rpm/binutils
SHA256
1
0
Fork 0
forked from pool/binutils
Code Pull Requests Activity
binutils/binutils-pr29370-pre1.diff
Michael Matz 8d8118647c - Add binutils-pr29370.diff and their prerequisites 
binutils-pr29370-pre1.diff and binutils-pr29370-pre2.diff
  for PR29370, aka CVE-2022-38128 [bsc#1203016]

OBS-URL: https://build.opensuse.org/package/show/devel:gcc/binutils?expand=0&rev=418
2022-09-01 11:54:53 +00:00

533 lines
16 KiB
Diff
Raw Blame History

From 175b91507b83ad42607d2f6dadaf55b7b511bdbe Mon Sep 17 00:00:00 2001
From: Alan Modra <amodra@gmail.com>
Date: Wed, 20 Jul 2022 18:28:50 +0930
Subject: [PATCH] miscellaneous dwarf.c tidies
* dwarf.c: Leading and trailing whitespace fixes.
(free_abbrev_list): New function.
(free_all_abbrevs): Use the above. Free cu_abbrev_map here too.
(process_abbrev_set): Print actual section name on error.
(get_type_abbrev_from_form): Add overflow check.
(free_debug_memory): Don't free cu_abbrev_map here..
(process_debug_info): ..or here. Warn on another case of not
finding a neeeded abbrev.
---
binutils/dwarf.c | 216 +++++++++++++++++++++++------------------------
1 file changed, 106 insertions(+), 110 deletions(-)
Index: binutils-2.39/binutils/dwarf.c
===================================================================
--- binutils-2.39.orig/binutils/dwarf.c 2022-09-01 13:51:14.818699924 +0200
+++ binutils-2.39/binutils/dwarf.c 2022-09-01 13:51:32.090984628 +0200
@@ -806,7 +806,7 @@ fetch_indexed_value (dwarf_vma idx,
pointer_size = 4;
bias = 12;
}
-
+
dwarf_vma offset = idx * pointer_size;
/* Offsets are biased by the size of the section header
@@ -901,38 +901,41 @@ record_abbrev_list_for_cu (dwarf_vma sta
next_free_abbrev_map_entry ++;
}
-static void
-free_all_abbrevs (void)
+static abbrev_list *
+free_abbrev_list (abbrev_list *list)
{
- abbrev_list * list;
+ abbrev_entry *abbrv = list->first_abbrev;
- for (list = abbrev_lists; list != NULL;)
+ while (abbrv)
{
- abbrev_list * next = list->next;
- abbrev_entry * abbrv;
+ abbrev_attr *attr = abbrv->first_attr;
- for (abbrv = list->first_abbrev; abbrv != NULL;)
+ while (attr)
{
- abbrev_entry * next_abbrev = abbrv->next;
- abbrev_attr * attr;
-
- for (attr = abbrv->first_attr; attr;)
- {
- abbrev_attr *next_attr = attr->next;
-
- free (attr);
- attr = next_attr;
- }
-
- free (abbrv);
- abbrv = next_abbrev;
+ abbrev_attr *next_attr = attr->next;
+ free (attr);
+ attr = next_attr;
}
- free (list);
- list = next;
+ abbrev_entry *next_abbrev = abbrv->next;
+ free (abbrv);
+ abbrv = next_abbrev;
}
- abbrev_lists = NULL;
+ abbrev_list *next = list->next;
+ free (list);
+ return next;
+}
+
+static void
+free_all_abbrevs (void)
+{
+ while (abbrev_lists)
+ abbrev_lists = free_abbrev_list (abbrev_lists);
+
+ free (cu_abbrev_map);
+ cu_abbrev_map = NULL;
+ next_free_abbrev_map_entry = 0;
}
static abbrev_list *
@@ -978,7 +981,7 @@ find_abbrev_map_by_offset (dwarf_vma off
&& cu_abbrev_map[i].end > offset)
return cu_abbrev_map + i;
- return NULL;
+ return NULL;
}
static void
@@ -1119,7 +1122,7 @@ process_abbrev_set (struct dwarf_section
}
/* Report the missing single zero which ends the section. */
- error (_(".debug_abbrev section not zero terminated\n"));
+ error (_("%s section not zero terminated\n"), section->name);
return NULL;
}
@@ -1842,7 +1845,7 @@ fetch_alt_indirect_string (dwarf_vma off
dwarf_vmatoa ("x", offset));
return _("<offset is too big>");
}
-
+
static const char *
get_AT_name (unsigned long attribute)
{
@@ -2124,7 +2127,8 @@ get_type_abbrev_from_form (unsigned long
case DW_FORM_ref4:
case DW_FORM_ref8:
case DW_FORM_ref_udata:
- if (uvalue + cu_offset > (size_t) (cu_end - section->start))
+ if (uvalue + cu_offset < uvalue
+ || uvalue + cu_offset > (size_t) (cu_end - section->start))
{
