perf probe: Add --range option to show a variable's location range

It is not easy for users to get the accurate byte offset or the line
number where a local variable can be probed.

With '--range' option, local variables in the scope of the probe point
are showed with a byte offset range, and can be added according to this
range information.

For example, there are some variables in the function
generic_perform_write():

  <generic_perform_write@mm/filemap.c:0>
  0  ssize_t generic_perform_write(struct file *file,
  1                                 struct iov_iter *i, loff_t pos)
  2  {
  3          struct address_space *mapping = file->f_mapping;
  4          const struct address_space_operations *a_ops = mapping->a_ops;
  ...
  42                 status = a_ops->write_begin(file, mapping, pos, bytes, flags,
                                               &page, &fsdata);
  44                 if (unlikely(status < 0))

But we fail when we try to probe the variable 'a_ops' at line 42 or 44.

  $ perf probe --add 'generic_perform_write:42 a_ops'
  Failed to find the location of a_ops at this address.
    Perhaps, it has been optimized out.

This is because the source code do not match the assembly, so a variable
may not be available in the source code line where it appears.

After this patch, we can lookup the accurate byte offset range of a
variable, 'INV' indicates that this variable is not valid at the given
point, but available in the scope:

  $ perf probe --vars 'generic_perform_write:42' --range
  Available variables at generic_perform_write:42
    @<generic_perform_write+141>
       [INV] ssize_t written @<generic_perform_write+[324-331]>
       [INV] struct address_space_operations*        a_ops   @<generic_perform_write+[55-61,170-176,223-246]>
       [VAL] (unknown_type)  fsdata  @<generic_perform_write+[70-307,346-411]>
       [VAL] loff_t  pos     @<generic_perform_write+[0-286,286-336,346-411]>
       [VAL] long int        status  @<generic_perform_write+[83-342,346-411]>
       [VAL] long unsigned int       bytes   @<generic_perform_write+[122-311,320-338,346-403,403-411]>
       [VAL] struct address_space*   mapping @<generic_perform_write+[35-344,346-411]>
       [VAL] struct iov_iter*        i       @<generic_perform_write+[0-340,346-411]>
       [VAL] struct page*    page    @<generic_perform_write+[70-307,346-411]>

Then it is more clear for us to add a probe with this variable:

  $ perf probe --add 'generic_perform_write+170 a_ops'
  Added new event:
    probe:generic_perform_write (on generic_perform_write+170 with a_ops)

Signed-off-by: He Kuang <hekuang@huawei.com>
Acked-by: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/r/1431336304-16863-2-git-send-email-hekuang@huawei.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>

tools/perf/builtin-probe.c

@@ -372,6 +372,8 @@ __cmd_probe(int argc, const char **argv, const char *prefix __maybe_unused)
 		     "Show accessible variables on PROBEDEF", opt_show_vars),
 	OPT_BOOLEAN('\0', "externs", &probe_conf.show_ext_vars,
 		    "Show external variables too (with --vars only)"),
+	OPT_BOOLEAN('\0', "range", &probe_conf.show_location_range,
+		"Show variables location range in scope (with --vars only)"),
 	OPT_STRING('k', "vmlinux", &symbol_conf.vmlinux_name,
 		   "file", "vmlinux pathname"),
 	OPT_STRING('s', "source", &symbol_conf.source_prefix,

tools/perf/util/dwarf-aux.c

@@ -913,3 +913,124 @@ int die_get_varname(Dwarf_Die *vr_die, struct strbuf *buf)
 	return 0;
 }
 
