11 files changed, 78 insertions, 29 deletions

diff --git a/tools/perf/Makefile b/tools/perf/Makefile
index 41abb90df50d..1950e19af1cf 100644
--- a/tools/perf/Makefile
+++ b/tools/perf/Makefile
@@ -5,6 +5,12 @@ endif
 # The default target of this Makefile is...
 all::
 
+ifneq ($(OUTPUT),)
+# check that the output directory actually exists
+OUTDIR := $(shell cd $(OUTPUT) && /bin/pwd)
+$(if $(OUTDIR),, $(error output directory "$(OUTPUT)" does not exist))
+endif
+
 # Define V=1 to have a more verbose compile.
 # Define V=2 to have an even more verbose compile.
 #
@@ -157,10 +163,6 @@ all::
 #
 # Define NO_DWARF if you do not want debug-info analysis feature at all.
 
-$(shell sh -c 'mkdir -p $(OUTPUT)scripts/{perl,python}/Perf-Trace-Util/' 2> /dev/null)
-$(shell sh -c 'mkdir -p $(OUTPUT)util/{ui/browsers,scripting-engines}/' 2> /dev/null)
-$(shell sh -c 'mkdir $(OUTPUT)bench' 2> /dev/null)
-
 $(OUTPUT)PERF-VERSION-FILE: .FORCE-PERF-VERSION-FILE
         @$(SHELL_PATH) util/PERF-VERSION-GEN $(OUTPUT)
 -include $(OUTPUT)PERF-VERSION-FILE
@@ -186,8 +188,6 @@ ifeq ($(ARCH),x86_64)
         ARCH := x86
 endif
 
-$(shell sh -c 'mkdir -p $(OUTPUT)arch/$(ARCH)/util/' 2> /dev/null)
-
 # CFLAGS and LDFLAGS are for the users to override from the command line.
 
 #
@@ -268,6 +268,7 @@ export prefix bindir sharedir sysconfdir
 CC = $(CROSS_COMPILE)gcc
 AR = $(CROSS_COMPILE)ar
 RM = rm -f
+MKDIR = mkdir
 TAR = tar
 FIND = find
 INSTALL = install
@@ -838,6 +839,7 @@ ifndef V
         QUIET_CC       = @echo '   ' CC $@;
         QUIET_AR       = @echo '   ' AR $@;
         QUIET_LINK     = @echo '   ' LINK $@;
+        QUIET_MKDIR    = @echo '   ' MKDIR $@;
         QUIET_BUILT_IN = @echo '   ' BUILTIN $@;
         QUIET_GEN      = @echo '   ' GEN $@;
         QUIET_SUBDIR0  = +@subdir=
@@ -935,15 +937,15 @@ $(OUTPUT)common-cmds.h: $(wildcard Documentation/perf-*.txt)
         $(QUIET_GEN). util/generate-cmdlist.sh > $@+ && mv $@+ $@
 
 $(patsubst %.sh,%,$(SCRIPT_SH)) : % : %.sh
-        $(QUIET_GEN)$(RM) $@ $@+ && \
+        $(QUIET_GEN)$(RM) $(OUTPUT)$@ $(OUTPUT)$@+ && \
         sed -e '1s|#!.*/sh|#!$(SHELL_PATH_SQ)|' \
             -e 's|@SHELL_PATH@|$(SHELL_PATH_SQ)|' \
             -e 's|@@PERL@@|$(PERL_PATH_SQ)|g' \
             -e 's/@@PERF_VERSION@@/$(PERF_VERSION)/g' \
             -e 's/@@NO_CURL@@/$(NO_CURL)/g' \
-            $@.sh >$@+ && \
-        chmod +x $@+ && \
-        mv $@+ $(OUTPUT)$@
+            $@.sh > $(OUTPUT)$@+ && \
+        chmod +x $(OUTPUT)$@+ && \
+        mv $(OUTPUT)$@+ $(OUTPUT)$@
 
 configure: configure.ac
         $(QUIET_GEN)$(RM) $@ $<+ && \
@@ -1012,6 +1014,14 @@ $(LIB_OBJS) $(BUILTIN_OBJS): $(LIB_H)
 $(patsubst perf-%$X,%.o,$(PROGRAMS)): $(LIB_H) $(wildcard */*.h)
 builtin-revert.o wt-status.o: wt-status.h
 
