
+{
+	free(field->type);
+	free(field->name);
+	free(field);
+}
+
 static void free_format_fields(struct format_field *field)
 {
 	struct format_field *next;
 
 	while (field) {
 		next = field->next;
-		free(field->type);
-		free(field->name);
-		free(field);
+		pevent_free_format_field(field);
 		field = next;
 	}
 }
@@ -6341,6 +6559,7 @@ void pevent_free(struct pevent *pevent)
 		free_handler(handle);
 	}
 
+	free(pevent->trace_clock);
 	free(pevent->events);
 	free(pevent->sort_events);