1 files changed, 362 insertions, 134 deletions

diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c
index 32d50492505d..8f36ce813bc5 100644
--- a/tools/perf/util/machine.c
+++ b/tools/perf/util/machine.c
@@ -24,6 +24,7 @@
 
 #include "sane_ctype.h"
 #include <symbol/kallsyms.h>
+#include <linux/mman.h>
 
 static void __machine__remove_thread(struct machine *machine, struct thread *th, bool lock);
 
@@ -81,8 +82,7 @@ int machine__init(struct machine *machine, const char *root_dir, pid_t pid)
 	machine->kptr_restrict_warned = false;
 	machine->comm_exec = false;
 	machine->kernel_start = 0;
-
-	memset(machine->vmlinux_maps, 0, sizeof(machine->vmlinux_maps));
+	machine->vmlinux_map = NULL;
 
 	machine->root_dir = strdup(root_dir);
 	if (machine->root_dir == NULL)
@@ -137,13 +137,11 @@ struct machine *machine__new_kallsyms(void)
 	struct machine *machine = machine__new_host();
 	/*
 	 * FIXME:
-	 * 1) MAP__FUNCTION will go away when we stop loading separate maps for
-	 *    functions and data objects.
-	 * 2) We should switch to machine__load_kallsyms(), i.e. not explicitely
+	 * 1) We should switch to machine__load_kallsyms(), i.e. not explicitely
 	 *    ask for not using the kcore parsing code, once this one is fixed
 	 *    to create a map per module.
 	 */
-	if (machine && machine__load_kallsyms(machine, "/proc/kallsyms", MAP__FUNCTION) <= 0) {
+	if (machine && machine__load_kallsyms(machine, "/proc/kallsyms") <= 0) {
 		machine__delete(machine);
 		machine = NULL;
 	}
@@ -410,23 +408,16 @@ out_err:
 }
 
 /*
- * Caller must eventually drop thread->refcnt returned with a successful
- * lookup/new thread inserted.
+ * Front-end cache - TID lookups come in blocks,
+ * so most of the time we dont have to look up
+ * the full rbtree:
  */
-static struct thread *____machine__findnew_thread(struct machine *machine,
-						  struct threads *threads,
-						  pid_t pid, pid_t tid,
-						  bool create)
+static struct thread*
+__threads__get_last_match(struct threads *threads, struct machine *machine,
+			  int pid, int tid)
 {
-	struct rb_node **p = &threads->entries.rb_node;
-	struct rb_node *parent = NULL;
 	struct thread *th;
 
-	/*
-	 * Front-end cache - TID lookups come in blocks,
-	 * so most of the time we dont have to look up
-	 * the full rbtree:
-	 */
 	th = threads->last_match;
 	if (th != NULL) {
 		if (th->tid == tid) {
@@ -437,12 +428,57 @@ static struct thread *____machine__findnew_thread(struct machine *machine,
 		threads->last_match = NULL;
 	}
 
+	return NULL;
+}
+
+static struct thread*
+threads__get_last_match(struct threads *threads, struct machine *machine,
+			int pid, int tid)
+{
+	struct thread *th = NULL;
+
+	if (perf_singlethreaded)
+		th = __threads__get_last_match(threads, machine, pid, tid);
+
+	return th;
+}
+
+static void
+__threads__set_last_match(struct threads *threads, struct thread *th)
+{
+	threads->last_match = th;
+}
+
+static void
+threads__set_last_match(struct threads *threads, struct thread *th)
+{
+	if (perf_singlethreaded)
+		__threads__set_last_match(threads, th);
+}
+
+/*
+ * Caller must eventually drop thread->refcnt returned with a successful
+ * lookup/new thread inserted.
+ */
+static struct thread *____machine__findnew_thread(struct machine *machine,
+						  struct threads *threads,
+						  pid_t pid, pid_t tid,
+						  bool create)
+{
+	struct rb_node **p = &threads->entries.rb_node;
+	struct rb_node *parent = NULL;
+	struct thread *th;
+
+	th = threads__get_last_match(threads, machine, pid, tid);
+	if (th)
+		return th;
+
 	while (*p != NULL) {
 		parent = *p;
 		th = rb_entry(parent, struct thread, rb_node);
 
