Merge tag 'perf-urgent-for-mingo-4.20-20181106' of...

Merge tag 'perf-urgent-for-mingo-4.20-20181106' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux into perf/urgent

Pull perf/urgent improvements and fixes from Arnaldo Carvalho de Melo:

Intel PT SQL viewer: (Adrian Hunter)

- Fall back to /usr/local/lib/libxed.so
- Add Selected branches report
- Add help window
- Fix table find when table re-ordered

Intel PT debug log (Adrian Hunter)

- Add more event information
- Add MTC and CYC timestamps

perf record: (Andi Kleen)

- Support weak groups, just like with 'perf stat'

perf trace: (Arnaldo Carvalho de Melo)

- Start augmenting raw_syscalls:{sys_enter,sys_exit}: goal is to have a
  generic, arch independent eBPF kernel component that is programmed with
  syscall table details, what to copy, how many bytes, pid, arg filters from the
  userspace via eBPF maps by the 'perf trace' tool that continues to use all its
  argument beautifiers, just taking advantage of the extra pointer contents.

JVMTI: (Gustavo Romero)

- Fix undefined symbol scnprintf in libperf-jvmti.so

perf top: (Jin Yao)

- Display the LBR stats in callchain entries

perf stat: (Thomas Richter)

- Handle different PMU names with common prefix

arm64: Will (Deacon)

- Fix arm64 tools build failure wrt smp_load_{acquire,release}.
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>

tools/arch/arm64/include/asm/barrier.h

@@ -17,31 +17,31 @@
 #define smp_store_release(p, v)						\
 do {									\
 	union { typeof(*p) __val; char __c[1]; } __u =			\
-		{ .__val = (__force typeof(*p)) (v) }; 		\
+		{ .__val = (v) }; 					\
 									\
 	switch (sizeof(*p)) {						\
 	case 1:								\
 		asm volatile ("stlrb %w1, %0"				\
 				: "=Q" (*p)				\
-				: "r" (*(__u8 *)__u.__c)	\
+				: "r" (*(__u8_alias_t *)__u.__c)	\
 				: "memory");				\
 		break;							\
 	case 2:								\
 		asm volatile ("stlrh %w1, %0"				\
 				: "=Q" (*p)				\
-				: "r" (*(__u16 *)__u.__c)	\
+				: "r" (*(__u16_alias_t *)__u.__c)	\
 				: "memory");				\
 		break;							\
 	case 4:								\
 		asm volatile ("stlr %w1, %0"				\
 				: "=Q" (*p)				\
-				: "r" (*(__u32 *)__u.__c)	\
+				: "r" (*(__u32_alias_t *)__u.__c)	\
 				: "memory");				\
 		break;							\
 	case 8:								\
 		asm volatile ("stlr %1, %0"				\
 				: "=Q" (*p)				\
-				: "r" (*(__u64 *)__u.__c)	\
+				: "r" (*(__u64_alias_t *)__u.__c)	\
 				: "memory");				\
 		break;							\
 	default:							\
@@ -53,27 +53,28 @@ do {								\
 
 #define smp_load_acquire(p)						\
 ({									\
-	union { typeof(*p) __val; char __c[1]; } __u;		\
+	union { typeof(*p) __val; char __c[1]; } __u =			\
+		{ .__c = { 0 } };					\
 									\
 	switch (sizeof(*p)) {						\
 	case 1:								\
 		asm volatile ("ldarb %w0, %1"				\
-			: "=r" (*(__u8 *)__u.__c)		\
+			: "=r" (*(__u8_alias_t *)__u.__c)		\
 			: "Q" (*p) : "memory");				\
 		break;							\
 	case 2:								\
 		asm volatile ("ldarh %w0, %1"				\
-			: "=r" (*(__u16 *)__u.__c)		\
+			: "=r" (*(__u16_alias_t *)__u.__c)		\
 			: "Q" (*p) : "memory");				\
 		break;							\
 	case 4:								\
 		asm volatile ("ldar %w0, %1"				\
-			: "=r" (*(__u32 *)__u.__c)		\
+			: "=r" (*(__u32_alias_t *)__u.__c)		\
 			: "Q" (*p) : "memory");				\
 		break;							\
 	case 8:								\
 		asm volatile ("ldar %0, %1"				\
-			: "=r" (*(__u64 *)__u.__c)		\
+			: "=r" (*(__u64_alias_t *)__u.__c)		\
 			: "Q" (*p) : "memory");				\
 		break;							\
 	default:							\

tools/perf/Documentation/perf-list.txt

@@ -55,7 +55,6 @@ counted. The following modifiers exist:
  S - read sample value (PERF_SAMPLE_READ)
  D - pin the event to the PMU
  W - group is weak and will fallback to non-group if not schedulable,
-     only supported in 'perf stat' for now.
 
 The 'p' modifier can be used for specifying how precise the instruction
 address should be. The 'p' modifier can be specified multiple times:

tools/perf/Makefile.perf

@@ -387,7 +387,7 @@ SHELL = $(SHELL_PATH)
 
 linux_uapi_dir := $(srctree)/tools/include/uapi/linux
 asm_generic_uapi_dir := $(srctree)/tools/include/uapi/asm-generic
-arch_asm_uapi_dir := $(srctree)/tools/arch/$(ARCH)/include/uapi/asm/
+arch_asm_uapi_dir := $(srctree)/tools/arch/$(SRCARCH)/include/uapi/asm/
 
 beauty_outdir := $(OUTPUT)trace/beauty/generated
 beauty_ioctl_outdir := $(beauty_outdir)/ioctl

tools/perf/builtin-record.c

@@ -391,7 +391,12 @@ static int record__open(struct record *rec)
 					ui__warning("%s\n", msg);
 				goto try_again;
 			}
-
+			if ((errno == EINVAL || errno == EBADF) &&
+			    pos->leader != pos &&
+			    pos->weak_group) {
+			        pos = perf_evlist__reset_weak_group(evlist, pos);
+				goto try_again;
+			}
 			rc = -errno;
 			perf_evsel__open_strerror(pos, &opts->target,
 						  errno, msg, sizeof(msg));

tools/perf/builtin-stat.c

@@ -383,32 +383,6 @@ static bool perf_evsel__should_store_id(struct perf_evsel *counter)
 	return STAT_RECORD || counter->attr.read_format & PERF_FORMAT_ID;
 }
 