warn (_("Unable to resolve ref form: uvalue %lx + cu_offset %lx > CU size %lx\n"),
uvalue, (long) cu_offset, (long) (cu_end - section->start));
@@ -2161,7 +2165,7 @@ get_type_abbrev_from_form (unsigned long
else
*map_return = NULL;
}
-
+
READ_ULEB (abbrev_number, data, section->start + section->size);
for (entry = map->list->first_abbrev; entry != NULL; entry = entry->next)
@@ -2750,10 +2754,10 @@ read_and_display_attr_value (unsigned lo
if (form == DW_FORM_loclistx)
{
if (dwo)
- {
- index = fetch_indexed_value (uvalue, loclists_dwo, 0);
- index += (offset_size == 8) ? 20 : 12;
- }
+ {
+ index = fetch_indexed_value (uvalue, loclists_dwo, 0);
+ index += (offset_size == 8) ? 20 : 12;
+ }
else if (debug_info_p == NULL)
{
index = fetch_indexed_value (uvalue, loclists, 0);
@@ -2771,21 +2775,21 @@ read_and_display_attr_value (unsigned lo
else if (form == DW_FORM_rnglistx)
{
if (dwo)
- {
- index = fetch_indexed_value (uvalue, rnglists_dwo, 0);
- index += (offset_size == 8) ? 20 : 12;
- }
+ {
+ index = fetch_indexed_value (uvalue, rnglists_dwo, 0);
+ index += (offset_size == 8) ? 20 : 12;
+ }
else
- {
- if (debug_info_p == NULL)
- base = 0;
- else
- base = debug_info_p->rnglists_base;
- /* We do not have a cached value this time, so we perform the
- computation manually. */
- index = fetch_indexed_value (uvalue, rnglists, base);
- index += base;
- }
+ {
+ if (debug_info_p == NULL)
+ base = 0;
+ else
+ base = debug_info_p->rnglists_base;
+ /* We do not have a cached value this time, so we perform the
+ computation manually. */
+ index = fetch_indexed_value (uvalue, rnglists, base);
+ index += base;
+ }
}
else
{
@@ -2811,7 +2815,7 @@ read_and_display_attr_value (unsigned lo
if (!do_loc)
printf ("%c<0x%s>", delimiter, dwarf_vmatoa ("x", uvalue + cu_offset));
break;
-
+
default:
warn (_("Unrecognized form: 0x%lx\n"), form);
/* What to do? Consume a byte maybe? */
@@ -2836,9 +2840,9 @@ read_and_display_attr_value (unsigned lo
case DW_AT_rnglists_base:
if (debug_info_p->rnglists_base)
warn (_("CU @ 0x%s has multiple rnglists_base values (0x%s and 0x%s)"),
- dwarf_vmatoa ("x", debug_info_p->cu_offset),
- dwarf_vmatoa ("x", debug_info_p->rnglists_base),
- dwarf_vmatoa ("x", uvalue));
+ dwarf_vmatoa ("x", debug_info_p->cu_offset),
+ dwarf_vmatoa ("x", debug_info_p->rnglists_base),
+ dwarf_vmatoa ("x", uvalue));
debug_info_p->rnglists_base = uvalue;
break;
case DW_AT_str_offsets_base:
@@ -2988,7 +2992,7 @@ read_and_display_attr_value (unsigned lo
case DW_FORM_strx3:
case DW_FORM_strx4:
add_dwo_name (fetch_indexed_string (uvalue, this_set, offset_size, false,
- debug_info_p->str_offsets_base),
+ debug_info_p->str_offsets_base),
cu_offset);
break;
case DW_FORM_string:
@@ -3022,7 +3026,7 @@ read_and_display_attr_value (unsigned lo
case DW_FORM_strx3:
case DW_FORM_strx4:
add_dwo_dir (fetch_indexed_string (uvalue, this_set, offset_size, false,
- debug_info_p->str_offsets_base),
+ debug_info_p->str_offsets_base),
cu_offset);
break;
case DW_FORM_string:
@@ -3653,11 +3657,8 @@ process_debug_info (struct dwarf_section
introduce (section, false);
free_all_abbrevs ();
- free (cu_abbrev_map);
- cu_abbrev_map = NULL;
- next_free_abbrev_map_entry = 0;
- /* In order to be able to resolve DW_FORM_ref_attr forms we need
+ /* In order to be able to resolve DW_FORM_ref_addr forms we need
to load *all* of the abbrevs for all CUs in this .debug_info
section. This does effectively mean that we (partially) read
every CU header twice. */
@@ -4028,12 +4029,11 @@ process_debug_info (struct dwarf_section
/* Scan through the abbreviation list until we reach the
correct entry. */
- if (list == NULL)
- continue;
-
- for (entry = list->first_abbrev; entry != NULL; entry = entry->next)
- if (entry->number == abbrev_number)
- break;
+ entry = NULL;
+ if (list != NULL)