 		if (th->tid == tid) {
-			threads->last_match = th;
+			threads__set_last_match(threads, th);
 			machine__update_thread_pid(machine, th, pid);
 			return thread__get(th);
 		}
@@ -479,7 +515,7 @@ static struct thread *____machine__findnew_thread(struct machine *machine,
 		 * It is now in the rbtree, get a ref
 		 */
 		thread__get(th);
-		threads->last_match = th;
+		threads__set_last_match(threads, th);
 		++threads->nr;
 	}
 
@@ -673,8 +709,7 @@ struct map *machine__findnew_module_map(struct machine *machine, u64 start,
 	if (kmod_path__parse_name(&m, filename))
 		return NULL;
 
-	map = map_groups__find_by_name(&machine->kmaps, MAP__FUNCTION,
-				       m.name);
+	map = map_groups__find_by_name(&machine->kmaps, m.name);
 	if (map) {
 		/*
 		 * If the map's dso is an offline module, give dso__load()
@@ -689,7 +724,7 @@ struct map *machine__findnew_module_map(struct machine *machine, u64 start,
 	if (dso == NULL)
 		goto out;
 
-	map = map__new2(start, dso, MAP__FUNCTION);
+	map = map__new2(start, dso);
 	if (map == NULL)
 		goto out;
 
@@ -810,8 +845,8 @@ struct process_args {
 	u64 start;
 };
 
-static void machine__get_kallsyms_filename(struct machine *machine, char *buf,
-					   size_t bufsz)
+void machine__get_kallsyms_filename(struct machine *machine, char *buf,
+				    size_t bufsz)
 {
 	if (machine__is_default_guest(machine))
 		scnprintf(buf, bufsz, "%s", symbol_conf.default_guest_kallsyms);
@@ -854,65 +889,171 @@ static int machine__get_running_kernel_start(struct machine *machine,
 	return 0;
 }
 