-static struct perf_evsel *perf_evsel__reset_weak_group(struct perf_evsel *evsel)
-{
-	struct perf_evsel *c2, *leader;
-	bool is_open = true;
-
-	leader = evsel->leader;
-	pr_debug("Weak group for %s/%d failed\n",
-			leader->name, leader->nr_members);
-
-	/*
-	 * for_each_group_member doesn't work here because it doesn't
-	 * include the first entry.
-	 */
-	evlist__for_each_entry(evsel_list, c2) {
-		if (c2 == evsel)
-			is_open = false;
-		if (c2->leader == leader) {
-			if (is_open)
-				perf_evsel__close(c2);
-			c2->leader = c2;
-			c2->nr_members = 0;
-		}
-	}
-	return leader;
-}
-
 static bool is_target_alive(struct target *_target,
 			    struct thread_map *threads)
 {
@@ -477,7 +451,7 @@ static int __run_perf_stat(int argc, const char **argv, int run_idx)
 			if ((errno == EINVAL || errno == EBADF) &&
 			    counter->leader != counter &&
 			    counter->weak_group) {
-				counter = perf_evsel__reset_weak_group(counter);
+				counter = perf_evlist__reset_weak_group(evsel_list, counter);
 				goto try_again;
 			}
 

tools/perf/builtin-top.c

@@ -1429,6 +1429,9 @@ int cmd_top(int argc, const char **argv)
 		}
 	}
 
+	if (opts->branch_stack && callchain_param.enabled)
+		symbol_conf.show_branchflag_count = true;
+
 	sort__mode = SORT_MODE__TOP;
 	/* display thread wants entries to be collapsed in a different tree */
 	perf_hpp_list.need_collapse = 1;

tools/perf/builtin-trace.c

@@ -108,6 +108,7 @@ struct trace {
 	} stats;
 	unsigned int		max_stack;
 	unsigned int		min_stack;
+	bool			raw_augmented_syscalls;
 	bool			not_ev_qualifier;
 	bool			live;
 	bool			full_time;
@@ -1724,13 +1725,28 @@ static int trace__fprintf_sample(struct trace *trace, struct perf_evsel *evsel,
 	return printed;
 }
 
-static void *syscall__augmented_args(struct syscall *sc, struct perf_sample *sample, int *augmented_args_size)
+static void *syscall__augmented_args(struct syscall *sc, struct perf_sample *sample, int *augmented_args_size, bool raw_augmented)
 {
 	void *augmented_args = NULL;
+	/*
+	 * For now with BPF raw_augmented we hook into raw_syscalls:sys_enter
+	 * and there we get all 6 syscall args plus the tracepoint common
+	 * fields (sizeof(long)) and the syscall_nr (another long). So we check
+	 * if that is the case and if so don't look after the sc->args_size,
+	 * but always after the full raw_syscalls:sys_enter payload, which is
+	 * fixed.
+	 *
+	 * We'll revisit this later to pass s->args_size to the BPF augmenter
+	 * (now tools/perf/examples/bpf/augmented_raw_syscalls.c, so that it
+	 * copies only what we need for each syscall, like what happens when we
+	 * use syscalls:sys_enter_NAME, so that we reduce the kernel/userspace
+	 * traffic to just what is needed for each syscall.
+	 */
+	int args_size = raw_augmented ? (8 * (int)sizeof(long)) : sc->args_size;
 
-	*augmented_args_size = sample->raw_size - sc->args_size;
+	*augmented_args_size = sample->raw_size - args_size;
 	if (*augmented_args_size > 0)
-		augmented_args = sample->raw_data + sc->args_size;
+		augmented_args = sample->raw_data + args_size;
 
 	return augmented_args;
 }
@@ -1780,7 +1796,7 @@ static int trace__sys_enter(struct trace *trace, struct perf_evsel *evsel,
 	 * here and avoid using augmented syscalls when the evsel is the raw_syscalls one.
 	 */
 	if (evsel != trace->syscalls.events.sys_enter)
-		augmented_args = syscall__augmented_args(sc, sample, &augmented_args_size);
+		augmented_args = syscall__augmented_args(sc, sample, &augmented_args_size, trace->raw_augmented_syscalls);
 	ttrace->entry_time = sample->time;
 	msg = ttrace->entry_str;
 	printed += scnprintf(msg + printed, trace__entry_str_size - printed, "%s(", sc->name);
@@ -1833,7 +1849,7 @@ static int trace__fprintf_sys_enter(struct trace *trace, struct perf_evsel *evse
 		goto out_put;
 
 	args = perf_evsel__sc_tp_ptr(evsel, args, sample);
-	augmented_args = syscall__augmented_args(sc, sample, &augmented_args_size);
+	augmented_args = syscall__augmented_args(sc, sample, &augmented_args_size, trace->raw_augmented_syscalls);
 	syscall__scnprintf_args(sc, msg, sizeof(msg), args, augmented_args, augmented_args_size, trace, thread);
 	fprintf(trace->output, "%s", msg);
 	err = 0;
@@ -3501,7 +3517,15 @@ int cmd_trace(int argc, const char **argv)
 		evsel->handler = trace__sys_enter;
 
 		evlist__for_each_entry(trace.evlist, evsel) {
+			bool raw_syscalls_sys_exit = strcmp(perf_evsel__name(evsel), "raw_syscalls:sys_exit") == 0;
+
+			if (raw_syscalls_sys_exit) {
+				trace.raw_augmented_syscalls = true;
+				goto init_augmented_syscall_tp;
+			}
+
 			if (strstarts(perf_evsel__name(evsel), "syscalls:sys_exit_")) {
+init_augmented_syscall_tp:
 				perf_evsel__init_augmented_syscall_tp(evsel);
 				perf_evsel__init_augmented_syscall_tp_ret(evsel);
 				evsel->handler = trace__sys_exit;