+ for (entry = list->first_abbrev; entry != NULL; entry = entry->next)
+ if (entry->number == abbrev_number)
+ break;
if (entry == NULL)
{
@@ -4057,7 +4057,7 @@ process_debug_info (struct dwarf_section
break;
case DW_TAG_compile_unit:
case DW_TAG_skeleton_unit:
- need_base_address = 1;
+ need_base_address = 1;
need_dwo_info = do_loc;
break;
case DW_TAG_entry_point:
@@ -4440,7 +4440,7 @@ display_debug_sup (struct dwarf_section
SAFE_BYTE_GET_AND_INC (is_supplementary, start, 1, end);
if (is_supplementary != 0 && is_supplementary != 1)
- warn (_("corrupt .debug_sup section: is_supplementary not 0 or 1\n"));
+ warn (_("corrupt .debug_sup section: is_supplementary not 0 or 1\n"));
sup_filename = start;
if (is_supplementary && sup_filename[0] != 0)
@@ -5619,7 +5619,7 @@ display_debug_lines_decoded (struct dwar
printf ("%s %11d %#18" DWARF_VMA_FMT "x",
newFileName, state_machine_regs.line,
state_machine_regs.address);
- }
+ }
else
{
if (xop == -DW_LNE_end_sequence)
@@ -6073,7 +6073,7 @@ display_debug_macro (struct dwarf_sectio
load_debug_section_with_follow (str, file);
load_debug_section_with_follow (line, file);
load_debug_section_with_follow (str_index, file);
-
+
introduce (section, false);
while (curr < end)
@@ -6525,7 +6525,7 @@ display_loc_list (struct dwarf_section *
/* Check base address specifiers. */
if (is_max_address (begin, pointer_size)
- && !is_max_address (end, pointer_size))
+ && !is_max_address (end, pointer_size))
{
base_address = end;
print_dwarf_vma (begin, pointer_size);
@@ -6703,7 +6703,7 @@ display_loclists_list (struct dwarf_sect
case DW_LLE_default_location:
begin = end = 0;
break;
-
+
case DW_LLE_offset_pair:
READ_ULEB (begin, start, section_end);
begin += base_address;
@@ -6999,7 +6999,7 @@ display_offset_entry_loclists (struct dw
unsigned char * start = section->start;
unsigned char * const end = start + section->size;
- introduce (section, false);
+ introduce (section, false);
do
{
@@ -7048,14 +7048,14 @@ display_offset_entry_loclists (struct dw
section->name, segment_selector_size);
return 0;
}
-
+
if (offset_entry_count == 0)
{
warn (_("The %s section contains a table without offset\n"),
section->name);
return 0;
}
-
+
printf (_("\n Offset Entries starting at 0x%lx:\n"),
(long)(start - section->start));
@@ -8217,7 +8217,7 @@ display_debug_rnglists (struct dwarf_sec
start = display_debug_rnglists_list
(start, end, address_size, offset, 0, offset_size);
if (start >= end)
- break;
+ break;
}
start = end;
@@ -8335,12 +8335,12 @@ display_debug_ranges (struct dwarf_secti
next = section_begin + offset + debug_info_p->rnglists_base;
/* If multiple DWARF entities reference the same range then we will
- have multiple entries in the `range_entries' list for the same
- offset. Thanks to the sort above these will all be consecutive in
- the `range_entries' list, so we can easily ignore duplicates
- here. */
+ have multiple entries in the `range_entries' list for the same
+ offset. Thanks to the sort above these will all be consecutive in
+ the `range_entries' list, so we can easily ignore duplicates
+ here. */
if (i > 0 && last_offset == offset)
- continue;
+ continue;
last_offset = offset;
if (dwarf_check != 0 && i > 0)
@@ -10274,7 +10274,7 @@ display_debug_names (struct dwarf_sectio
printf (_("Out of %lu items there are %zu bucket clashes"
" (longest of %zu entries).\n"),
(unsigned long) name_count, hash_clash_count, longest_clash);
-
+
if (name_count != buckets_filled + hash_clash_count)
warn (_("The name_count (%lu) is not the same as the used bucket_count (%lu) + the hash clash count (%lu)"),
(unsigned long) name_count,
@@ -10378,7 +10378,7 @@ display_debug_names (struct dwarf_sectio
break;
if (tagno >= 0)
printf ("%s<%lu>",
- (tagno == 0 && second_abbrev_tag == 0 ? " " : "\n\t"),