+int machine__create_extra_kernel_map(struct machine *machine,
+				     struct dso *kernel,
+				     struct extra_kernel_map *xm)
+{
+	struct kmap *kmap;
+	struct map *map;
+
+	map = map__new2(xm->start, kernel);
+	if (!map)
+		return -1;
+
+	map->end   = xm->end;
+	map->pgoff = xm->pgoff;
+
+	kmap = map__kmap(map);
+
+	kmap->kmaps = &machine->kmaps;
+	strlcpy(kmap->name, xm->name, KMAP_NAME_LEN);
+
+	map_groups__insert(&machine->kmaps, map);
+
+	pr_debug2("Added extra kernel map %s %" PRIx64 "-%" PRIx64 "\n",
+		  kmap->name, map->start, map->end);
+
+	map__put(map);
+
+	return 0;
+}
+
+static u64 find_entry_trampoline(struct dso *dso)
+{
+	/* Duplicates are removed so lookup all aliases */
+	const char *syms[] = {
+		"_entry_trampoline",
+		"__entry_trampoline_start",
+		"entry_SYSCALL_64_trampoline",
+	};
+	struct symbol *sym = dso__first_symbol(dso);
+	unsigned int i;
+
+	for (; sym; sym = dso__next_symbol(sym)) {
+		if (sym->binding != STB_GLOBAL)
+			continue;
+		for (i = 0; i < ARRAY_SIZE(syms); i++) {
+			if (!strcmp(sym->name, syms[i]))
+				return sym->start;
+		}
+	}
+
+	return 0;
+}
+
+/*
+ * These values can be used for kernels that do not have symbols for the entry
+ * trampolines in kallsyms.
+ */
+#define X86_64_CPU_ENTRY_AREA_PER_CPU	0xfffffe0000000000ULL
+#define X86_64_CPU_ENTRY_AREA_SIZE	0x2c000
+#define X86_64_ENTRY_TRAMPOLINE		0x6000
+
+/* Map x86_64 PTI entry trampolines */
+int machine__map_x86_64_entry_trampolines(struct machine *machine,
+					  struct dso *kernel)
+{
+	struct map_groups *kmaps = &machine->kmaps;
+	struct maps *maps = &kmaps->maps;
+	int nr_cpus_avail, cpu;
+	bool found = false;
+	struct map *map;
+	u64 pgoff;
+
+	/*
+	 * In the vmlinux case, pgoff is a virtual address which must now be
+	 * mapped to a vmlinux offset.
+	 */
+	for (map = maps__first(maps); map; map = map__next(map)) {
+		struct kmap *kmap = __map__kmap(map);
+		struct map *dest_map;
+
+		if (!kmap || !is_entry_trampoline(kmap->name))
+			continue;
+
+		dest_map = map_groups__find(kmaps, map->pgoff);
+		if (dest_map != map)
+			map->pgoff = dest_map->map_ip(dest_map, map->pgoff);
+		found = true;
+	}
+	if (found || machine->trampolines_mapped)
+		return 0;
+
+	pgoff = find_entry_trampoline(kernel);
+	if (!pgoff)
+		return 0;
+
+	nr_cpus_avail = machine__nr_cpus_avail(machine);
+
+	/* Add a 1 page map for each CPU's entry trampoline */
+	for (cpu = 0; cpu < nr_cpus_avail; cpu++) {
+		u64 va = X86_64_CPU_ENTRY_AREA_PER_CPU +
+			 cpu * X86_64_CPU_ENTRY_AREA_SIZE +
+			 X86_64_ENTRY_TRAMPOLINE;
+		struct extra_kernel_map xm = {
+			.start = va,
+			.end   = va + page_size,
+			.pgoff = pgoff,
+		};
+
+		strlcpy(xm.name, ENTRY_TRAMPOLINE_NAME, KMAP_NAME_LEN);
+
+		if (machine__create_extra_kernel_map(machine, kernel, &xm) < 0)
+			return -1;
+	}
+
+	machine->trampolines_mapped = nr_cpus_avail;
+
+	return 0;
+}
+
+int __weak machine__create_extra_kernel_maps(struct machine *machine __maybe_unused,
+					     struct dso *kernel __maybe_unused)
+{
+	return 0;
+}
+
 static int
 __machine__create_kernel_maps(struct machine *machine, struct dso *kernel)
 {
-	int type;
+	struct kmap *kmap;
+	struct map *map;
 
 	/* In case of renewal the kernel map, destroy previous one */
 	machine__destroy_kernel_maps(machine);
 
-	for (type = 0; type < MAP__NR_TYPES; ++type) {
-		struct kmap *kmap;
-		struct map *map;
-
-		machine->vmlinux_maps[type] = map__new2(0, kernel, type);
-		if (machine->vmlinux_maps[type] == NULL)
-			return -1;
+	machine->vmlinux_map = map__new2(0, kernel);
+	if (machine->vmlinux_map == NULL)
+		return -1;
 
-		machine->vmlinux_maps[type]->map_ip =
-			machine->vmlinux_maps[type]->unmap_ip =
-				identity__map_ip;
-		map = __machine__kernel_map(machine, type);
-		kmap = map__kmap(map);
-		if (!kmap)
-			return -1;
+	machine->vmlinux_map->map_ip = machine->vmlinux_map->unmap_ip = identity__map_ip;
+	map = machine__kernel_map(machine);
+	kmap = map__kmap(map);
+	if (!kmap)
+		return -1;
 
-		kmap->kmaps = &machine->kmaps;
-		map_groups__insert(&machine->kmaps, map);
-	}
+	kmap->kmaps = &machine->kmaps;
+	map_groups__insert(&machine->kmaps, map);
 
 	return 0;
 }
 
 void machine__destroy_kernel_maps(struct machine *machine)
 {
-	int type;
-
-	for (type = 0; type < MAP__NR_TYPES; ++type) {
-		struct kmap *kmap;
-		struct map *map = __machine__kernel_map(machine, type);
-
-		if (map == NULL)
-			continue;
+	struct kmap *kmap;
+	struct map *map = machine__kernel_map(machine);
 