tools/perf/examples/bpf/augmented_raw_syscalls.c

+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Augment the raw_syscalls tracepoints with the contents of the pointer arguments.
+ *
+ * Test it with:
+ *
+ * perf trace -e tools/perf/examples/bpf/augmented_raw_syscalls.c cat /etc/passwd > /dev/null
+ *
+ * This exactly matches what is marshalled into the raw_syscall:sys_enter
+ * payload expected by the 'perf trace' beautifiers.
+ *
+ * For now it just uses the existing tracepoint augmentation code in 'perf
+ * trace', in the next csets we'll hook up these with the sys_enter/sys_exit
+ * code that will combine entry/exit in a strace like way.
+ */
+
+#include <stdio.h>
+#include <linux/socket.h>
+
+/* bpf-output associated map */
+struct bpf_map SEC("maps") __augmented_syscalls__ = {
+	.type = BPF_MAP_TYPE_PERF_EVENT_ARRAY,
+	.key_size = sizeof(int),
+	.value_size = sizeof(u32),
+	.max_entries = __NR_CPUS__,
+};
+
+struct syscall_enter_args {
+	unsigned long long common_tp_fields;
+	long		   syscall_nr;
+	unsigned long	   args[6];
+};
+
+struct syscall_exit_args {
+	unsigned long long common_tp_fields;
+	long		   syscall_nr;
+	long		   ret;
+};
+
+struct augmented_filename {
+	unsigned int	size;
+	int		reserved;
+	char		value[256];
+};
+
+#define SYS_OPEN 2
+#define SYS_OPENAT 257
+
+SEC("raw_syscalls:sys_enter")
+int sys_enter(struct syscall_enter_args *args)
+{
+	struct {
+		struct syscall_enter_args args;
+		struct augmented_filename filename;
+	} augmented_args;
+	unsigned int len = sizeof(augmented_args);
+	const void *filename_arg = NULL;
+
+	probe_read(&augmented_args.args, sizeof(augmented_args.args), args);
+	/*
+	 * Yonghong and Edward Cree sayz:
+	 *
+	 * https://www.spinics.net/lists/netdev/msg531645.html
+	 *
+	 * >>   R0=inv(id=0) R1=inv2 R6=ctx(id=0,off=0,imm=0) R7=inv64 R10=fp0,call_-1
+	 * >> 10: (bf) r1 = r6
+	 * >> 11: (07) r1 += 16
+	 * >> 12: (05) goto pc+2
+	 * >> 15: (79) r3 = *(u64 *)(r1 +0)
+	 * >> dereference of modified ctx ptr R1 off=16 disallowed
+	 * > Aha, we at least got a different error message this time.
+	 * > And indeed llvm has done that optimisation, rather than the more obvious
+	 * > 11: r3 = *(u64 *)(r1 +16)
+	 * > because it wants to have lots of reads share a single insn.  You may be able
+	 * > to defeat that optimisation by adding compiler barriers, idk.  Maybe someone
+	 * > with llvm knowledge can figure out how to stop it (ideally, llvm would know
+	 * > when it's generating for bpf backend and not do that).  -O0?  ¯\_(ツ)_/¯
+	 *
+	 * The optimization mostly likes below:
+	 *
+	 *	br1:
+	 * 	...
+	 *	r1 += 16
+	 *	goto merge
+	 *	br2:
+	 *	...
+	 *	r1 += 20
+	 *	goto merge
+	 *	merge:
+	 *	*(u64 *)(r1 + 0)
+	 *
+	 * The compiler tries to merge common loads. There is no easy way to
+	 * stop this compiler optimization without turning off a lot of other
+	 * optimizations. The easiest way is to add barriers:
+	 *
+	 * 	 __asm__ __volatile__("": : :"memory")
+	 *
+	 * 	 after the ctx memory access to prevent their down stream merging.
+	 */
+	switch (augmented_args.args.syscall_nr) {
+	case SYS_OPEN:	 filename_arg = (const void *)args->args[0];
+			__asm__ __volatile__("": : :"memory");
+			 break;
+	case SYS_OPENAT: filename_arg = (const void *)args->args[1];
+			 break;
+	}
+
+	if (filename_arg != NULL) {
+		augmented_args.filename.reserved = 0;
+		augmented_args.filename.size = probe_read_str(&augmented_args.filename.value,
+							      sizeof(augmented_args.filename.value),
+							      filename_arg);
+		if (augmented_args.filename.size < sizeof(augmented_args.filename.value)) {
+			len -= sizeof(augmented_args.filename.value) - augmented_args.filename.size;
+			len &= sizeof(augmented_args.filename.value) - 1;
+		}
+	} else {
+		len = sizeof(augmented_args.args);
+	}
+
+	perf_event_output(args, &__augmented_syscalls__, BPF_F_CURRENT_CPU, &augmented_args, len);
+	return 0;
+}
+
+SEC("raw_syscalls:sys_exit")
+int sys_exit(struct syscall_exit_args *args)
+{
+	return 1; /* 0 as soon as we start copying data returned by the kernel, e.g. 'read' */
+}
+
+license(GPL);

tools/perf/jvmti/jvmti_agent.c

@@ -125,7 +125,7 @@ perf_get_timestamp(void)
 }
 
 static int
-debug_cache_init(void)
+create_jit_cache_dir(void)
 {
 	char str[32];
 	char *base, *p;
@@ -144,8 +144,13 @@ debug_cache_init(void)
 
 	strftime(str, sizeof(str), JIT_LANG"-jit-%Y%m%d", &tm);
 
-	snprintf(jit_path, PATH_MAX - 1, "%s/.debug/", base);
-
+	ret = snprintf(jit_path, PATH_MAX, "%s/.debug/", base);
+	if (ret >= PATH_MAX) {
+		warnx("jvmti: cannot generate jit cache dir because %s/.debug/"
+			" is too long, please check the cwd, JITDUMPDIR, and"
+			" HOME variables", base);
+		return -1;
+	}
 	ret = mkdir(jit_path, 0755);
 	if (ret == -1) {
 		if (errno != EEXIST) {
@@ -154,20 +159,32 @@ debug_cache_init(void)
 		}
 	}
 
