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

Pull perf tooling updates form Ingo Molnar:
 "A final batch of perf tooling changes: mostly fixes and small
  improvements"

* 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (29 commits)
  perf session: Add comment for perf_session__register_idle_thread()
  perf thread-stack: Fix thread stack processing for the idle task
  perf thread-stack: Allocate an array of thread stacks
  perf thread-stack: Factor out thread_stack__init()
  perf thread-stack: Allow for a thread stack array
  perf thread-stack: Avoid direct reference to the thread's stack
  perf thread-stack: Tidy thread_stack__bottom() usage
  perf thread-stack: Simplify some code in thread_stack__process()
  tools gpio: Allow overriding CFLAGS
  tools power turbostat: Override CFLAGS assignments and add LDFLAGS to build command
  tools thermal tmon: Allow overriding CFLAGS assignments
  tools power x86_energy_perf_policy: Override CFLAGS assignments and add LDFLAGS to build command
  perf c2c: Increase the HITM ratio limit for displayed cachelines
  perf c2c: Change the default coalesce setup
  perf trace beauty ioctl: Beautify USBDEVFS_ commands
  perf trace beauty: Export function to get the files for a thread
  perf trace: Wire up ioctl's USBDEBFS_ cmd table generator
  perf beauty ioctl: Add generator for USBDEVFS_ ioctl commands
  tools headers uapi: Grab a copy of usbdevice_fs.h
  perf trace: Store the major number for a file when storing its pathname
  ...

20 files changed, 400 insertions, 147 deletions

diff --git a/tools/perf/Makefile.config b/tools/perf/Makefile.config
index 07c1857c3d7a..b441c88cafa1 100644
--- a/tools/perf/Makefile.config
+++ b/tools/perf/Makefile.config
@@ -702,18 +702,20 @@ endif
 
 ifeq ($(feature-libbfd), 1)
   EXTLIBS += -lbfd
+else
+  # we are on a system that requires -liberty and (maybe) -lz
+  # to link against -lbfd; test each case individually here
 
   # call all detections now so we get correct
   # status in VF output
-  $(call feature_check,liberty)
-  $(call feature_check,liberty-z)
-  $(call feature_check,cplus-demangle)
+  $(call feature_check,libbfd-liberty)
+  $(call feature_check,libbfd-liberty-z)
 
-  ifeq ($(feature-liberty), 1)
-    EXTLIBS += -liberty
+  ifeq ($(feature-libbfd-liberty), 1)
+    EXTLIBS += -lbfd -liberty
   else
-    ifeq ($(feature-liberty-z), 1)
-      EXTLIBS += -liberty -lz
+    ifeq ($(feature-libbfd-liberty-z), 1)
+      EXTLIBS += -lbfd -liberty -lz
     endif
   endif
 endif
@@ -723,24 +725,24 @@ ifdef NO_DEMANGLE
 else
   ifdef HAVE_CPLUS_DEMANGLE_SUPPORT
     EXTLIBS += -liberty
-    CFLAGS += -DHAVE_CPLUS_DEMANGLE_SUPPORT
   else
-    ifneq ($(feature-libbfd), 1)
-      ifneq ($(feature-liberty), 1)
-        ifneq ($(feature-liberty-z), 1)
-          # we dont have neither HAVE_CPLUS_DEMANGLE_SUPPORT
-          # or any of 'bfd iberty z' trinity
-          ifeq ($(feature-cplus-demangle), 1)
-            EXTLIBS += -liberty
-            CFLAGS += -DHAVE_CPLUS_DEMANGLE_SUPPORT
-          else
-            msg := $(warning No bfd.h/libbfd found, please install binutils-dev[el]/zlib-static/libiberty-dev to gain symbol demangling)
-            CFLAGS += -DNO_DEMANGLE
-          endif
-        endif
+    ifeq ($(filter -liberty,$(EXTLIBS)),)
+      $(call feature_check,cplus-demangle)
+
+      # we dont have neither HAVE_CPLUS_DEMANGLE_SUPPORT
+      # or any of 'bfd iberty z' trinity
+      ifeq ($(feature-cplus-demangle), 1)
+        EXTLIBS += -liberty
+      else
+        msg := $(warning No bfd.h/libbfd found, please install binutils-dev[el]/zlib-static/libiberty-dev to gain symbol demangling)
+        CFLAGS += -DNO_DEMANGLE
       endif
     endif
   endif
+
+  ifneq ($(filter -liberty,$(EXTLIBS)),)
+    CFLAGS += -DHAVE_CPLUS_DEMANGLE_SUPPORT
+  endif
 endif
 
 ifneq ($(filter -lbfd,$(EXTLIBS)),)
diff --git a/tools/perf/Makefile.perf b/tools/perf/Makefile.perf
index bd23e3f30895..ff29c3372ec3 100644
--- a/tools/perf/Makefile.perf
+++ b/tools/perf/Makefile.perf
@@ -497,6 +497,12 @@ prctl_option_tbl := $(srctree)/tools/perf/trace/beauty/prctl_option.sh
 $(prctl_option_array): $(prctl_hdr_dir)/prctl.h $(prctl_option_tbl)
 	$(Q)$(SHELL) '$(prctl_option_tbl)' $(prctl_hdr_dir) > $@
 