-		kmap = map__kmap(map);
-		map_groups__remove(&machine->kmaps, map);
-		if (kmap && kmap->ref_reloc_sym) {
-			/*
-			 * ref_reloc_sym is shared among all maps, so free just
-			 * on one of them.
-			 */
-			if (type == MAP__FUNCTION) {
-				zfree((char **)&kmap->ref_reloc_sym->name);
-				zfree(&kmap->ref_reloc_sym);
-			} else
-				kmap->ref_reloc_sym = NULL;
-		}
+	if (map == NULL)
+		return;
 
-		map__put(machine->vmlinux_maps[type]);
-		machine->vmlinux_maps[type] = NULL;
+	kmap = map__kmap(map);
+	map_groups__remove(&machine->kmaps, map);
+	if (kmap && kmap->ref_reloc_sym) {
+		zfree((char **)&kmap->ref_reloc_sym->name);
+		zfree(&kmap->ref_reloc_sym);
 	}
+
+	map__zput(machine->vmlinux_map);
 }
 
 int machines__create_guest_kernel_maps(struct machines *machines)
@@ -989,32 +1130,31 @@ int machines__create_kernel_maps(struct machines *machines, pid_t pid)
 	return machine__create_kernel_maps(machine);
 }
 
-int machine__load_kallsyms(struct machine *machine, const char *filename,
-			     enum map_type type)
+int machine__load_kallsyms(struct machine *machine, const char *filename)
 {
 	struct map *map = machine__kernel_map(machine);
 	int ret = __dso__load_kallsyms(map->dso, filename, map, true);
 
 	if (ret > 0) {
-		dso__set_loaded(map->dso, type);
+		dso__set_loaded(map->dso);
 		/*
 		 * Since /proc/kallsyms will have multiple sessions for the
 		 * kernel, with modules between them, fixup the end of all
 		 * sections.
 		 */
-		__map_groups__fixup_end(&machine->kmaps, type);
+		map_groups__fixup_end(&machine->kmaps);
 	}
 
 	return ret;
 }
 
-int machine__load_vmlinux_path(struct machine *machine, enum map_type type)
+int machine__load_vmlinux_path(struct machine *machine)
 {
 	struct map *map = machine__kernel_map(machine);
 	int ret = dso__load_vmlinux_path(map->dso, map);
 
 	if (ret > 0)
-		dso__set_loaded(map->dso, type);
+		dso__set_loaded(map->dso);
 
 	return ret;
 }
@@ -1055,10 +1195,9 @@ static bool is_kmod_dso(struct dso *dso)
 static int map_groups__set_module_path(struct map_groups *mg, const char *path,
 				       struct kmod_path *m)
 {
-	struct map *map;
 	char *long_name;
+	struct map *map = map_groups__find_by_name(mg, m->name);
 
-	map = map_groups__find_by_name(mg, MAP__FUNCTION, m->name);
 	if (map == NULL)
 		return 0;
 
@@ -1073,8 +1212,10 @@ static int map_groups__set_module_path(struct map_groups *mg, const char *path,
 	 * Full name could reveal us kmod compression, so
 	 * we need to update the symtab_type if needed.
 	 */
-	if (m->comp && is_kmod_dso(map->dso))
+	if (m->comp && is_kmod_dso(map->dso)) {
 		map->dso->symtab_type++;
+		map->dso->comp = m->comp;
+	}
 
 	return 0;
 }
@@ -1207,19 +1348,14 @@ static int machine__create_modules(struct machine *machine)
 static void machine__set_kernel_mmap(struct machine *machine,
 				     u64 start, u64 end)
 {
-	int i;
-
-	for (i = 0; i < MAP__NR_TYPES; i++) {
-		machine->vmlinux_maps[i]->start = start;
-		machine->vmlinux_maps[i]->end   = end;
-
-		/*
-		 * Be a bit paranoid here, some perf.data file came with
-		 * a zero sized synthesized MMAP event for the kernel.
-		 */
-		if (start == 0 && end == 0)
-			machine->vmlinux_maps[i]->end = ~0ULL;
-	}
+	machine->vmlinux_map->start = start;
+	machine->vmlinux_map->end   = end;
+	/*
+	 * Be a bit paranoid here, some perf.data file came with
+	 * a zero sized synthesized MMAP event for the kernel.
+	 */
+	if (start == 0 && end == 0)
+		machine->vmlinux_map->end = ~0ULL;
 }
 