-	snprintf(jit_path, PATH_MAX - 1, "%s/.debug/jit", base);
+	ret = snprintf(jit_path, PATH_MAX, "%s/.debug/jit", base);
+	if (ret >= PATH_MAX) {
+		warnx("jvmti: cannot generate jit cache dir because"
+			" %s/.debug/jit is too long, please check the cwd,"
+			" JITDUMPDIR, and HOME variables", base);
+		return -1;
+	}
 	ret = mkdir(jit_path, 0755);
 	if (ret == -1) {
 		if (errno != EEXIST) {
-			warn("cannot create jit cache dir %s", jit_path);
+			warn("jvmti: cannot create jit cache dir %s", jit_path);
 			return -1;
 		}
 	}
 
-	snprintf(jit_path, PATH_MAX - 1, "%s/.debug/jit/%s.XXXXXXXX", base, str);
-
+	ret = snprintf(jit_path, PATH_MAX, "%s/.debug/jit/%s.XXXXXXXX", base, str);
+	if (ret >= PATH_MAX) {
+		warnx("jvmti: cannot generate jit cache dir because"
+			" %s/.debug/jit/%s.XXXXXXXX is too long, please check"
+			" the cwd, JITDUMPDIR, and HOME variables",
+			base, str);
+		return -1;
+	}
 	p = mkdtemp(jit_path);
 	if (p != jit_path) {
-		warn("cannot create jit cache dir %s", jit_path);
+		warn("jvmti: cannot create jit cache dir %s", jit_path);
 		return -1;
 	}
 
@@ -228,7 +245,7 @@ void *jvmti_open(void)
 {
 	char dump_path[PATH_MAX];
 	struct jitheader header;
-	int fd;
+	int fd, ret;
 	FILE *fp;
 
 	init_arch_timestamp();
@@ -245,12 +262,22 @@ void *jvmti_open(void)
 
 	memset(&header, 0, sizeof(header));
 
-	debug_cache_init();
+	/*
+	 * jitdump file dir
+	 */
+	if (create_jit_cache_dir() < 0)
+		return NULL;
 
 	/*
 	 * jitdump file name
 	 */
-	scnprintf(dump_path, PATH_MAX, "%s/jit-%i.dump", jit_path, getpid());
+	ret = snprintf(dump_path, PATH_MAX, "%s/jit-%i.dump", jit_path, getpid());
+	if (ret >= PATH_MAX) {
+		warnx("jvmti: cannot generate jitdump file full path because"
+			" %s/jit-%i.dump is too long, please check the cwd,"
+			" JITDUMPDIR, and HOME variables", jit_path, getpid());
+		return NULL;
+	}
 
 	fd = open(dump_path, O_CREAT|O_TRUNC|O_RDWR, 0666);
 	if (fd == -1)

tools/perf/scripts/python/exported-sql-viewer.py

@@ -119,6 +119,14 @@ def dsoname(name):
 		return "[kernel]"
 	return name
 
+def findnth(s, sub, n, offs=0):
+	pos = s.find(sub)
+	if pos < 0:
+		return pos
+	if n <= 1:
+		return offs + pos
+	return findnth(s[pos + 1:], sub, n - 1, offs + pos + 1)
+
 # Percent to one decimal place
 
 def PercentToOneDP(n, d):
@@ -1464,6 +1472,317 @@ class BranchWindow(QMdiSubWindow):
 		else:
 			self.find_bar.NotFound()
 