+usbdevfs_ioctl_array := $(beauty_ioctl_outdir)/usbdevfs_ioctl_array.c
+usbdevfs_ioctl_tbl := $(srctree)/tools/perf/trace/beauty/usbdevfs_ioctl.sh
+
+$(usbdevfs_ioctl_array): $(linux_uapi_dir)/usbdevice_fs.h $(usbdevfs_ioctl_tbl)
+	$(Q)$(SHELL) '$(usbdevfs_ioctl_tbl)' $(linux_uapi_dir) > $@
+
 x86_arch_prctl_code_array := $(beauty_outdir)/x86_arch_prctl_code_array.c
 x86_arch_prctl_code_tbl := $(srctree)/tools/perf/trace/beauty/x86_arch_prctl.sh
 
@@ -624,6 +630,7 @@ prepare: $(OUTPUT)PERF-VERSION-FILE $(OUTPUT)common-cmds.h archheaders $(drm_ioc
 	$(mount_flags_array) \
 	$(perf_ioctl_array) \
 	$(prctl_option_array) \
+	$(usbdevfs_ioctl_array) \
 	$(x86_arch_prctl_code_array) \
 	$(rename_flags_array) \
 	$(arch_errno_name_array)
@@ -923,6 +930,7 @@ clean:: $(LIBTRACEEVENT)-clean $(LIBAPI)-clean $(LIBBPF)-clean $(LIBSUBCMD)-clea
 		$(OUTPUT)$(vhost_virtio_ioctl_array) \
 		$(OUTPUT)$(perf_ioctl_array) \
 		$(OUTPUT)$(prctl_option_array) \
+		$(OUTPUT)$(usbdevfs_ioctl_array) \
 		$(OUTPUT)$(x86_arch_prctl_code_array) \
 		$(OUTPUT)$(rename_flags_array) \
 		$(OUTPUT)$(arch_errno_name_array)
diff --git a/tools/perf/builtin-c2c.c b/tools/perf/builtin-c2c.c
index f3aa9d02a5ab..d340d2e42776 100644
--- a/tools/perf/builtin-c2c.c
+++ b/tools/perf/builtin-c2c.c
@@ -68,7 +68,7 @@ struct c2c_hist_entry {
 	struct hist_entry	he;
 };
 
-static char const *coalesce_default = "pid,iaddr";
+static char const *coalesce_default = "iaddr";
 
 struct perf_c2c {
 	struct perf_tool	tool;
@@ -1878,7 +1878,7 @@ static int c2c_hists__reinit(struct c2c_hists *c2c_hists,
 	return hpp_list__parse(&c2c_hists->list, output, sort);
 }
 
-#define DISPLAY_LINE_LIMIT  0.0005
+#define DISPLAY_LINE_LIMIT  0.001
 
 static bool he__display(struct hist_entry *he, struct c2c_stats *stats)
 {
diff --git a/tools/perf/builtin-script.c b/tools/perf/builtin-script.c
index 3728b50e52e2..d079f36d342d 100644
--- a/tools/perf/builtin-script.c
+++ b/tools/perf/builtin-script.c
@@ -1073,9 +1073,18 @@ static int perf_sample__fprintf_brstackinsn(struct perf_sample *sample,
 