 int machine__create_kernel_maps(struct machine *machine)
@@ -1234,9 +1370,8 @@ int machine__create_kernel_maps(struct machine *machine)
 		return -1;
 
 	ret = __machine__create_kernel_maps(machine, kernel);
-	dso__put(kernel);
 	if (ret < 0)
-		return -1;
+		goto out_put;
 
 	if (symbol_conf.use_modules && machine__create_modules(machine) < 0) {
 		if (machine__is_host(machine))
@@ -1249,9 +1384,10 @@ int machine__create_kernel_maps(struct machine *machine)
 
 	if (!machine__get_running_kernel_start(machine, &name, &addr)) {
 		if (name &&
-		    maps__set_kallsyms_ref_reloc_sym(machine->vmlinux_maps, name, addr)) {
+		    map__set_kallsyms_ref_reloc_sym(machine->vmlinux_map, name, addr)) {
 			machine__destroy_kernel_maps(machine);
-			return -1;
+			ret = -1;
+			goto out_put;
 		}
 
 		/* we have a real start address now, so re-order the kmaps */
@@ -1267,12 +1403,16 @@ int machine__create_kernel_maps(struct machine *machine)
 		map__put(map);
 	}
 
+	if (machine__create_extra_kernel_maps(machine, kernel))
+		pr_debug("Problems creating extra kernel maps, continuing anyway...\n");
+
 	/* update end address of the kernel map using adjacent module address */
 	map = map__next(machine__kernel_map(machine));
 	if (map)
 		machine__set_kernel_mmap(machine, addr, map->start);
-
-	return 0;
+out_put:
+	dso__put(kernel);
+	return ret;
 }
 
 static bool machine__uses_kcore(struct machine *machine)
@@ -1287,6 +1427,32 @@ static bool machine__uses_kcore(struct machine *machine)
 	return false;
 }
 
+static bool perf_event__is_extra_kernel_mmap(struct machine *machine,
+					     union perf_event *event)
+{
+	return machine__is(machine, "x86_64") &&
+	       is_entry_trampoline(event->mmap.filename);
+}
+
+static int machine__process_extra_kernel_map(struct machine *machine,
+					     union perf_event *event)
+{
+	struct map *kernel_map = machine__kernel_map(machine);
+	struct dso *kernel = kernel_map ? kernel_map->dso : NULL;
+	struct extra_kernel_map xm = {
+		.start = event->mmap.start,
+		.end   = event->mmap.start + event->mmap.len,
+		.pgoff = event->mmap.pgoff,
+	};
+
+	if (kernel == NULL)
+		return -1;
+
+	strlcpy(xm.name, event->mmap.filename, KMAP_NAME_LEN);
+
+	return machine__create_extra_kernel_map(machine, kernel, &xm);
+}
+
 static int machine__process_kernel_mmap_event(struct machine *machine,
 					      union perf_event *event)
 {
@@ -1379,9 +1545,9 @@ static int machine__process_kernel_mmap_event(struct machine *machine,
 		 * time /proc/sys/kernel/kptr_restrict was non zero.
 		 */
 		if (event->mmap.pgoff != 0) {
-			maps__set_kallsyms_ref_reloc_sym(machine->vmlinux_maps,
-							 symbol_name,
-							 event->mmap.pgoff);
+			map__set_kallsyms_ref_reloc_sym(machine->vmlinux_map,
+							symbol_name,
+							event->mmap.pgoff);
 		}
 