+# Dialog data item converted and validated using a SQL table
+
+class SQLTableDialogDataItem():
+
+	def __init__(self, glb, label, placeholder_text, table_name, match_column, column_name1, column_name2, parent):
+		self.glb = glb
+		self.label = label
+		self.placeholder_text = placeholder_text
+		self.table_name = table_name
+		self.match_column = match_column
+		self.column_name1 = column_name1
+		self.column_name2 = column_name2
+		self.parent = parent
+
+		self.value = ""
+
+		self.widget = QLineEdit()
+		self.widget.editingFinished.connect(self.Validate)
+		self.widget.textChanged.connect(self.Invalidate)
+		self.red = False
+		self.error = ""
+		self.validated = True
+
+		self.last_id = 0
+		self.first_time = 0
+		self.last_time = 2 ** 64
+		if self.table_name == "<timeranges>":
+			query = QSqlQuery(self.glb.db)
+			QueryExec(query, "SELECT id, time FROM samples ORDER BY id DESC LIMIT 1")
+			if query.next():
+				self.last_id = int(query.value(0))
+				self.last_time = int(query.value(1))
+			QueryExec(query, "SELECT time FROM samples WHERE time != 0 ORDER BY id LIMIT 1")
+			if query.next():
+				self.first_time = int(query.value(0))
+			if placeholder_text:
+				placeholder_text += ", between " + str(self.first_time) + " and " + str(self.last_time)
+
+		if placeholder_text:
+			self.widget.setPlaceholderText(placeholder_text)
+
+	def ValueToIds(self, value):
+		ids = []
+		query = QSqlQuery(self.glb.db)
+		stmt = "SELECT id FROM " + self.table_name + " WHERE " + self.match_column + " = '" + value + "'"
+		ret = query.exec_(stmt)
+		if ret:
+			while query.next():
+				ids.append(str(query.value(0)))
+		return ids
+
+	def IdBetween(self, query, lower_id, higher_id, order):
+		QueryExec(query, "SELECT id FROM samples WHERE id > " + str(lower_id) + " AND id < " + str(higher_id) + " ORDER BY id " + order + " LIMIT 1")
+		if query.next():
+			return True, int(query.value(0))
+		else:
+			return False, 0
+
+	def BinarySearchTime(self, lower_id, higher_id, target_time, get_floor):
+		query = QSqlQuery(self.glb.db)
+		while True:
+			next_id = int((lower_id + higher_id) / 2)
+			QueryExec(query, "SELECT time FROM samples WHERE id = " + str(next_id))
+			if not query.next():
+				ok, dbid = self.IdBetween(query, lower_id, next_id, "DESC")
+				if not ok:
+					ok, dbid = self.IdBetween(query, next_id, higher_id, "")
+					if not ok:
+						return str(higher_id)
+				next_id = dbid
+				QueryExec(query, "SELECT time FROM samples WHERE id = " + str(next_id))
+			next_time = int(query.value(0))
+			if get_floor:
+				if target_time > next_time:
+					lower_id = next_id
+				else:
+					higher_id = next_id
+				if higher_id <= lower_id + 1:
+					return str(higher_id)
+			else:
+				if target_time >= next_time:
+					lower_id = next_id
+				else:
+					higher_id = next_id
+				if higher_id <= lower_id + 1:
+					return str(lower_id)
+
+	def ConvertRelativeTime(self, val):
+		print "val ", val
+		mult = 1
+		suffix = val[-2:]
+		if suffix == "ms":
+			mult = 1000000
+		elif suffix == "us":
+			mult = 1000
+		elif suffix == "ns":
+			mult = 1
+		else:
+			return val
+		val = val[:-2].strip()
+		if not self.IsNumber(val):
+			return val
+		val = int(val) * mult
+		if val >= 0:
+			val += self.first_time
+		else:
+			val += self.last_time
+		return str(val)
+
+	def ConvertTimeRange(self, vrange):
+		print "vrange ", vrange
+		if vrange[0] == "":
+			vrange[0] = str(self.first_time)
+		if vrange[1] == "":
+			vrange[1] = str(self.last_time)
+		vrange[0] = self.ConvertRelativeTime(vrange[0])
+		vrange[1] = self.ConvertRelativeTime(vrange[1])
+		print "vrange2 ", vrange
+		if not self.IsNumber(vrange[0]) or not self.IsNumber(vrange[1]):
+			return False
+		print "ok1"
+		beg_range = max(int(vrange[0]), self.first_time)
+		end_range = min(int(vrange[1]), self.last_time)
+		if beg_range > self.last_time or end_range < self.first_time:
+			return False
+		print "ok2"
+		vrange[0] = self.BinarySearchTime(0, self.last_id, beg_range, True)
+		vrange[1] = self.BinarySearchTime(1, self.last_id + 1, end_range, False)
+		print "vrange3 ", vrange
+		return True
+
+	def AddTimeRange(self, value, ranges):
+		print "value ", value
+		n = value.count("-")
+		if n == 1:
+			pass
+		elif n == 2:
+			if value.split("-")[1].strip() == "":
+				n = 1
+		elif n == 3:
+			n = 2
+		else:
+			return False
+		pos = findnth(value, "-", n)
+		vrange = [value[:pos].strip() ,value[pos+1:].strip()]
+		if self.ConvertTimeRange(vrange):
+			ranges.append(vrange)
+			return True
+		return False
+
+	def InvalidValue(self, value):
+		self.value = ""
+		palette = QPalette()
+		palette.setColor(QPalette.Text,Qt.red)
+		self.widget.setPalette(palette)
+		self.red = True
+		self.error = self.label + " invalid value '" + value + "'"
+		self.parent.ShowMessage(self.error)
+
+	def IsNumber(self, value):
+		try:
+			x = int(value)
+		except:
+			x = 0
+		return str(x) == value
+
+	def Invalidate(self):
+		self.validated = False