+ (tagno == 0 && second_abbrev_tag == 0 ? " " : "\n\t"),
(unsigned long) abbrev_tag);
for (entry = abbrev_lookup;
@@ -10907,7 +10907,7 @@ process_cu_tu_index (struct dwarf_sectio
Check for integer overflow (can occur when size_t is 32-bit)
with overlarge ncols or nused values. */
if (nused == -1u
- || _mul_overflow ((size_t) ncols, 4, &temp)
+ || _mul_overflow ((size_t) ncols, 4, &temp)
|| _mul_overflow ((size_t) nused + 1, temp, &total)
|| total > (size_t) (limit - ppool))
{
@@ -10915,7 +10915,7 @@ process_cu_tu_index (struct dwarf_sectio
section->name);
return 0;
}
-
+
if (do_display)
{
printf (_(" Offset table\n"));
@@ -11419,8 +11419,8 @@ add_separate_debug_file (const char * fi
static bool
debuginfod_fetch_separate_debug_info (struct dwarf_section * section,
- char ** filename,
- void * file)
+ char ** filename,
+ void * file)
{
size_t build_id_len;
unsigned char * build_id;
@@ -11438,14 +11438,14 @@ debuginfod_fetch_separate_debug_info (st
filelen = strnlen ((const char *)section->start, section->size);
if (filelen == section->size)
- /* Corrupt debugaltlink. */
- return false;
+ /* Corrupt debugaltlink. */
+ return false;
build_id = section->start + filelen + 1;
build_id_len = section->size - (filelen + 1);
if (build_id_len == 0)
- return false;
+ return false;
}
else
return false;
@@ -11457,25 +11457,25 @@ debuginfod_fetch_separate_debug_info (st
client = debuginfod_begin ();
if (client == NULL)
- return false;
+ return false;
/* Query debuginfod servers for the target file. If found its path
- will be stored in filename. */
+ will be stored in filename. */
fd = debuginfod_find_debuginfo (client, build_id, build_id_len, filename);
debuginfod_end (client);
/* Only free build_id if we allocated space for a hex string
- in get_build_id (). */
+ in get_build_id (). */
if (build_id_len == 0)
- free (build_id);
+ free (build_id);
if (fd >= 0)
- {
- /* File successfully retrieved. Close fd since we want to
- use open_debug_file () on filename instead. */
- close (fd);
- return true;
- }
+ {
+ /* File successfully retrieved. Close fd since we want to
+ use open_debug_file () on filename instead. */
+ close (fd);
+ return true;
+ }
}
return false;
@@ -11488,7 +11488,7 @@ load_separate_debug_info (const char *
parse_func_type parse_func,
check_func_type check_func,
void * func_data,
- void * file ATTRIBUTE_UNUSED)
+ void * file ATTRIBUTE_UNUSED)
{
const char * separate_filename;
char * debug_filename;
@@ -11604,11 +11604,11 @@ load_separate_debug_info (const char *
& tmp_filename,
file))
{
- /* File successfully downloaded from server, replace
- debug_filename with the file's path. */
- free (debug_filename);
- debug_filename = tmp_filename;
- goto found;
+ /* File successfully downloaded from server, replace
+ debug_filename with the file's path. */
+ free (debug_filename);
+ debug_filename = tmp_filename;
+ goto found;
}
}
#endif
@@ -11775,12 +11775,12 @@ load_build_id_debug_file (const char * m
/* In theory we should extract the contents of the section into
a note structure and then check the fields. For now though
just use hard coded offsets instead:
-
+
Field Bytes Contents
NSize 0...3 4
DSize 4...7 8+
Type 8..11 3 (NT_GNU_BUILD_ID)
- Name 12.15 GNU\0
+ Name 12.15 GNU\0
Data 16.... */
/* FIXME: Check the name size, name and type fields. */
@@ -11792,7 +11792,7 @@ load_build_id_debug_file (const char * m
warn (_(".note.gnu.build-id data size is too small\n"));
return;
}
-
+
if (build_id_size > (section->size - 16))
{
warn (_(".note.gnu.build-id data size is too bug\n"));
@@ -12088,10 +12088,6 @@ free_debug_memory (void)
free_all_abbrevs ();
- free (cu_abbrev_map);
- cu_abbrev_map = NULL;
- next_free_abbrev_map_entry = 0;
-
free (shndx_pool);
shndx_pool = NULL;
shndx_pool_size = 0;
View Git Blame Copy Permalink