 	/*
 	 * Print final block upto sample
+	 *
+	 * Due to pipeline delays the LBRs might be missing a branch
+	 * or two, which can result in very large or negative blocks
+	 * between final branch and sample. When this happens just
+	 * continue walking after the last TO until we hit a branch.
 	 */
 	start = br->entries[0].to;
 	end = sample->ip;
+	if (end < start) {
+		/* Missing jump. Scan 128 bytes for the next branch */
+		end = start + 128;
+	}
 	len = grab_bb(buffer, start, end, machine, thread, &x.is64bit, &x.cpumode, true);
 	printed += ip__fprintf_sym(start, thread, x.cpumode, x.cpu, &lastsym, attr, fp);
 	if (len <= 0) {
@@ -1084,7 +1093,6 @@ static int perf_sample__fprintf_brstackinsn(struct perf_sample *sample,
 			      machine, thread, &x.is64bit, &x.cpumode, false);
 		if (len <= 0)
 			goto out;
-
 		printed += fprintf(fp, "\t%016" PRIx64 "\t%s\n", sample->ip,
 			dump_insn(&x, sample->ip, buffer, len, NULL));
 		if (PRINT_FIELD(SRCCODE))
@@ -1096,6 +1104,13 @@ static int perf_sample__fprintf_brstackinsn(struct perf_sample *sample,
 				   dump_insn(&x, start + off, buffer + off, len - off, &ilen));
 		if (ilen == 0)
 			break;
+		if (arch_is_branch(buffer + off, len - off, x.is64bit) && start + off != sample->ip) {
+			/*
+			 * Hit a missing branch. Just stop.
+			 */
+			printed += fprintf(fp, "\t... not reaching sample ...\n");
+			break;
+		}
 		if (PRINT_FIELD(SRCCODE))
 			print_srccode(thread, x.cpumode, start + off);
 	}
@@ -1167,7 +1182,7 @@ static int perf_sample__fprintf_callindent(struct perf_sample *sample,
 					   struct addr_location *al, FILE *fp)
 {
 	struct perf_event_attr *attr = &evsel->attr;
-	size_t depth = thread_stack__depth(thread);
+	size_t depth = thread_stack__depth(thread, sample->cpu);
 	const char *name = NULL;
 	static int spacing;
 	int len = 0;
@@ -1701,7 +1716,7 @@ static bool show_event(struct perf_sample *sample,
 		       struct thread *thread,
 		       struct addr_location *al)
 {
-	int depth = thread_stack__depth(thread);
+	int depth = thread_stack__depth(thread, sample->cpu);
 
 	if (!symbol_conf.graph_function)
 		return true;
diff --git a/tools/perf/builtin-trace.c b/tools/perf/builtin-trace.c
index ebde59e61133..adbf28183560 100644
--- a/tools/perf/builtin-trace.c
+++ b/tools/perf/builtin-trace.c
@@ -60,6 +60,7 @@
 #include <linux/stringify.h>
 #include <linux/time64.h>
 #include <fcntl.h>
+#include <sys/sysmacros.h>
 
 #include "sane_ctype.h"
 
@@ -112,8 +113,9 @@ struct trace {
 	} stats;
 	unsigned int		max_stack;
 	unsigned int		min_stack;
-	bool			sort_events;
+	int			raw_augmented_syscalls_args_size;
 	bool			raw_augmented_syscalls;
+	bool			sort_events;
 	bool			not_ev_qualifier;
 	bool			live;
 	bool			full_time;
@@ -283,12 +285,17 @@ out_delete:
 	return -ENOENT;
 }
 
-static int perf_evsel__init_augmented_syscall_tp(struct perf_evsel *evsel)
+static int perf_evsel__init_augmented_syscall_tp(struct perf_evsel *evsel, struct perf_evsel *tp)
 {
 	struct syscall_tp *sc = evsel->priv = malloc(sizeof(struct syscall_tp));
 
-	if (evsel->priv != NULL) {       /* field, sizeof_field, offsetof_field */
-		if (__tp_field__init_uint(&sc->id, sizeof(long), sizeof(long long), evsel->needs_swap))
+	if (evsel->priv != NULL) {
+		struct tep_format_field *syscall_id = perf_evsel__field(tp, "id");
+		if (syscall_id == NULL)
+			syscall_id = perf_evsel__field(tp, "__syscall_nr");
+		if (syscall_id == NULL)
+			goto out_delete;
+		if (__tp_field__init_uint(&sc->id, syscall_id->size, syscall_id->offset, evsel->needs_swap))
 			goto out_delete;
 
 		return 0;
@@ -974,9 +981,9 @@ struct thread_trace {
 		char	      *name;
 	} filename;
 	struct {
-		int	  max;
-		char	  **table;
-	} paths;
+		int	      max;
+		struct file   *table;
+	} files;
 
 	struct intlist *syscall_stats;
 };
@@ -986,7 +993,7 @@ static struct thread_trace *thread_trace__new(void)
 	struct thread_trace *ttrace =  zalloc(sizeof(struct thread_trace));
 
 	if (ttrace)
-		ttrace->paths.max = -1;
+		ttrace->files.max = -1;
 
 	ttrace->syscall_stats = intlist__new(NULL);
 
@@ -1030,30 +1037,48 @@ void syscall_arg__set_ret_scnprintf(struct syscall_arg *arg,
 
 static const size_t trace__entry_str_size = 2048;
 