+
+	def Validate(self):
+		input_string = self.widget.text()
+		self.validated = True
+		if self.red:
+			palette = QPalette()
+			self.widget.setPalette(palette)
+			self.red = False
+		if not len(input_string.strip()):
+			self.error = ""
+			self.value = ""
+			return
+		if self.table_name == "<timeranges>":
+			ranges = []
+			for value in [x.strip() for x in input_string.split(",")]:
+				if not self.AddTimeRange(value, ranges):
+					return self.InvalidValue(value)
+			ranges = [("(" + self.column_name1 + " >= " + r[0] + " AND " + self.column_name1 + " <= " + r[1] + ")") for r in ranges]
+			self.value = " OR ".join(ranges)
+		elif self.table_name == "<ranges>":
+			singles = []
+			ranges = []
+			for value in [x.strip() for x in input_string.split(",")]:
+				if "-" in value:
+					vrange = value.split("-")
+					if len(vrange) != 2 or not self.IsNumber(vrange[0]) or not self.IsNumber(vrange[1]):
+						return self.InvalidValue(value)
+					ranges.append(vrange)
+				else:
+					if not self.IsNumber(value):
+						return self.InvalidValue(value)
+					singles.append(value)
+			ranges = [("(" + self.column_name1 + " >= " + r[0] + " AND " + self.column_name1 + " <= " + r[1] + ")") for r in ranges]
+			if len(singles):
+				ranges.append(self.column_name1 + " IN (" + ",".join(singles) + ")")
+			self.value = " OR ".join(ranges)
+		elif self.table_name:
+			all_ids = []
+			for value in [x.strip() for x in input_string.split(",")]:
+				ids = self.ValueToIds(value)
+				if len(ids):
+					all_ids.extend(ids)
+				else:
+					return self.InvalidValue(value)
+			self.value = self.column_name1 + " IN (" + ",".join(all_ids) + ")"
+			if self.column_name2:
+				self.value = "( " + self.value + " OR " + self.column_name2 + " IN (" + ",".join(all_ids) + ") )"
+		else:
+			self.value = input_string.strip()
+		self.error = ""
+		self.parent.ClearMessage()
+
+	def IsValid(self):
+		if not self.validated:
+			self.Validate()
+		if len(self.error):
+			self.parent.ShowMessage(self.error)
+			return False
+		return True
+
+# Selected branch report creation dialog
+
+class SelectedBranchDialog(QDialog):
+
+	def __init__(self, glb, parent=None):
+		super(SelectedBranchDialog, self).__init__(parent)
+
+		self.glb = glb
+
+		self.name = ""
+		self.where_clause = ""
+
+		self.setWindowTitle("Selected Branches")
+		self.setMinimumWidth(600)
+
+		items = (
+			("Report name:", "Enter a name to appear in the window title bar", "", "", "", ""),
+			("Time ranges:", "Enter time ranges", "<timeranges>", "", "samples.id", ""),
+			("CPUs:", "Enter CPUs or ranges e.g. 0,5-6", "<ranges>", "", "cpu", ""),
+			("Commands:", "Only branches with these commands will be included", "comms", "comm", "comm_id", ""),
+			("PIDs:", "Only branches with these process IDs will be included", "threads", "pid", "thread_id", ""),
+			("TIDs:", "Only branches with these thread IDs will be included", "threads", "tid", "thread_id", ""),
+			("DSOs:", "Only branches with these DSOs will be included", "dsos", "short_name", "samples.dso_id", "to_dso_id"),
+			("Symbols:", "Only branches with these symbols will be included", "symbols", "name", "symbol_id", "to_symbol_id"),
+			("Raw SQL clause: ", "Enter a raw SQL WHERE clause", "", "", "", ""),
+			)
+		self.data_items = [SQLTableDialogDataItem(glb, *x, parent=self) for x in items]
+
+		self.grid = QGridLayout()
+
+		for row in xrange(len(self.data_items)):
+			self.grid.addWidget(QLabel(self.data_items[row].label), row, 0)
+			self.grid.addWidget(self.data_items[row].widget, row, 1)
+
+		self.status = QLabel()
+
+		self.ok_button = QPushButton("Ok", self)
+		self.ok_button.setDefault(True)
+		self.ok_button.released.connect(self.Ok)
+		self.ok_button.setSizePolicy(QSizePolicy.Fixed, QSizePolicy.Fixed)
+
+		self.cancel_button = QPushButton("Cancel", self)
+		self.cancel_button.released.connect(self.reject)
+		self.cancel_button.setSizePolicy(QSizePolicy.Fixed, QSizePolicy.Fixed)
+
+		self.hbox = QHBoxLayout()
+		#self.hbox.addStretch()
+		self.hbox.addWidget(self.status)
+		self.hbox.addWidget(self.ok_button)
+		self.hbox.addWidget(self.cancel_button)
+
+		self.vbox = QVBoxLayout()
+		self.vbox.addLayout(self.grid)
+		self.vbox.addLayout(self.hbox)
+
+		self.setLayout(self.vbox);
+
+	def Ok(self):
+		self.name = self.data_items[0].value
+		if not self.name:
+			self.ShowMessage("Report name is required")
+			return
+		for d in self.data_items:
+			if not d.IsValid():
+				return
+		for d in self.data_items[1:]:
+			if len(d.value):
+				if len(self.where_clause):
+					self.where_clause += " AND "
+				self.where_clause += d.value
+		if len(self.where_clause):
+			self.where_clause = " AND ( " + self.where_clause + " ) "
+		else:
+			self.ShowMessage("No selection")
+			return
+		self.accept()
+
+	def ShowMessage(self, msg):
+		self.status.setText("<font color=#FF0000>" + msg)
+
+	def ClearMessage(self):
+		self.status.setText("")
+
 # Event list
 