 		if (machine__is_default_guest(machine)) {
@@ -1390,6 +1556,8 @@ static int machine__process_kernel_mmap_event(struct machine *machine,
 			 */
 			dso__load(kernel, machine__kernel_map(machine));
 		}
+	} else if (perf_event__is_extra_kernel_mmap(machine, event)) {
+		return machine__process_extra_kernel_map(machine, event);
 	}
 	return 0;
 out_problem:
@@ -1402,7 +1570,6 @@ int machine__process_mmap2_event(struct machine *machine,
 {
 	struct thread *thread;
 	struct map *map;
-	enum map_type type;
 	int ret = 0;
 
 	if (dump_trace)
@@ -1421,11 +1588,6 @@ int machine__process_mmap2_event(struct machine *machine,
 	if (thread == NULL)
 		goto out_problem;
 
-	if (event->header.misc & PERF_RECORD_MISC_MMAP_DATA)
-		type = MAP__VARIABLE;
-	else
-		type = MAP__FUNCTION;
-
 	map = map__new(machine, event->mmap2.start,
 			event->mmap2.len, event->mmap2.pgoff,
 			event->mmap2.maj,
@@ -1433,7 +1595,7 @@ int machine__process_mmap2_event(struct machine *machine,
 			event->mmap2.ino_generation,
 			event->mmap2.prot,
 			event->mmap2.flags,
-			event->mmap2.filename, type, thread);
+			event->mmap2.filename, thread);
 
 	if (map == NULL)
 		goto out_problem_map;
@@ -1460,7 +1622,7 @@ int machine__process_mmap_event(struct machine *machine, union perf_event *event
 {
 	struct thread *thread;
 	struct map *map;
-	enum map_type type;
+	u32 prot = 0;
 	int ret = 0;
 
 	if (dump_trace)
@@ -1479,16 +1641,14 @@ int machine__process_mmap_event(struct machine *machine, union perf_event *event
 	if (thread == NULL)
 		goto out_problem;
 
-	if (event->header.misc & PERF_RECORD_MISC_MMAP_DATA)
-		type = MAP__VARIABLE;
-	else
-		type = MAP__FUNCTION;
+	if (!(event->header.misc & PERF_RECORD_MISC_MMAP_DATA))
+		prot = PROT_EXEC;
 
 	map = map__new(machine, event->mmap.start,
 			event->mmap.len, event->mmap.pgoff,
-			0, 0, 0, 0, 0, 0,
+			0, 0, 0, 0, prot, 0,
 			event->mmap.filename,
-			type, thread);
+			thread);
 
 	if (map == NULL)
 		goto out_problem_map;
@@ -1515,7 +1675,7 @@ static void __machine__remove_thread(struct machine *machine, struct thread *th,
 	struct threads *threads = machine__threads(machine, th->tid);
 
 	if (threads->last_match == th)
-		threads->last_match = NULL;
+		threads__set_last_match(threads, NULL);
 
 	BUG_ON(refcount_read(&th->refcnt) == 0);
 	if (lock)
@@ -1548,6 +1708,7 @@ int machine__process_fork_event(struct machine *machine, union perf_event *event
 	struct thread *parent = machine__findnew_thread(machine,
 							event->fork.ppid,
 							event->fork.ptid);
+	bool do_maps_clone = true;
 	int err = 0;
 
 	if (dump_trace)
@@ -1576,9 +1737,25 @@ int machine__process_fork_event(struct machine *machine, union perf_event *event
 
 	thread = machine__findnew_thread(machine, event->fork.pid,
 					 event->fork.tid);
+	/*
+	 * When synthesizing FORK events, we are trying to create thread
+	 * objects for the already running tasks on the machine.
+	 *
+	 * Normally, for a kernel FORK event, we want to clone the parent's
+	 * maps because that is what the kernel just did.
+	 *
+	 * But when synthesizing, this should not be done.  If we do, we end up
+	 * with overlapping maps as we process the sythesized MMAP2 events that
+	 * get delivered shortly thereafter.
+	 *
+	 * Use the FORK event misc flags in an internal way to signal this
+	 * situation, so we can elide the map clone when appropriate.
+	 */
+	if (event->fork.header.misc & PERF_RECORD_MISC_FORK_EXEC)
+		do_maps_clone = false;
 
