This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
[PATCH 5/9] [libbacktrace] Unify function name preference handling
- From: Tom de Vries <tdevries at suse dot de>
- To: gcc-patches at gcc dot gnu dot org
- Cc: Ian Lance Taylor <ian at airs dot com>
- Date: Tue, 11 Dec 2018 11:14:07 +0100
- Subject: [PATCH 5/9] [libbacktrace] Unify function name preference handling
- References: <20181211101411.7067-1-tdevries@suse.de>
Both read_function_entry and read_referenced_name implement a priority scheme
for names. The priorities are:
- 1st: DW_AT_linkage_name
- 2nd: Name from DW_AT_abstract_origin or DW_AT_specification
- 3rd: DW_AT_name.
Ensure both functions fully adhere to it.
2018-11-21 Tom de Vries <tdevries@suse.de>
* dwarf.c (read_referenced_name): Don't allow DW_AT_name to override any
name.
(read_function_entry): Same. Don't allow name found via
DW_AT_abstract_origin or case DW_AT_specification to override linkage
name.
---
libbacktrace/dwarf.c | 33 +++++++++++++++++++++++----------
1 file changed, 23 insertions(+), 10 deletions(-)
diff --git a/libbacktrace/dwarf.c b/libbacktrace/dwarf.c
index 341b84605c0..2483295beb4 100644
--- a/libbacktrace/dwarf.c
+++ b/libbacktrace/dwarf.c
@@ -2175,18 +2175,25 @@ read_referenced_name (struct dwarf_data *ddata, struct unit *u,
switch (abbrev->attrs[i].name)
{
case DW_AT_name:
- /* We prefer the linkage name if get one. */
+ /* Third name preference: don't override. A name we found in some
+ other way, will normally be more useful -- e.g., this name is
+ normally not mangled. */
+ if (ret != NULL)
+ break;
if (val.encoding == ATTR_VAL_STRING)
ret = val.u.string;
break;
case DW_AT_linkage_name:
case DW_AT_MIPS_linkage_name:
+ /* First name preference: override all. */
if (val.encoding == ATTR_VAL_STRING)
return val.u.string;
break;
case DW_AT_specification:
+ /* Second name preference: override DW_AT_name, don't override
+ DW_AT_linkage_name. */
if (abbrev->attrs[i].form == DW_FORM_ref_addr
|| abbrev->attrs[i].form == DW_FORM_ref_sig8)
{
@@ -2339,6 +2346,7 @@ read_function_entry (struct backtrace_state *state, struct dwarf_data *ddata,
int highpc_is_relative;
uint64_t ranges;
int have_ranges;
+ int have_linkage_name;
code = read_uleb128 (unit_buf);
if (code == 0)
@@ -2375,6 +2383,7 @@ read_function_entry (struct backtrace_state *state, struct dwarf_data *ddata,
highpc_is_relative = 0;
ranges = 0;
have_ranges = 0;
+ have_linkage_name = 0;
for (i = 0; i < abbrev->num_attrs; ++i)
{
struct attr_val val;
@@ -2423,6 +2432,10 @@ read_function_entry (struct backtrace_state *state, struct dwarf_data *ddata,
case DW_AT_abstract_origin:
case DW_AT_specification:
+ /* Second name preference: override DW_AT_name, don't override
+ DW_AT_linkage_name. */
+ if (have_linkage_name)
+ break;
if (abbrev->attrs[i].form == DW_FORM_ref_addr
|| abbrev->attrs[i].form == DW_FORM_ref_sig8)
{
@@ -2444,21 +2457,21 @@ read_function_entry (struct backtrace_state *state, struct dwarf_data *ddata,
break;
case DW_AT_name:
+ /* Third name preference: don't override. */
+ if (function->name != NULL)
+ break;
if (val.encoding == ATTR_VAL_STRING)
- {
- /* Don't override a name we found in some other
- way, as it will normally be more
- useful--e.g., this name is normally not
- mangled. */
- if (function->name == NULL)
- function->name = val.u.string;
- }
+ function->name = val.u.string;
break;
case DW_AT_linkage_name:
case DW_AT_MIPS_linkage_name:
+ /* First name preference: override all. */
if (val.encoding == ATTR_VAL_STRING)
- function->name = val.u.string;
+ {
+ function->name = val.u.string;
+ have_linkage_name = 1;
+ }
break;
case DW_AT_low_pc:
--
2.16.4