+# we compile into subdirectories. if the target directory is not the source directory, they might not exists. So
+# we depend the various files onto their directories.
+DIRECTORY_DEPS = $(LIB_OBJS) $(BUILTIN_OBJS) $(OUTPUT)PERF-VERSION-FILE $(OUTPUT)common-cmds.h
+$(DIRECTORY_DEPS): | $(sort $(dir $(DIRECTORY_DEPS)))
+# In the second step, we make a rule to actually create these directories
+$(sort $(dir $(DIRECTORY_DEPS))):
+        $(QUIET_MKDIR)$(MKDIR) -p $@ 2>/dev/null
+
 $(LIB_FILE): $(LIB_OBJS)
         $(QUIET_AR)$(RM) $@ && $(AR) rcs $@ $(LIB_OBJS)
 
diff --git a/tools/perf/feature-tests.mak b/tools/perf/feature-tests.mak
index ddb68e601f0e..7a7b60859053 100644
--- a/tools/perf/feature-tests.mak
+++ b/tools/perf/feature-tests.mak
@@ -113,7 +113,7 @@ endef
 # try-cc
 # Usage: option = $(call try-cc, source-to-build, cc-options)
 try-cc = $(shell sh -c                                            \
-        'TMP="$(TMPOUT).$$$$";                                    \
+        'TMP="$(OUTPUT)$(TMPOUT).$$$$";                           \
          echo "$(1)" |                                            \
          $(CC) -x c - $(2) -o "$$TMP" > /dev/null 2>&1 && echo y; \
          rm -f "$$TMP"')
diff --git a/tools/perf/perf.h b/tools/perf/perf.h
index ef7aa0a0c526..95aaf565c704 100644
--- a/tools/perf/perf.h
+++ b/tools/perf/perf.h
@@ -73,6 +73,18 @@ void get_term_dimensions(struct winsize *ws);
 #define cpu_relax()     asm volatile("":::"memory")
 #endif
 
+#ifdef __mips__
+#include "../../arch/mips/include/asm/unistd.h"
+#define rmb()           asm volatile(                                   \
+                                ".set   mips2\n\t"                      \
+                                "sync\n\t"                              \
+                                ".set   mips0"                          \
+                                : /* no output */                       \
+                                : /* no input */                        \
+                                : "memory")
+#define cpu_relax()     asm volatile("" ::: "memory")
+#endif
+
 #include <time.h>
 #include <unistd.h>
 #include <sys/types.h>
diff --git a/tools/perf/util/callchain.h b/tools/perf/util/callchain.h
index 624a96c636fd..6de4313924fb 100644
--- a/tools/perf/util/callchain.h
+++ b/tools/perf/util/callchain.h
@@ -50,6 +50,7 @@ static inline void callchain_init(struct callchain_node *node)
         INIT_LIST_HEAD(&node->children);
         INIT_LIST_HEAD(&node->val);
 
+        node->children_hit = 0;
         node->parent = NULL;
         node->hit = 0;
 }
diff --git a/tools/perf/util/probe-event.c b/tools/perf/util/probe-event.c
index e72f05c3bef0..fcc16e4349df 100644
--- a/tools/perf/util/probe-event.c
+++ b/tools/perf/util/probe-event.c
@@ -1539,6 +1539,7 @@ static int convert_to_probe_trace_events(struct perf_probe_event *pev,
                 goto error;
         }
         tev->point.offset = pev->point.offset;
+        tev->point.retprobe = pev->point.retprobe;
         tev->nargs = pev->nargs;
         if (tev->nargs) {
                 tev->args = zalloc(sizeof(struct probe_trace_arg)
diff --git a/tools/perf/util/probe-finder.c b/tools/perf/util/probe-finder.c
index 525136684d4e..32b81f707ff5 100644
--- a/tools/perf/util/probe-finder.c
+++ b/tools/perf/util/probe-finder.c
@@ -686,6 +686,25 @@ static int find_variable(Dwarf_Die *sp_die, struct probe_finder *pf)
         char buf[32], *ptr;
         int ret, nscopes;
 
+        if (!is_c_varname(pf->pvar->var)) {
+                /* Copy raw parameters */
+                pf->tvar->value = strdup(pf->pvar->var);
+                if (pf->tvar->value == NULL)
+                        return -ENOMEM;
+                if (pf->pvar->type) {
+                        pf->tvar->type = strdup(pf->pvar->type);
+                        if (pf->tvar->type == NULL)
+                                return -ENOMEM;
+                }
+                if (pf->pvar->name) {
+                        pf->tvar->name = strdup(pf->pvar->name);
+                        if (pf->tvar->name == NULL)
+                                return -ENOMEM;
+                } else
+                        pf->tvar->name = NULL;
+                return 0;
+        }
+
         if (pf->pvar->name)
                 pf->tvar->name = strdup(pf->pvar->name);
         else {
@@ -700,19 +719,6 @@ static int find_variable(Dwarf_Die *sp_die, struct probe_finder *pf)
         if (pf->tvar->name == NULL)
                 return -ENOMEM;
 
-        if (!is_c_varname(pf->pvar->var)) {
-                /* Copy raw parameters */
-                pf->tvar->value = strdup(pf->pvar->var);
-                if (pf->tvar->value == NULL)
-                        return -ENOMEM;
-                if (pf->pvar->type) {
-                        pf->tvar->type = strdup(pf->pvar->type);
-                        if (pf->tvar->type == NULL)
-                                return -ENOMEM;
-                }
-                return 0;
-        }
-
         pr_debug("Searching '%s' variable in context.\n",
                  pf->pvar->var);
         /* Search child die for local variables and parameters. */
@@ -783,6 +789,16 @@ static int convert_probe_point(Dwarf_Die *sp_die, struct probe_finder *pf)
                 /* This function has no name. */
                 tev->point.offset = (unsigned long)pf->addr;
 