 	if (thread == NULL || parent == NULL ||
-	    thread__fork(thread, parent, sample->time) < 0) {
+	    thread__fork(thread, parent, sample->time, do_maps_clone) < 0) {
 		dump_printf("problem processing PERF_RECORD_FORK, skipping event.\n");
 		err = -1;
 	}
@@ -1664,7 +1841,7 @@ static void ip__resolve_ams(struct thread *thread,
 	 * Thus, we have to try consecutively until we find a match
 	 * or else, the symbol is unknown
 	 */
-	thread__find_cpumode_addr_location(thread, MAP__FUNCTION, ip, &al);
+	thread__find_cpumode_addr_location(thread, ip, &al);
 
 	ams->addr = ip;
 	ams->al_addr = al.addr;
@@ -1681,15 +1858,7 @@ static void ip__resolve_data(struct thread *thread,
 
 	memset(&al, 0, sizeof(al));
 
-	thread__find_addr_location(thread, m, MAP__VARIABLE, addr, &al);
-	if (al.map == NULL) {
-		/*
-		 * some shared data regions have execute bit set which puts
-		 * their mapping in the MAP__FUNCTION type array.
-		 * Check there as a fallback option before dropping the sample.
-		 */
-		thread__find_addr_location(thread, m, MAP__FUNCTION, addr, &al);
-	}
+	thread__find_symbol(thread, m, addr, &al);
 
 	ams->addr = addr;
 	ams->al_addr = al.addr;
@@ -1758,8 +1927,7 @@ static int add_callchain_ip(struct thread *thread,
 	al.filtered = 0;
 	al.sym = NULL;
 	if (!cpumode) {
-		thread__find_cpumode_addr_location(thread, MAP__FUNCTION,
-						   ip, &al);
+		thread__find_cpumode_addr_location(thread, ip, &al);
 	} else {
 		if (ip >= PERF_CONTEXT_MAX) {
 			switch (ip) {
@@ -1784,8 +1952,7 @@ static int add_callchain_ip(struct thread *thread,
 			}
 			return 0;
 		}
-		thread__find_addr_location(thread, *cpumode, MAP__FUNCTION,
-					   ip, &al);
+		thread__find_symbol(thread, *cpumode, ip, &al);
 	}
 
 	if (al.sym != NULL) {
@@ -1810,7 +1977,7 @@ static int add_callchain_ip(struct thread *thread,
 	}
 
 	srcline = callchain_srcline(al.map, al.sym, al.addr);
-	return callchain_cursor_append(cursor, al.addr, al.map, al.sym,
+	return callchain_cursor_append(cursor, ip, al.map, al.sym,
 				       branch, flags, nr_loop_iter,
 				       iter_cycles, branch_from, srcline);
 }
@@ -1990,6 +2157,27 @@ static int resolve_lbr_callchain_sample(struct thread *thread,
 	return 0;
 }
 
+static int find_prev_cpumode(struct ip_callchain *chain, struct thread *thread,
+			     struct callchain_cursor *cursor,
+			     struct symbol **parent,
+			     struct addr_location *root_al,
+			     u8 *cpumode, int ent)
+{
+	int err = 0;
+
+	while (--ent >= 0) {
+		u64 ip = chain->ips[ent];
+
+		if (ip >= PERF_CONTEXT_MAX) {
+			err = add_callchain_ip(thread, cursor, parent,
+					       root_al, cpumode, ip,
+					       false, NULL, NULL, 0);
+			break;
+		}
+	}
+	return err;
+}
+
 static int thread__resolve_callchain_sample(struct thread *thread,
 					    struct callchain_cursor *cursor,
 					    struct perf_evsel *evsel,
@@ -2096,6 +2284,12 @@ static int thread__resolve_callchain_sample(struct thread *thread,
 	}
 