-static int trace__set_fd_pathname(struct thread *thread, int fd, const char *pathname)
+static struct file *thread_trace__files_entry(struct thread_trace *ttrace, int fd)
 {
-	struct thread_trace *ttrace = thread__priv(thread);
-
-	if (fd > ttrace->paths.max) {
-		char **npath = realloc(ttrace->paths.table, (fd + 1) * sizeof(char *));
+	if (fd > ttrace->files.max) {
+		struct file *nfiles = realloc(ttrace->files.table, (fd + 1) * sizeof(struct file));
 
-		if (npath == NULL)
-			return -1;
+		if (nfiles == NULL)
+			return NULL;
 
-		if (ttrace->paths.max != -1) {
-			memset(npath + ttrace->paths.max + 1, 0,
-			       (fd - ttrace->paths.max) * sizeof(char *));
+		if (ttrace->files.max != -1) {
+			memset(nfiles + ttrace->files.max + 1, 0,
+			       (fd - ttrace->files.max) * sizeof(struct file));
 		} else {
-			memset(npath, 0, (fd + 1) * sizeof(char *));
+			memset(nfiles, 0, (fd + 1) * sizeof(struct file));
 		}
 
-		ttrace->paths.table = npath;
-		ttrace->paths.max   = fd;
+		ttrace->files.table = nfiles;
+		ttrace->files.max   = fd;
 	}
 
-	ttrace->paths.table[fd] = strdup(pathname);
+	return ttrace->files.table + fd;
+}
 
-	return ttrace->paths.table[fd] != NULL ? 0 : -1;
+struct file *thread__files_entry(struct thread *thread, int fd)
+{
+	return thread_trace__files_entry(thread__priv(thread), fd);
+}
+
+static int trace__set_fd_pathname(struct thread *thread, int fd, const char *pathname)
+{
+	struct thread_trace *ttrace = thread__priv(thread);
+	struct file *file = thread_trace__files_entry(ttrace, fd);
+
+	if (file != NULL) {
+		struct stat st;
+		if (stat(pathname, &st) == 0)
+			file->dev_maj = major(st.st_rdev);
+		file->pathname = strdup(pathname);
+		if (file->pathname)
+			return 0;
+	}
+
+	return -1;
 }
 
 static int thread__read_fd_path(struct thread *thread, int fd)
@@ -1093,7 +1118,7 @@ static const char *thread__fd_path(struct thread *thread, int fd,
 	if (fd < 0)
 		return NULL;
 
-	if ((fd > ttrace->paths.max || ttrace->paths.table[fd] == NULL)) {
+	if ((fd > ttrace->files.max || ttrace->files.table[fd].pathname == NULL)) {
 		if (!trace->live)
 			return NULL;
 		++trace->stats.proc_getname;
@@ -1101,7 +1126,7 @@ static const char *thread__fd_path(struct thread *thread, int fd,
 			return NULL;
 	}
 
-	return ttrace->paths.table[fd];
+	return ttrace->files.table[fd].pathname;
 }
 
 size_t syscall_arg__scnprintf_fd(char *bf, size_t size, struct syscall_arg *arg)
@@ -1140,8 +1165,8 @@ static size_t syscall_arg__scnprintf_close_fd(char *bf, size_t size,
 	size_t printed = syscall_arg__scnprintf_fd(bf, size, arg);
 	struct thread_trace *ttrace = thread__priv(arg->thread);
 
-	if (ttrace && fd >= 0 && fd <= ttrace->paths.max)
-		zfree(&ttrace->paths.table[fd]);
+	if (ttrace && fd >= 0 && fd <= ttrace->files.max)
+		zfree(&ttrace->files.table[fd].pathname);
 
 	return printed;
 }
@@ -1768,16 +1793,16 @@ 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, bool raw_augmented)
+static void *syscall__augmented_args(struct syscall *sc, struct perf_sample *sample, int *augmented_args_size, int raw_augmented_args_size)
 {
 	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.
+	 * and there we get all 6 syscall args plus the tracepoint common fields
+	 * that gets calculated at the start 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
@@ -1785,7 +1810,7 @@ static void *syscall__augmented_args(struct syscall *sc, struct perf_sample *sam
 	 * 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;
+	int args_size = raw_augmented_args_size ?: sc->args_size;
 