 def GetEventList(db):
@@ -1656,7 +1975,7 @@ class TableWindow(QMdiSubWindow, ResizeColumnsToContentsBase):
 	def FindDone(self, row):
 		self.find_bar.Idle()
 		if row >= 0:
-			self.view.setCurrentIndex(self.model.index(row, 0, QModelIndex()))
+			self.view.setCurrentIndex(self.model.mapFromSource(self.data_model.index(row, 0, QModelIndex())))
 		else:
 			self.find_bar.NotFound()
 
@@ -1765,6 +2084,149 @@ class WindowMenu():
 	def setActiveSubWindow(self, nr):
 		self.mdi_area.setActiveSubWindow(self.mdi_area.subWindowList()[nr - 1])
 
+# Help text
+
+glb_help_text = """
+<h1>Contents</h1>
+<style>
+p.c1 {
+    text-indent: 40px;
+}
+p.c2 {
+    text-indent: 80px;
+}
+}
+</style>
+<p class=c1><a href=#reports>1. Reports</a></p>
+<p class=c2><a href=#callgraph>1.1 Context-Sensitive Call Graph</a></p>
+<p class=c2><a href=#allbranches>1.2 All branches</a></p>
+<p class=c2><a href=#selectedbranches>1.3 Selected branches</a></p>
+<p class=c1><a href=#tables>2. Tables</a></p>
+<h1 id=reports>1. Reports</h1>
+<h2 id=callgraph>1.1 Context-Sensitive Call Graph</h2>
+The result is a GUI window with a tree representing a context-sensitive
+call-graph. Expanding a couple of levels of the tree and adjusting column
+widths to suit will display something like:
+<pre>
+                                         Call Graph: pt_example
+Call Path                          Object      Count   Time(ns)  Time(%)  Branch Count   Branch Count(%)
+v- ls
+    v- 2638:2638
+        v- _start                  ld-2.19.so    1     10074071   100.0         211135            100.0
+          |- unknown               unknown       1        13198     0.1              1              0.0
+          >- _dl_start             ld-2.19.so    1      1400980    13.9          19637              9.3
+          >- _d_linit_internal     ld-2.19.so    1       448152     4.4          11094              5.3
+          v-__libc_start_main@plt  ls            1      8211741    81.5         180397             85.4
+             >- _dl_fixup          ld-2.19.so    1         7607     0.1            108              0.1
+             >- __cxa_atexit       libc-2.19.so  1        11737     0.1             10              0.0
+             >- __libc_csu_init    ls            1        10354     0.1             10              0.0
+             |- _setjmp            libc-2.19.so  1            0     0.0              4              0.0
+             v- main               ls            1      8182043    99.6         180254             99.9
+</pre>
+<h3>Points to note:</h3>
+<ul>
+<li>The top level is a command name (comm)</li>
+<li>The next level is a thread (pid:tid)</li>
+<li>Subsequent levels are functions</li>
+<li>'Count' is the number of calls</li>
+<li>'Time' is the elapsed time until the function returns</li>
+<li>Percentages are relative to the level above</li>
+<li>'Branch Count' is the total number of branches for that function and all functions that it calls
+</ul>
+<h3>Find</h3>
+Ctrl-F displays a Find bar which finds function names by either an exact match or a pattern match.
+The pattern matching symbols are ? for any character and * for zero or more characters.
+<h2 id=allbranches>1.2 All branches</h2>
+The All branches report displays all branches in chronological order.
+Not all data is fetched immediately. More records can be fetched using the Fetch bar provided.
+<h3>Disassembly</h3>
+Open a branch to display disassembly. This only works if:
+<ol>
+<li>The disassembler is available. Currently, only Intel XED is supported - see <a href=#xed>Intel XED Setup</a></li>
+<li>The object code is available. Currently, only the perf build ID cache is searched for object code.
+The default directory ~/.debug can be overridden by setting environment variable PERF_BUILDID_DIR.
+One exception is kcore where the DSO long name is used (refer dsos_view on the Tables menu),
+or alternatively, set environment variable PERF_KCORE to the kcore file name.</li>
+</ol>
+<h4 id=xed>Intel XED Setup</h4>
+To use Intel XED, libxed.so must be present.  To build and install libxed.so:
+<pre>
+git clone https://github.com/intelxed/mbuild.git mbuild
+git clone https://github.com/intelxed/xed
+cd xed
+./mfile.py --share
+sudo ./mfile.py --prefix=/usr/local install
+sudo ldconfig
+</pre>
+<h3>Find</h3>
+Ctrl-F displays a Find bar which finds substrings by either an exact match or a regular expression match.
+Refer to Python documentation for the regular expression syntax.
+All columns are searched, but only currently fetched rows are searched.
+<h2 id=selectedbranches>1.3 Selected branches</h2>
+This is the same as the <a href=#allbranches>All branches</a> report but with the data reduced
+by various selection criteria. A dialog box displays available criteria which are AND'ed together.
+<h3>1.3.1 Time ranges</h3>
+The time ranges hint text shows the total time range. Relative time ranges can also be entered in
+ms, us or ns. Also, negative values are relative to the end of trace.  Examples:
+<pre>
+	81073085947329-81073085958238	From 81073085947329 to 81073085958238
+	100us-200us		From 100us to 200us
+	10ms-			From 10ms to the end
+	-100ns			The first 100ns
+	-10ms-			The last 10ms
+</pre>
+N.B. Due to the granularity of timestamps, there could be no branches in any given time range.
+<h1 id=tables>2. Tables</h1>
+The Tables menu shows all tables and views in the database. Most tables have an associated view
+which displays the information in a more friendly way. Not all data for large tables is fetched
+immediately. More records can be fetched using the Fetch bar provided. Columns can be sorted,
+but that can be slow for large tables.
+<p>There are also tables of database meta-information.
+For SQLite3 databases, the sqlite_master table is included.
+For PostgreSQL databases, information_schema.tables/views/columns are included.
+<h3>Find</h3>
+Ctrl-F displays a Find bar which finds substrings by either an exact match or a regular expression match.
+Refer to Python documentation for the regular expression syntax.
+All columns are searched, but only currently fetched rows are searched.
+<p>N.B. Results are found in id order, so if the table is re-ordered, find-next and find-previous
+will go to the next/previous result in id order, instead of display order.
+"""
+
+# Help window
+
+class HelpWindow(QMdiSubWindow):
+
+	def __init__(self, glb, parent=None):
+		super(HelpWindow, self).__init__(parent)
+
+		self.text = QTextBrowser()
+		self.text.setHtml(glb_help_text)
+		self.text.setReadOnly(True)
+		self.text.setOpenExternalLinks(True)
+
+		self.setWidget(self.text)
+
+		AddSubWindow(glb.mainwindow.mdi_area, self, "Exported SQL Viewer Help")
+
+# Main window that only displays the help text
+
+class HelpOnlyWindow(QMainWindow):
+
+	def __init__(self, parent=None):
+		super(HelpOnlyWindow, self).__init__(parent)
+
+		self.setMinimumSize(200, 100)
+		self.resize(800, 600)
+		self.setWindowTitle("Exported SQL Viewer Help")
+		self.setWindowIcon(self.style().standardIcon(QStyle.SP_MessageBoxInformation))
+
+		self.text = QTextBrowser()
+		self.text.setHtml(glb_help_text)
+		self.text.setReadOnly(True)
+		self.text.setOpenExternalLinks(True)
+
+		self.setCentralWidget(self.text)
+
 # Font resize
 
 def ResizeFont(widget, diff):
@@ -1851,6 +2313,9 @@ class MainWindow(QMainWindow):
 
 		self.window_menu = WindowMenu(self.mdi_area, menu)
 
+		help_menu = menu.addMenu("&Help")
+		help_menu.addAction(CreateAction("&Exported SQL Viewer Help", "Helpful information", self.Help, self, QKeySequence.HelpContents))
+
 	def Find(self):
 		win = self.mdi_area.activeSubWindow()
 		if win:
@@ -1888,6 +2353,8 @@ class MainWindow(QMainWindow):
 			if event == "branches":
 				label = "All branches" if branches_events == 1 else "All branches " + "(id=" + dbid + ")"
 				reports_menu.addAction(CreateAction(label, "Create a new window displaying branch events", lambda x=dbid: self.NewBranchView(x), self))
+				label = "Selected branches" if branches_events == 1 else "Selected branches " + "(id=" + dbid + ")"
+				reports_menu.addAction(CreateAction(label, "Create a new window displaying branch events", lambda x=dbid: self.NewSelectedBranchView(x), self))
 
 	def TableMenu(self, tables, menu):
 		table_menu = menu.addMenu("&Tables")
@@ -1900,9 +2367,18 @@ class MainWindow(QMainWindow):
 	def NewBranchView(self, event_id):
 		BranchWindow(self.glb, event_id, "", "", self)
 
+	def NewSelectedBranchView(self, event_id):
+		dialog = SelectedBranchDialog(self.glb, self)
+		ret = dialog.exec_()
+		if ret:
+			BranchWindow(self.glb, event_id, dialog.name, dialog.where_clause, self)
+
 	def NewTableView(self, table_name):
 		TableWindow(self.glb, table_name, self)
 