 check_calls:
+	if (callchain_param.order != ORDER_CALLEE) {
+		err = find_prev_cpumode(chain, thread, cursor, parent, root_al,
+					&cpumode, chain->nr - first_call);
+		if (err)
+			return (err < 0) ? err : 0;
+	}
 	for (i = first_call, nr_entries = 0;
 	     i < chain_nr && nr_entries < max_stack; i++) {
 		u64 ip;
@@ -2110,9 +2304,15 @@ check_calls:
 			continue;
 #endif
 		ip = chain->ips[j];
-
 		if (ip < PERF_CONTEXT_MAX)
                        ++nr_entries;
+		else if (callchain_param.order != ORDER_CALLEE) {
+			err = find_prev_cpumode(chain, thread, cursor, parent,
+						root_al, &cpumode, j);
+			if (err)
+				return (err < 0) ? err : 0;
+			continue;
+		}
 
 		err = add_callchain_ip(thread, cursor, parent,
 				       root_al, &cpumode, ip,
@@ -2136,7 +2336,8 @@ static int append_inlines(struct callchain_cursor *cursor,
 	if (!symbol_conf.inline_name || !map || !sym)
 		return ret;
 
-	addr = map__rip_2objdump(map, ip);
+	addr = map__map_ip(map, ip);
+	addr = map__rip_2objdump(map, addr);
 
 	inline_node = inlines__tree_find(&map->dso->inlined_nodes, addr);
 	if (!inline_node) {
@@ -2162,6 +2363,7 @@ static int unwind_entry(struct unwind_entry *entry, void *arg)
 {
 	struct callchain_cursor *cursor = arg;
 	const char *srcline = NULL;
+	u64 addr = entry->ip;
 
 	if (symbol_conf.hide_unresolved && entry->sym == NULL)
 		return 0;
@@ -2169,7 +2371,14 @@ static int unwind_entry(struct unwind_entry *entry, void *arg)
 	if (append_inlines(cursor, entry->map, entry->sym, entry->ip) == 0)
 		return 0;
 
-	srcline = callchain_srcline(entry->map, entry->sym, entry->ip);
+	/*
+	 * Convert entry->ip from a virtual address to an offset in
+	 * its corresponding binary.
+	 */
+	if (entry->map)
+		addr = map__map_ip(entry->map, entry->ip);
+
+	srcline = callchain_srcline(entry->map, entry->sym, addr);
 	return callchain_cursor_append(cursor, entry->ip,
 				       entry->map, entry->sym,
 				       false, NULL, 0, 0, 0, srcline);
@@ -2342,6 +2551,20 @@ int machine__set_current_tid(struct machine *machine, int cpu, pid_t pid,
 	return 0;
 }
 
+/*
+ * Compares the raw arch string. N.B. see instead perf_env__arch() if a
+ * normalized arch is needed.
+ */
+bool machine__is(struct machine *machine, const char *arch)
+{
+	return machine && !strcmp(perf_env__raw_arch(machine->env), arch);
+}
+
+int machine__nr_cpus_avail(struct machine *machine)
+{
+	return machine ? perf_env__nr_cpus_avail(machine->env) : 0;
+}
+
 int machine__get_kernel_start(struct machine *machine)
 {
 	struct map *map = machine__kernel_map(machine);
@@ -2358,7 +2581,12 @@ int machine__get_kernel_start(struct machine *machine)
 	machine->kernel_start = 1ULL << 63;
 	if (map) {
 		err = map__load(map);
-		if (!err)
+		/*
+		 * On x86_64, PTI entry trampolines are less than the
+		 * start of kernel text, but still above 2^63. So leave
+		 * kernel_start = 1ULL << 63 for x86_64.
+		 */
+		if (!err && !machine__is(machine, "x86_64"))
 			machine->kernel_start = map->start;
 	}
 	return err;
@@ -2373,7 +2601,7 @@ char *machine__resolve_kernel_addr(void *vmachine, unsigned long long *addrp, ch
 {
 	struct machine *machine = vmachine;
 	struct map *map;
-	struct symbol *sym = map_groups__find_symbol(&machine->kmaps, MAP__FUNCTION, *addrp, &map);
+	struct symbol *sym = machine__find_kernel_symbol(machine, *addrp, &map);
 
 	if (sym == NULL)
 		return NULL;