 	*augmented_args_size = sample->raw_size - args_size;
 	if (*augmented_args_size > 0)
@@ -1839,7 +1864,7 @@ 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, trace->raw_augmented_syscalls);
+		augmented_args = syscall__augmented_args(sc, sample, &augmented_args_size, trace->raw_augmented_syscalls_args_size);
 	ttrace->entry_time = sample->time;
 	msg = ttrace->entry_str;
 	printed += scnprintf(msg + printed, trace__entry_str_size - printed, "%s(", sc->name);
@@ -1897,7 +1922,7 @@ 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, trace->raw_augmented_syscalls);
+	augmented_args = syscall__augmented_args(sc, sample, &augmented_args_size, trace->raw_augmented_syscalls_args_size);
 	syscall__scnprintf_args(sc, msg, sizeof(msg), args, augmented_args, augmented_args_size, trace, thread);
 	fprintf(trace->output, "%s", msg);
 	err = 0;
@@ -2686,7 +2711,9 @@ static int trace__set_ev_qualifier_filter(struct trace *trace)
 {
 	if (trace->syscalls.map)
 		return trace__set_ev_qualifier_bpf_filter(trace);
-	return trace__set_ev_qualifier_tp_filter(trace);
+	if (trace->syscalls.events.sys_enter)
+		return trace__set_ev_qualifier_tp_filter(trace);
+	return 0;
 }
 
 static int bpf_map__set_filter_pids(struct bpf_map *map __maybe_unused,
@@ -3812,13 +3839,6 @@ int cmd_trace(int argc, const char **argv)
 	 * syscall.
 	 */
 	if (trace.syscalls.events.augmented) {
-		evsel = trace.syscalls.events.augmented;
-
-		if (perf_evsel__init_augmented_syscall_tp(evsel) ||
-		    perf_evsel__init_augmented_syscall_tp_args(evsel))
-			goto out;
-		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;
 
@@ -3827,9 +3847,41 @@ int cmd_trace(int argc, const char **argv)
 				goto init_augmented_syscall_tp;
 			}
 
+			if (strcmp(perf_evsel__name(evsel), "raw_syscalls:sys_enter") == 0) {
+				struct perf_evsel *augmented = trace.syscalls.events.augmented;
+				if (perf_evsel__init_augmented_syscall_tp(augmented, evsel) ||
+				    perf_evsel__init_augmented_syscall_tp_args(augmented))
+					goto out;
+				augmented->handler = trace__sys_enter;
+			}
+
 			if (strstarts(perf_evsel__name(evsel), "syscalls:sys_exit_")) {
+				struct syscall_tp *sc;
 init_augmented_syscall_tp:
-				perf_evsel__init_augmented_syscall_tp(evsel);
+				if (perf_evsel__init_augmented_syscall_tp(evsel, evsel))
+					goto out;
+				sc = evsel->priv;
+				/*
+				 * 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 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.
+				 */
+				if (trace.raw_augmented_syscalls)
+					trace.raw_augmented_syscalls_args_size = (6 + 1) * sizeof(long) + sc->id.offset;
 				perf_evsel__init_augmented_syscall_tp_ret(evsel);
 				evsel->handler = trace__sys_exit;
 			}
diff --git a/tools/perf/check-headers.sh b/tools/perf/check-headers.sh
index 8e811ea0cf85..6cb98f8570a2 100755
--- a/tools/perf/check-headers.sh
+++ b/tools/perf/check-headers.sh
@@ -14,6 +14,7 @@ include/uapi/linux/perf_event.h
 include/uapi/linux/prctl.h
 include/uapi/linux/sched.h
 include/uapi/linux/stat.h
+include/uapi/linux/usbdevice_fs.h
 include/uapi/linux/vhost.h
 include/uapi/sound/asound.h
 include/linux/bits.h
diff --git a/tools/perf/trace/beauty/beauty.h b/tools/perf/trace/beauty/beauty.h
index 83c5b202e00e..139d485a6f16 100644
--- a/tools/perf/trace/beauty/beauty.h
+++ b/tools/perf/trace/beauty/beauty.h
@@ -32,6 +32,13 @@ size_t strarray__scnprintf_flags(struct strarray *sa, char *bf, size_t size, boo
 struct trace;
 struct thread;
 
+struct file {
+	char *pathname;
+	int  dev_maj;
+};
+
+struct file *thread__files_entry(struct thread *thread, int fd);
+
 struct strarrays {
 	int		nr_entries;
 	struct strarray **entries;
diff --git a/tools/perf/trace/beauty/ioctl.c b/tools/perf/trace/beauty/ioctl.c
index 9efeb6a936c2..620350d41209 100644
--- a/tools/perf/trace/beauty/ioctl.c
+++ b/tools/perf/trace/beauty/ioctl.c
@@ -112,6 +112,17 @@ static size_t ioctl__scnprintf_perf_cmd(int nr, int dir, char *bf, size_t size)
 	return scnprintf(bf, size, "(%#x, %#x, %#x)", 0xAE, nr, dir);
 }
 