+/**
+ * die_get_var_innermost_scope - Get innermost scope range of given variable DIE
+ * @sp_die: a subprogram DIE
+ * @vr_die: a variable DIE
+ * @buf: a strbuf for variable byte offset range
+ *
+ * Get the innermost scope range of @vr_die and stores it in @buf as
+ * "@<function_name+[NN-NN,NN-NN]>".
+ */
+static int die_get_var_innermost_scope(Dwarf_Die *sp_die, Dwarf_Die *vr_die,
+				struct strbuf *buf)
+{
+	Dwarf_Die *scopes;
+	int count;
+	size_t offset = 0;
+	Dwarf_Addr base;
+	Dwarf_Addr start, end;
+	Dwarf_Addr entry;
+	int ret;
+	bool first = true;
+	const char *name;
+
+	ret = dwarf_entrypc(sp_die, &entry);
+	if (ret)
+		return ret;
+
+	name = dwarf_diename(sp_die);
+	if (!name)
+		return -ENOENT;
+
+	count = dwarf_getscopes_die(vr_die, &scopes);
+
+	/* (*SCOPES)[1] is the DIE for the scope containing that scope */
+	if (count <= 1) {
+		ret = -EINVAL;
+		goto out;
+	}
+
+	while ((offset = dwarf_ranges(&scopes[1], offset, &base,
+				&start, &end)) > 0) {
+		start -= entry;
+		end -= entry;
+
+		if (first) {
+			strbuf_addf(buf, "@<%s+[%lu-%lu",
+				name, start, end);
+			first = false;
+		} else {
+			strbuf_addf(buf, ",%lu-%lu",
+				start, end);
+		}
+	}
+
+	if (!first)
+		strbuf_addf(buf, "]>");
+
+out:
+	free(scopes);
+	return ret;
+}
+
+/**
+ * die_get_var_range - Get byte offset range of given variable DIE
+ * @sp_die: a subprogram DIE
+ * @vr_die: a variable DIE
+ * @buf: a strbuf for type and variable name and byte offset range
+ *
+ * Get the byte offset range of @vr_die and stores it in @buf as
+ * "@<function_name+[NN-NN,NN-NN]>".
+ */
+int die_get_var_range(Dwarf_Die *sp_die, Dwarf_Die *vr_die, struct strbuf *buf)
+{
+	int ret = 0;
+	Dwarf_Addr base;
+	Dwarf_Addr start, end;
+	Dwarf_Addr entry;
+	Dwarf_Op *op;
+	size_t nops;
+	size_t offset = 0;
+	Dwarf_Attribute attr;
+	bool first = true;
+	const char *name;
+
+	ret = dwarf_entrypc(sp_die, &entry);
+	if (ret)
+		return ret;
+
+	name = dwarf_diename(sp_die);
+	if (!name)
+		return -ENOENT;
+
+	if (dwarf_attr(vr_die, DW_AT_location, &attr) == NULL)
+		return -EINVAL;
+
+	while ((offset = dwarf_getlocations(
+				&attr, offset, &base,
+				&start, &end, &op, &nops)) > 0) {
+		if (start == 0) {
+			/* Single Location Descriptions */
+			ret = die_get_var_innermost_scope(sp_die, vr_die, buf);
+			return ret;
+		}
+
+		/* Location Lists */
+		start -= entry;
+		end -= entry;
+		if (first) {
+			strbuf_addf(buf, "@<%s+[%lu-%lu",
+				name, start, end);
+			first = false;
+		} else {
+			strbuf_addf(buf, ",%lu-%lu",
+				start, end);
+		}
+	}
+
+	if (!first)
+		strbuf_addf(buf, "]>");
+
+	return ret;
+}

tools/perf/util/dwarf-aux.h

@@ -121,4 +121,6 @@ extern int die_get_typename(Dwarf_Die *vr_die, struct strbuf *buf);
 
 /* Get the name and type of given variable DIE, stored as "type\tname" */
 extern int die_get_varname(Dwarf_Die *vr_die, struct strbuf *buf);
+extern int die_get_var_range(Dwarf_Die *sp_die, Dwarf_Die *vr_die,
+			struct strbuf *buf);
 #endif

tools/perf/util/probe-event.h

@@ -9,6 +9,7 @@
 /* Probe related configurations */
 struct probe_conf {
 	bool	show_ext_vars;
+	bool	show_location_range;
 	bool	force_add;
 	bool	no_inlines;
 	int	max_probes;

tools/perf/util/probe-finder.c

@@ -177,7 +177,7 @@ static int convert_variable_location(Dwarf_Die *vr_die, Dwarf_Addr addr,
 	Dwarf_Word offs = 0;
 	bool ref = false;
 	const char *regs;
-	int ret;
+	int ret, ret2 = 0;
 