+	def Help(self):
+		HelpWindow(self.glb, self)
+
 # XED Disassembler
 
 class xed_state_t(Structure):
@@ -1929,7 +2405,12 @@ class XEDInstruction():
 class LibXED():
 
 	def __init__(self):
+		try:
 			self.libxed = CDLL("libxed.so")
+		except:
+			self.libxed = None
+		if not self.libxed:
+			self.libxed = CDLL("/usr/local/lib/libxed.so")
 
 		self.xed_tables_init = self.libxed.xed_tables_init
 		self.xed_tables_init.restype = None
@@ -2097,10 +2578,16 @@ class DBRef():
 
 def Main():
 	if (len(sys.argv) < 2):
-		print >> sys.stderr, "Usage is: exported-sql-viewer.py <database name>"
+		print >> sys.stderr, "Usage is: exported-sql-viewer.py {<database name> | --help-only}"
 		raise Exception("Too few arguments")
 
 	dbname = sys.argv[1]
+	if dbname == "--help-only":
+		app = QApplication(sys.argv)
+		mainwindow = HelpOnlyWindow()
+		mainwindow.show()
+		err = app.exec_()
+		sys.exit(err)
 
 	is_sqlite3 = False
 	try:

tools/perf/tests/attr/test-record-group-sampling

@@ -37,4 +37,3 @@ sample_freq=0
 sample_period=0
 freq=0
 write_backward=0
-sample_id_all=0

tools/perf/util/evlist.c

@@ -1810,3 +1810,30 @@ void perf_evlist__force_leader(struct perf_evlist *evlist)
 		leader->forced_leader = true;
 	}
 }
+
+struct perf_evsel *perf_evlist__reset_weak_group(struct perf_evlist *evsel_list,
+						 struct perf_evsel *evsel)
+{
+	struct perf_evsel *c2, *leader;
+	bool is_open = true;
+
+	leader = evsel->leader;
+	pr_debug("Weak group for %s/%d failed\n",
+			leader->name, leader->nr_members);
+
+	/*
+	 * for_each_group_member doesn't work here because it doesn't
+	 * include the first entry.
+	 */
+	evlist__for_each_entry(evsel_list, c2) {
+		if (c2 == evsel)
+			is_open = false;
+		if (c2->leader == leader) {
+			if (is_open)
+				perf_evsel__close(c2);
+			c2->leader = c2;
+			c2->nr_members = 0;
+		}
+	}
+	return leader;
+}

tools/perf/util/evlist.h

@@ -312,4 +312,7 @@ bool perf_evlist__exclude_kernel(struct perf_evlist *evlist);
 
 void perf_evlist__force_leader(struct perf_evlist *evlist);
 
+struct perf_evsel *perf_evlist__reset_weak_group(struct perf_evlist *evlist,
+						 struct perf_evsel *evsel);
+
 #endif /* __PERF_EVLIST_H */

tools/perf/util/evsel.c

@@ -956,7 +956,6 @@ void perf_evsel__config(struct perf_evsel *evsel, struct record_opts *opts,
 		attr->sample_freq    = 0;
 		attr->sample_period  = 0;
 		attr->write_backward = 0;
-		attr->sample_id_all  = 0;
 	}
 
 	if (opts->no_samples)

tools/perf/util/intel-pt-decoder/intel-pt-decoder.c

@@ -1474,6 +1474,8 @@ static void intel_pt_calc_mtc_timestamp(struct intel_pt_decoder *decoder)
 		decoder->have_calc_cyc_to_tsc = false;
 		intel_pt_calc_cyc_to_tsc(decoder, true);
 	}
+
+	intel_pt_log_to("Setting timestamp", decoder->timestamp);
 }
 
 static void intel_pt_calc_cbr(struct intel_pt_decoder *decoder)
@@ -1514,6 +1516,8 @@ static void intel_pt_calc_cyc_timestamp(struct intel_pt_decoder *decoder)
 		decoder->timestamp = timestamp;
 
 	decoder->timestamp_insn_cnt = 0;
+
+	intel_pt_log_to("Setting timestamp", decoder->timestamp);
 }
 
 /* Walk PSB+ packets when already in sync. */

tools/perf/util/intel-pt-decoder/intel-pt-log.c

@@ -31,6 +31,11 @@ static FILE *f;
 static char log_name[MAX_LOG_NAME];
 bool intel_pt_enable_logging;
 
+void *intel_pt_log_fp(void)
+{
+	return f;
+}
+
 void intel_pt_log_enable(void)
 {
 	intel_pt_enable_logging = true;

tools/perf/util/intel-pt-decoder/intel-pt-log.h

@@ -22,6 +22,7 @@
 
 struct intel_pt_pkt;
 
+void *intel_pt_log_fp(void);
 void intel_pt_log_enable(void);
 void intel_pt_log_disable(void);
 void intel_pt_log_set_name(const char *name);

tools/perf/util/intel-pt.c

@@ -206,6 +206,16 @@ static void intel_pt_dump_event(struct intel_pt *pt, unsigned char *buf,
 	intel_pt_dump(pt, buf, len);
 }
 
+static void intel_pt_log_event(union perf_event *event)
+{
+	FILE *f = intel_pt_log_fp();
+
+	if (!intel_pt_enable_logging || !f)
+		return;
+
+	perf_event__fprintf(event, f);
+}
+
 static int intel_pt_do_fix_overlap(struct intel_pt *pt, struct auxtrace_buffer *a,
 				   struct auxtrace_buffer *b)
 {
@@ -2010,9 +2020,9 @@ static int intel_pt_process_event(struct perf_session *session,
 		 event->header.type == PERF_RECORD_SWITCH_CPU_WIDE)
 		err = intel_pt_context_switch(pt, event, sample);
 
-	intel_pt_log("event %s (%u): cpu %d time %"PRIu64" tsc %#"PRIx64"\n",
-		     perf_event__name(event->header.type), event->header.type,
-		     sample->cpu, sample->time, timestamp);
+	intel_pt_log("event %u: cpu %d time %"PRIu64" tsc %#"PRIx64" ",
+		     event->header.type, sample->cpu, sample->time, timestamp);
+	intel_pt_log_event(event);
 
 	return err;
 }

tools/perf/util/pmu.c

@@ -773,7 +773,7 @@ static void pmu_add_cpu_aliases(struct list_head *head, struct perf_pmu *pmu)
 
 		if (!is_arm_pmu_core(name)) {
 			pname = pe->pmu ? pe->pmu : "cpu";
-			if (strncmp(pname, name, strlen(pname)))
+			if (strcmp(pname, name))
 				continue;
 		}