+static size_t ioctl__scnprintf_usbdevfs_cmd(int nr, int dir, char *bf, size_t size)
+{
+#include "trace/beauty/generated/ioctl/usbdevfs_ioctl_array.c"
+	static DEFINE_STRARRAY(usbdevfs_ioctl_cmds, "");
+
+	if (nr < strarray__usbdevfs_ioctl_cmds.nr_entries && strarray__usbdevfs_ioctl_cmds.entries[nr] != NULL)
+		return scnprintf(bf, size, "USBDEVFS_%s", strarray__usbdevfs_ioctl_cmds.entries[nr]);
+
+	return scnprintf(bf, size, "(%c, %#x, %#x)", 'U', nr, dir);
+}
+
 static size_t ioctl__scnprintf_cmd(unsigned long cmd, char *bf, size_t size, bool show_prefix)
 {
 	const char *prefix = "_IOC_";
@@ -157,9 +168,20 @@ static size_t ioctl__scnprintf_cmd(unsigned long cmd, char *bf, size_t size, boo
 	return printed + scnprintf(bf + printed, size - printed, ", %#x, %#x, %#x)", type, nr, sz);
 }
 
+#ifndef USB_DEVICE_MAJOR
+#define USB_DEVICE_MAJOR 189
+#endif // USB_DEVICE_MAJOR
+
 size_t syscall_arg__scnprintf_ioctl_cmd(char *bf, size_t size, struct syscall_arg *arg)
 {
 	unsigned long cmd = arg->val;
+	unsigned int fd = syscall_arg__val(arg, 0);
+	struct file *file = thread__files_entry(arg->thread, fd);
+
+	if (file != NULL) {
+		if (file->dev_maj == USB_DEVICE_MAJOR)
+			return ioctl__scnprintf_usbdevfs_cmd(_IOC_NR(cmd), _IOC_DIR(cmd), bf, size);
+	}
 
 	return ioctl__scnprintf_cmd(cmd, bf, size, arg->show_string_prefix);
 }
diff --git a/tools/perf/trace/beauty/mmap.c b/tools/perf/trace/beauty/mmap.c
index eb31089790e3..859a8a9db2c6 100644
--- a/tools/perf/trace/beauty/mmap.c
+++ b/tools/perf/trace/beauty/mmap.c
@@ -18,8 +18,8 @@ static size_t syscall_arg__scnprintf_mmap_prot(char *bf, size_t size,
 	}
 
 	P_MMAP_PROT(READ);
-	P_MMAP_PROT(EXEC);
 	P_MMAP_PROT(WRITE);
+	P_MMAP_PROT(EXEC);
 	P_MMAP_PROT(SEM);
 	P_MMAP_PROT(GROWSDOWN);
 	P_MMAP_PROT(GROWSUP);
diff --git a/tools/perf/trace/beauty/seccomp.c b/tools/perf/trace/beauty/seccomp.c
index 4600c28a3cfe..637722e2796b 100644
--- a/tools/perf/trace/beauty/seccomp.c
+++ b/tools/perf/trace/beauty/seccomp.c
@@ -9,7 +9,7 @@
 static size_t syscall_arg__scnprintf_seccomp_op(char *bf, size_t size, struct syscall_arg *arg)
 {
 	bool show_prefix = arg->show_string_prefix;
-	const char *prefix = "SECOMP_SET_MODE_";
+	const char *prefix = "SECCOMP_SET_MODE_";
 	int op = arg->val;
 	size_t printed = 0;
 
@@ -34,7 +34,7 @@ static size_t syscall_arg__scnprintf_seccomp_flags(char *bf, size_t size,
 						   struct syscall_arg *arg)
 {
 	bool show_prefix = arg->show_string_prefix;
-	const char *prefix = "SECOMP_FILTER_FLAG_";
+	const char *prefix = "SECCOMP_FILTER_FLAG_";
 	int printed = 0, flags = arg->val;
 