 	if (dwarf_attr(vr_die, DW_AT_external, &attr) != NULL)
 		goto static_var;
@@ -187,9 +187,19 @@ static int convert_variable_location(Dwarf_Die *vr_die, Dwarf_Addr addr,
 		return -EINVAL;	/* Broken DIE ? */
 	if (dwarf_getlocation_addr(&attr, addr, &op, &nops, 1) <= 0) {
 		ret = dwarf_entrypc(sp_die, &tmp);
-		if (ret || addr != tmp ||
-		    dwarf_tag(vr_die) != DW_TAG_formal_parameter ||
-		    dwarf_highpc(sp_die, &tmp))
+		if (ret)
+			return -ENOENT;
+
+		if (probe_conf.show_location_range &&
+			(dwarf_tag(vr_die) == DW_TAG_variable)) {
+			ret2 = -ERANGE;
+		} else if (addr != tmp ||
+			dwarf_tag(vr_die) != DW_TAG_formal_parameter) {
+			return -ENOENT;
+		}
+
+		ret = dwarf_highpc(sp_die, &tmp);
+		if (ret)
 			return -ENOENT;
 		/*
 		 * This is fuzzed by fentry mcount. We try to find the
@@ -210,7 +220,7 @@ found:
 	if (op->atom == DW_OP_addr) {
 static_var:
 		if (!tvar)
-			return 0;
+			return ret2;
 		/* Static variables on memory (not stack), make @varname */
 		ret = strlen(dwarf_diename(vr_die));
 		tvar->value = zalloc(ret + 2);
@@ -220,7 +230,7 @@ static_var:
 		tvar->ref = alloc_trace_arg_ref((long)offs);
 		if (tvar->ref == NULL)
 			return -ENOMEM;
-		return 0;
+		return ret2;
 	}
 
 	/* If this is based on frame buffer, set the offset */
@@ -250,14 +260,14 @@ static_var:
 	}
 
 	if (!tvar)
-		return 0;
+		return ret2;
 
 	regs = get_arch_regstr(regn);
 	if (!regs) {
 		/* This should be a bug in DWARF or this tool */
 		pr_warning("Mapping for the register number %u "
 			   "missing on this architecture.\n", regn);
-		return -ERANGE;
+		return -ENOTSUP;
 	}
 
 	tvar->value = strdup(regs);
@@ -269,7 +279,7 @@ static_var:
 		if (tvar->ref == NULL)
 			return -ENOMEM;
 	}
-	return 0;
+	return ret2;
 }
 
 #define BYTES_TO_BITS(nb)	((nb) * BITS_PER_LONG / sizeof(long))
@@ -1270,13 +1280,34 @@ static int collect_variables_cb(Dwarf_Die *die_mem, void *data)
 		ret = convert_variable_location(die_mem, af->pf.addr,
 						af->pf.fb_ops, &af->pf.sp_die,
 						NULL);
-		if (ret == 0) {
+		if (ret == 0 || ret == -ERANGE) {
+			int ret2;
+			bool externs = !af->child;
 			struct strbuf buf;
 
 			strbuf_init(&buf, 64);
-			ret = die_get_varname(die_mem, &buf);
-			pr_debug2("Add new var: %s\n", buf.buf);
-			if (ret == 0) {
+
+			if (probe_conf.show_location_range) {
+				if (!externs) {
+					if (ret)
+						strbuf_addf(&buf, "[INV]\t");
+					else
+						strbuf_addf(&buf, "[VAL]\t");
+				} else
+					strbuf_addf(&buf, "[EXT]\t");
+			}
+
+			ret2 = die_get_varname(die_mem, &buf);
+
+			if (!ret2 && probe_conf.show_location_range &&
+				!externs) {
+				strbuf_addf(&buf, "\t");
+				ret2 = die_get_var_range(&af->pf.sp_die,
+							die_mem, &buf);
+			}
+
+			pr_debug("Add new var: %s\n", buf.buf);
+			if (ret2 == 0) {
 				strlist__add(vl->vars,
 					strbuf_detach(&buf, NULL));
 			}