+        /* Return probe must be on the head of a subprogram */
+        if (pf->pev->point.retprobe) {
+                if (tev->point.offset != 0) {
+                        pr_warning("Return probe must be on the head of"
+                                   " a real function\n");
+                        return -EINVAL;
+                }
+                tev->point.retprobe = true;
+        }
+
         pr_debug("Probe point found: %s+%lu\n", tev->point.symbol,
                  tev->point.offset);
 
diff --git a/tools/perf/util/symbol.c b/tools/perf/util/symbol.c
index 1a367734e016..b2f5ae97f33d 100644
--- a/tools/perf/util/symbol.c
+++ b/tools/perf/util/symbol.c
@@ -2268,6 +2268,9 @@ static int setup_list(struct strlist **list, const char *list_str,
 
 int symbol__init(void)
 {
+        if (symbol_conf.initialized)
+                return 0;
+
         elf_version(EV_CURRENT);
         if (symbol_conf.sort_by_name)
                 symbol_conf.priv_size += (sizeof(struct symbol_name_rb_node) -
@@ -2293,6 +2296,7 @@ int symbol__init(void)
                        symbol_conf.sym_list_str, "symbol") < 0)
                 goto out_free_comm_list;
 
+        symbol_conf.initialized = true;
         return 0;
 
 out_free_dso_list:
@@ -2304,11 +2308,14 @@ out_free_comm_list:
 
 void symbol__exit(void)
 {
+        if (!symbol_conf.initialized)
+                return;
         strlist__delete(symbol_conf.sym_list);
         strlist__delete(symbol_conf.dso_list);
         strlist__delete(symbol_conf.comm_list);
         vmlinux_path__exit();
         symbol_conf.sym_list = symbol_conf.dso_list = symbol_conf.comm_list = NULL;
+        symbol_conf.initialized = false;
 }
 
 int machines__create_kernel_maps(struct rb_root *self, pid_t pid)
diff --git a/tools/perf/util/symbol.h b/tools/perf/util/symbol.h
index b7a8da4af5a0..ea95c2756f05 100644
--- a/tools/perf/util/symbol.h
+++ b/tools/perf/util/symbol.h
@@ -69,7 +69,8 @@ struct symbol_conf {
                         show_nr_samples,
                         use_callchain,
                         exclude_other,
-                        show_cpu_utilization;
+                        show_cpu_utilization,
+                        initialized;
         const char      *vmlinux_name,
                         *source_prefix,
                         *field_sep;
diff --git a/tools/perf/util/trace-event-scripting.c b/tools/perf/util/trace-event-scripting.c
index 7ea983acfaea..f7af2fca965d 100644
--- a/tools/perf/util/trace-event-scripting.c
+++ b/tools/perf/util/trace-event-scripting.c
@@ -97,7 +97,7 @@ void setup_python_scripting(void)
         register_python_scripting(&python_scripting_unsupported_ops);
 }
 #else
-struct scripting_ops python_scripting_ops;
+extern struct scripting_ops python_scripting_ops;
 
 void setup_python_scripting(void)
 {
@@ -158,7 +158,7 @@ void setup_perl_scripting(void)
         register_perl_scripting(&perl_scripting_unsupported_ops);
 }
 #else
-struct scripting_ops perl_scripting_ops;
+extern struct scripting_ops perl_scripting_ops;
 
 void setup_perl_scripting(void)
 {
diff --git a/tools/perf/util/ui/browsers/annotate.c b/tools/perf/util/ui/browsers/annotate.c
index 55ff792459ac..a90273e63f4f 100644
--- a/tools/perf/util/ui/browsers/annotate.c
+++ b/tools/perf/util/ui/browsers/annotate.c
@@ -146,6 +146,7 @@ static int annotate_browser__run(struct annotate_browser *self,
                 return -1;
 
         newtFormAddHotKey(self->b.form, NEWT_KEY_LEFT);
+        newtFormAddHotKey(self->b.form, NEWT_KEY_RIGHT);
 
         nd = self->curr_hot;
         if (nd) {
@@ -178,7 +179,7 @@ static int annotate_browser__run(struct annotate_browser *self,
         }
 out:
         ui_browser__hide(&self->b);
-        return 0;
+        return es->u.key;
 }
 
 int hist_entry__tui_annotate(struct hist_entry *self)
diff --git a/tools/perf/util/ui/browsers/hists.c b/tools/perf/util/ui/browsers/hists.c
index dafdf6775d77..6866aa4c41e0 100644
--- a/tools/perf/util/ui/browsers/hists.c
+++ b/tools/perf/util/ui/browsers/hists.c
@@ -773,7 +773,7 @@ int hists__browse(struct hists *self, const char *helpline, const char *ev_name)
 
                         switch (key) {
                         case 'a':
-                                if (browser->selection->map == NULL &&
+                                if (browser->selection->map == NULL ||
                                     browser->selection->map->dso->annotate_warned)
                                         continue;
                                 goto do_annotate;