 #define	P_FLAG(n) \
diff --git a/tools/perf/trace/beauty/usbdevfs_ioctl.sh b/tools/perf/trace/beauty/usbdevfs_ioctl.sh
new file mode 100755
index 000000000000..930b80f422e8
--- /dev/null
+++ b/tools/perf/trace/beauty/usbdevfs_ioctl.sh
@@ -0,0 +1,19 @@
+#!/bin/sh
+# SPDX-License-Identifier: LGPL-2.1
+
+[ $# -eq 1 ] && header_dir=$1 || header_dir=tools/include/uapi/linux/
+
+printf "static const char *usbdevfs_ioctl_cmds[] = {\n"
+regex="^#[[:space:]]*define[[:space:]]+USBDEVFS_(\w+)[[:space:]]+_IO[WR]{0,2}\([[:space:]]*'U'[[:space:]]*,[[:space:]]*([[:digit:]]+).*"
+egrep $regex ${header_dir}/usbdevice_fs.h | egrep -v 'USBDEVFS_\w+32[[:space:]]' | \
+	sed -r "s/$regex/\2 \1/g"	| \
+	sort | xargs printf "\t[%s] = \"%s\",\n"
+printf "};\n\n"
+printf "#if 0\n"
+printf "static const char *usbdevfs_ioctl_32_cmds[] = {\n"
+regex="^#[[:space:]]*define[[:space:]]+USBDEVFS_(\w+)[[:space:]]+_IO[WR]{0,2}\([[:space:]]*'U'[[:space:]]*,[[:space:]]*([[:digit:]]+).*"
+egrep $regex ${header_dir}/usbdevice_fs.h | egrep 'USBDEVFS_\w+32[[:space:]]' | \
+	sed -r "s/$regex/\2 \1/g"	| \
+	sort | xargs printf "\t[%s] = \"%s\",\n"
+printf "};\n"
+printf "#endif\n"
diff --git a/tools/perf/util/dump-insn.c b/tools/perf/util/dump-insn.c
index 10988d3de7ce..2bd8585db93c 100644
--- a/tools/perf/util/dump-insn.c
+++ b/tools/perf/util/dump-insn.c
@@ -13,3 +13,11 @@ const char *dump_insn(struct perf_insn *x __maybe_unused,
 		*lenp = 0;
 	return "?";
 }
+
+__weak
+int arch_is_branch(const unsigned char *buf __maybe_unused,
+		   size_t len __maybe_unused,
+		   int x86_64 __maybe_unused)
+{
+	return 0;
+}
diff --git a/tools/perf/util/dump-insn.h b/tools/perf/util/dump-insn.h
index 0e06280a8860..650125061530 100644
--- a/tools/perf/util/dump-insn.h
+++ b/tools/perf/util/dump-insn.h
@@ -20,4 +20,6 @@ struct perf_insn {
 
 const char *dump_insn(struct perf_insn *x, u64 ip,
 		      u8 *inbuf, int inlen, int *lenp);
+int arch_is_branch(const unsigned char *buf, size_t len, int x86_64);
+
 #endif
diff --git a/tools/perf/util/intel-bts.c b/tools/perf/util/intel-bts.c
index 7b27d77306c2..ee6ca65f81f4 100644
--- a/tools/perf/util/intel-bts.c
+++ b/tools/perf/util/intel-bts.c
@@ -451,7 +451,7 @@ static int intel_bts_process_buffer(struct intel_bts_queue *btsq,
 			continue;
 		intel_bts_get_branch_type(btsq, branch);
 		if (btsq->bts->synth_opts.thread_stack)
-			thread_stack__event(thread, btsq->sample_flags,
+			thread_stack__event(thread, btsq->cpu, btsq->sample_flags,
 					    le64_to_cpu(branch->from),
 					    le64_to_cpu(branch->to),
 					    btsq->intel_pt_insn.length,
@@ -523,7 +523,7 @@ static int intel_bts_process_queue(struct intel_bts_queue *btsq, u64 *timestamp)
 	    !btsq->bts->synth_opts.thread_stack && thread &&
 	    (!old_buffer || btsq->bts->sampling_mode ||
 	     (btsq->bts->snapshot_mode && !buffer->consecutive)))
-		thread_stack__set_trace_nr(thread, buffer->buffer_nr + 1);
+		thread_stack__set_trace_nr(thread, btsq->cpu, buffer->buffer_nr + 1);
 
 	err = intel_bts_process_buffer(btsq, buffer, thread);
 
diff --git a/tools/perf/util/intel-pt-decoder/intel-pt-insn-decoder.c b/tools/perf/util/intel-pt-decoder/intel-pt-insn-decoder.c
index 54818828023b..1c0e289f01e6 100644
--- a/tools/perf/util/intel-pt-decoder/intel-pt-insn-decoder.c
+++ b/tools/perf/util/intel-pt-decoder/intel-pt-insn-decoder.c
@@ -180,6 +180,14 @@ int intel_pt_get_insn(const unsigned char *buf, size_t len, int x86_64,
 	return 0;
 }
 
+int arch_is_branch(const unsigned char *buf, size_t len, int x86_64)
+{
+	struct intel_pt_insn in;
+	if (intel_pt_get_insn(buf, len, x86_64, &in) < 0)
+		return -1;
+	return in.branch != INTEL_PT_BR_NO_BRANCH;
+}
+
 const char *dump_insn(struct perf_insn *x, uint64_t ip __maybe_unused,
 		      u8 *inbuf, int inlen, int *lenp)
 {
diff --git a/tools/perf/util/intel-pt.c b/tools/perf/util/intel-pt.c
index 149ff361ca78..2e72373ec6df 100644
--- a/tools/perf/util/intel-pt.c
+++ b/tools/perf/util/intel-pt.c
@@ -1174,7 +1174,7 @@ static void intel_pt_prep_sample(struct intel_pt *pt,
 	intel_pt_prep_b_sample(pt, ptq, event, sample);
 
 	if (pt->synth_opts.callchain) {
-		thread_stack__sample(ptq->thread, ptq->chain,
+		thread_stack__sample(ptq->thread, ptq->cpu, ptq->chain,
 				     pt->synth_opts.callchain_sz + 1,
 				     sample->ip, pt->kernel_start);
 		sample->callchain = ptq->chain;
@@ -1526,11 +1526,11 @@ static int intel_pt_sample(struct intel_pt_queue *ptq)
 		return 0;
 
 	if (pt->synth_opts.callchain || pt->synth_opts.thread_stack)
-		thread_stack__event(ptq->thread, ptq->flags, state->from_ip,
+		thread_stack__event(ptq->thread, ptq->cpu, ptq->flags, state->from_ip,
 				    state->to_ip, ptq->insn_len,
 				    state->trace_nr);
 	else
-		thread_stack__set_trace_nr(ptq->thread, state->trace_nr);
+		thread_stack__set_trace_nr(ptq->thread, ptq->cpu, state->trace_nr);
 
 	if (pt->sample_branches) {
 		err = intel_pt_synth_branch_sample(ptq);
diff --git a/tools/perf/util/python.c b/tools/perf/util/python.c
index 47628e85c5eb..dda0ac978b1e 100644
--- a/tools/perf/util/python.c
+++ b/tools/perf/util/python.c
@@ -939,7 +939,8 @@ static PyObject *pyrf_evlist__get_pollfd(struct pyrf_evlist *pevlist,
 
 		file = PyFile_FromFile(fp, "perf", "r", NULL);
 #else
-		file = PyFile_FromFd(evlist->pollfd.entries[i].fd, "perf", "r", -1, NULL, NULL, NULL, 1);
+		file = PyFile_FromFd(evlist->pollfd.entries[i].fd, "perf", "r", -1,
+				     NULL, NULL, NULL, 0);
 #endif
 		if (file == NULL)
 			goto free_list;
diff --git a/tools/perf/util/session.c b/tools/perf/util/session.c
index 78a067777144..5456c84c7dd1 100644
--- a/tools/perf/util/session.c
+++ b/tools/perf/util/session.c
@@ -1527,6 +1527,13 @@ struct thread *perf_session__findnew(struct perf_session *session, pid_t pid)
 	return machine__findnew_thread(&session->machines.host, -1, pid);
 }
 
+/*
+ * Threads are identified by pid and tid, and the idle task has pid == tid == 0.
+ * So here a single thread is created for that, but actually there is a separate
+ * idle task per cpu, so there should be one 'struct thread' per cpu, but there
+ * is only 1. That causes problems for some tools, requiring workarounds. For
+ * example get_idle_thread() in builtin-sched.c, or thread_stack__per_cpu().
+ */
 int perf_session__register_idle_thread(struct perf_session *session)
 {
 	struct thread *thread;
diff --git a/tools/perf/util/thread-stack.c b/tools/perf/util/thread-stack.c
index 61a4286a74dc..d52f27f373ce 100644
--- a/tools/perf/util/thread-stack.c
+++ b/tools/perf/util/thread-stack.c
@@ -15,6 +15,7 @@
 
 #include <linux/rbtree.h>
 #include <linux/list.h>
+#include <linux/log2.h>
 #include <errno.h>
 #include "thread.h"
 #include "event.h"
@@ -60,6 +61,7 @@ struct thread_stack_entry {
  * @last_time: last timestamp
  * @crp: call/return processor
  * @comm: current comm
+ * @arr_sz: size of array if this is the first element of an array
  */
 struct thread_stack {
 	struct thread_stack_entry *stack;
@@ -71,8 +73,19 @@ struct thread_stack {
 	u64 last_time;
 	struct call_return_processor *crp;
 	struct comm *comm;
+	unsigned int arr_sz;
 };
 
+/*
+ * Assume pid == tid == 0 identifies the idle task as defined by
+ * perf_session__register_idle_thread(). The idle task is really 1 task per cpu,
+ * and therefore requires a stack for each cpu.
+ */
+static inline bool thread_stack__per_cpu(struct thread *thread)
+{
+	return !(thread->tid || thread->pid_);
+}
+
 static int thread_stack__grow(struct thread_stack *ts)
 {
 	struct thread_stack_entry *new_stack;
@@ -91,19 +104,14 @@ static int thread_stack__grow(struct thread_stack *ts)
 	return 0;
 }
 
-static struct thread_stack *thread_stack__new(struct thread *thread,
-					      struct call_return_processor *crp)
+static int thread_stack__in