diff --git a/SOURCES/binutils.unicode.patch b/SOURCES/binutils.unicode.patch new file mode 100644 index 0000000..1a55dad --- /dev/null +++ b/SOURCES/binutils.unicode.patch @@ -0,0 +1,1932 @@ +diff -rup binutils.orig/binutils/NEWS binutils-2.30/binutils/NEWS +--- binutils.orig/binutils/NEWS 2021-10-20 12:39:41.715008273 +0100 ++++ binutils-2.30/binutils/NEWS 2021-10-20 12:41:30.895327968 +0100 +@@ -1,5 +1,14 @@ + -*- text -*- + ++* Tools which display names or strings (readelf, strings, nm, objdump) ++ have a new command line option which controls how unicode characters are ++ handled. By default they are treated as normal for the tool. Using ++ --unicode=show will display them according to the current locale. ++ Using --unicode=hex will display them as hex byte values, whilst ++ --unicode=escape will display them as escape sequences. In addition ++ using --unicode=highlight will display them as unicode escape sequences ++ highlighted in red (if supported by the output device). ++ + * Add support for disassembling netronome Flow Processor (NFP) firmware files. + + Changes in 2.30: +diff -rup binutils.orig/binutils/doc/binutils.texi binutils-2.30/binutils/doc/binutils.texi +--- binutils.orig/binutils/doc/binutils.texi 2021-10-20 12:39:41.701008360 +0100 ++++ binutils-2.30/binutils/doc/binutils.texi 2021-10-20 13:00:11.801327501 +0100 +@@ -762,6 +762,7 @@ nm [@option{-A}|@option{-o}|@option{--pr + [@option{-D}|@option{--dynamic}] [@option{-f}@var{format}|@option{--format=}@var{format}] + [@option{-g}|@option{--extern-only}] [@option{-h}|@option{--help}] + [@option{-l}|@option{--line-numbers}] [@option{--inlines}] ++ [@option{-U} @var{method}] [@option{--unicode=}@var{method}] + [@option{-n}|@option{-v}|@option{--numeric-sort}] + [@option{-P}|@option{--portability}] [@option{-p}|@option{--no-sort}] + [@option{-r}|@option{--reverse-sort}] [@option{-S}|@option{--print-size}] +@@ -1026,6 +1027,21 @@ Use @var{radix} as the radix for printin + @cindex undefined symbols + Display only undefined symbols (those external to each object file). + ++@item -U @var{[d|i|l|e|x|h]} ++@itemx --unicode=@var{[default|invalid|locale|escape|hex|highlight]} ++Controls the display of UTF-8 encoded mulibyte characters in strings. ++The default (@option{--unicode=default}) is to give them no special ++treatment. The @option{--unicode=locale} option displays the sequence ++in the current locale, which may or may not support them. The options ++@option{--unicode=hex} and @option{--unicode=invalid} display them as ++hex byte sequences enclosed by either angle brackets or curly braces. ++ ++The @option{--unicode=escape} option displays them as escape sequences ++(@var{\uxxxx}) and the @option{--unicode=highlight} option displays ++them as escape sequences highlighted in red (if supported by the ++output device). The colouring is intended to draw attention to the ++presence of unicode sequences where they might not be expected. ++ + @item -V + @itemx --version + Show the version number of @command{nm} and exit. +@@ -2093,6 +2109,7 @@ objdump [@option{-a}|@option{--archive-h + [@option{--prefix=}@var{prefix}] + [@option{--prefix-strip=}@var{level}] + [@option{--insn-width=}@var{width}] ++ [@option{-U} @var{method}] [@option{--unicode=}@var{method}] + [@option{-V}|@option{--version}] + [@option{-H}|@option{--help}] + @var{objfile}@dots{} +@@ -2697,6 +2714,21 @@ When displaying symbols include those wh + special in some way and which would not normally be of interest to the + user. + ++@item -U @var{[d|i|l|e|x|h]} ++@itemx --unicode=@var{[default|invalid|locale|escape|hex|highlight]} ++Controls the display of UTF-8 encoded mulibyte characters in strings. ++The default (@option{--unicode=default}) is to give them no special ++treatment. The @option{--unicode=locale} option displays the sequence ++in the current locale, which may or may not support them. The options ++@option{--unicode=hex} and @option{--unicode=invalid} display them as ++hex byte sequences enclosed by either angle brackets or curly braces. ++ ++The @option{--unicode=escape} option displays them as escape sequences ++(@var{\uxxxx}) and the @option{--unicode=highlight} option displays ++them as escape sequences highlighted in red (if supported by the ++output device). The colouring is intended to draw attention to the ++presence of unicode sequences where they might not be expected. ++ + @item -V + @itemx --version + Print the version number of @command{objdump} and exit. +@@ -2952,6 +2984,7 @@ strings [@option{-afovV}] [@option{-}@va + [@option{-n} @var{min-len}] [@option{--bytes=}@var{min-len}] + [@option{-t} @var{radix}] [@option{--radix=}@var{radix}] + [@option{-e} @var{encoding}] [@option{--encoding=}@var{encoding}] ++ [@option{-U} @var{method}] [@option{--unicode=}@var{method}] + [@option{-}] [@option{--all}] [@option{--print-file-name}] + [@option{-T} @var{bfdname}] [@option{--target=}@var{bfdname}] + [@option{-w}] [@option{--include-all-whitespace}] +@@ -3043,6 +3076,28 @@ single-8-bit-byte characters, @samp{b} = + littleendian. Useful for finding wide character strings. (@samp{l} + and @samp{b} apply to, for example, Unicode UTF-16/UCS-2 encodings). + ++@item -U @var{[d|i|l|e|x|h]} ++@itemx --unicode=@var{[default|invalid|locale|escape|hex|highlight]} ++Controls the display of UTF-8 encoded mulibyte characters in strings. ++The default (@option{--unicode=default}) is to give them no special ++treatment, and instead rely upon the setting of the ++@option{--encoding} option. The other values for this option ++automatically enable @option{--encoding=S}. ++ ++The @option{--unicode=invalid} option treats them as non-graphic ++characters and hence not part of a valid string. All the remaining ++options treat them as valid string characters. ++ ++The @option{--unicode=locale} option displays them in the current ++locale, which may or may not support UTF-8 encoding. The ++@option{--unicode=hex} option displays them as hex byte sequences ++enclosed between @var{<>} characters. The @option{--unicode=escape} ++option displays them as escape sequences (@var{\uxxxx}) and the ++@option{--unicode=highlight} option displays them as escape sequences ++highlighted in red (if supported by the output device). The colouring ++is intended to draw attention to the presence of unicode sequences ++where they might not be expected. ++ + @item -T @var{bfdname} + @itemx --target=@var{bfdname} + @cindex object code format +@@ -4603,6 +4658,7 @@ readelf [@option{-a}|@option{--all}] + [@option{-v}|@option{--version}] + [@option{-W}|@option{--wide}] + [@option{-H}|@option{--help}] ++ [@option{-U} @var{method}|@option{--unicode=}@var{method}] + @var{elffile}@dots{} + @c man end + @end smallexample +@@ -4692,6 +4748,28 @@ Displays the entries in dynamic symbol t + has one. The output format is the same as the format used by the + @option{--syms} option. + ++@item -U @var{[d|i|l|e|x|h]} ++@itemx --unicode=[default|invalid|locale|escape|hex|highlight] ++Controls the display of non-ASCII characters in identifier names. ++The default (@option{--unicode=locale} or @option{--unicode=default}) is ++to treat them as multibyte characters and display them in the current ++locale. All other versions of this option treat the bytes as UTF-8 ++encoded values and attempt to interpret them. If they cannot be ++interpreted or if the @option{--unicode=invalid} option is used then ++they are displayed as a sequence of hex bytes, encloses in curly ++parethesis characters. ++ ++Using the @option{--unicode=escape} option will display the characters ++as as unicode escape sequences (@var{\uxxxx}). Using the ++@option{--unicode=hex} will display the characters as hex byte ++sequences enclosed between angle brackets. ++ ++Using the @option{--unicode=highlight} will display the characters as ++unicode escape sequences but it will also highlighted them in red, ++assuming that colouring is supported by the output device. The ++colouring is intended to draw attention to the presence of unicode ++sequences when they might not be expected. ++ + @item -e + @itemx --headers + Display all the headers in the file. Equivalent to @option{-h -l -S}. +diff -rup binutils.orig/binutils/nm.c binutils-2.30/binutils/nm.c +--- binutils.orig/binutils/nm.c 2021-10-20 12:39:41.681008485 +0100 ++++ binutils-2.30/binutils/nm.c 2021-10-20 12:52:26.731237376 +0100 +@@ -38,6 +38,11 @@ + #include "bucomm.h" + #include "plugin-api.h" + #include "plugin.h" ++#include "safe-ctype.h" ++ ++#ifndef streq ++#define streq(a,b) (strcmp ((a),(b)) == 0) ++#endif + + /* When sorting by size, we use this structure to hold the size and a + pointer to the minisymbol. */ +@@ -194,6 +199,18 @@ static const char *plugin_target = NULL; + static bfd *lineno_cache_bfd; + static bfd *lineno_cache_rel_bfd; + ++typedef enum unicode_display_type ++{ ++ unicode_default = 0, ++ unicode_locale, ++ unicode_escape, ++ unicode_hex, ++ unicode_highlight, ++ unicode_invalid ++} unicode_display_type; ++ ++static unicode_display_type unicode_display = unicode_default; ++ + #define OPTION_TARGET 200 + #define OPTION_PLUGIN (OPTION_TARGET + 1) + #define OPTION_SIZE_SORT (OPTION_PLUGIN + 1) +@@ -224,6 +241,7 @@ static struct option long_options[] = + {"target", required_argument, 0, OPTION_TARGET}, + {"defined-only", no_argument, &defined_only, 1}, + {"undefined-only", no_argument, &undefined_only, 1}, ++ {"unicode", required_argument, NULL, 'U'}, + {"version", no_argument, &show_version, 1}, + {"with-symbol-versions", no_argument, &with_symbol_versions, 1}, + {0, no_argument, 0, 0} +@@ -271,6 +289,8 @@ usage (FILE *stream, int status) + -t, --radix=RADIX Use RADIX for printing symbol values\n\ + --target=BFDNAME Specify the target object format as BFDNAME\n\ + -u, --undefined-only Display only undefined symbols\n\ ++ -U {d|s|i|x|e|h} Specify how to treat UTF-8 encoded unicode characters\n\ ++ --unicode={default|show|invalid|hex|escape|highlight}\n\ + --with-symbol-versions Display version strings after symbol names\n\ + -X 32_64 (ignored)\n\ + @FILE Read options from FILE\n\ +@@ -399,6 +419,189 @@ get_coff_symbol_type (const struct inter + return bufp; + } + ++/* Convert a potential UTF-8 encoded sequence in IN into characters in OUT. ++ The conversion format is controlled by the unicode_display variable. ++ Returns the number of characters added to OUT. ++ Returns the number of bytes consumed from IN in CONSUMED. ++ Always consumes at least one byte and displays at least one character. */ ++ ++static unsigned int ++display_utf8 (const unsigned char * in, char * out, unsigned int * consumed) ++{ ++ char * orig_out = out; ++ unsigned int nchars = 0; ++ ++ if (unicode_display == unicode_default) ++ goto invalid; ++ ++ if (in[0] < 0xc0) ++ goto invalid; ++ ++ if ((in[1] & 0xc0) != 0x80) ++ goto invalid; ++ ++ if ((in[0] & 0x20) == 0) ++ { ++ nchars = 2; ++ goto valid; ++ } ++ ++ if ((in[2] & 0xc0) != 0x80) ++ goto invalid; ++ ++ if ((in[0] & 0x10) == 0) ++ { ++ nchars = 3; ++ goto valid; ++ } ++ ++ if ((in[3] & 0xc0) != 0x80) ++ goto invalid; ++ ++ nchars = 4; ++ ++ valid: ++ switch (unicode_display) ++ { ++ case unicode_locale: ++ /* Copy the bytes into the output buffer as is. */ ++ memcpy (out, in, nchars); ++ out += nchars; ++ break; ++ ++ case unicode_invalid: ++ case unicode_hex: ++ { ++ unsigned int j; ++ ++ out += sprintf (out, "%c", unicode_display == unicode_hex ? '<' : '{'); ++ for (j = 0; j < nchars; j++) ++ out += sprintf (out, "%02x", in [j]); ++ out += sprintf (out, "%c", unicode_display == unicode_hex ? '>' : '}'); ++ } ++ break; ++ ++ case unicode_highlight: ++ if (isatty (1)) ++ out += sprintf (out, "\x1B[31;47m"); /* Red. */ ++ /* Fall through. */ ++ case unicode_escape: ++ switch (nchars) ++ { ++ case 2: ++ out += sprintf (out, "\\u%02x%02x", ++ ((in[0] & 0x1c) >> 2), ++ ((in[0] & 0x03) << 6) | (in[1] & 0x3f)); ++ break; ++ ++ case 3: ++ out += sprintf (out, "\\u%02x%02x", ++ ((in[0] & 0x0f) << 4) | ((in[1] & 0x3c) >> 2), ++ ((in[1] & 0x03) << 6) | ((in[2] & 0x3f))); ++ break; ++ ++ case 4: ++ out += sprintf (out, "\\u%02x%02x%02x", ++ ((in[0] & 0x07) << 6) | ((in[1] & 0x3c) >> 2), ++ ((in[1] & 0x03) << 6) | ((in[2] & 0x3c) >> 2), ++ ((in[2] & 0x03) << 6) | ((in[3] & 0x3f))); ++ break; ++ default: ++ /* URG. */ ++ break; ++ } ++ ++ if (unicode_display == unicode_highlight && isatty (1)) ++ out += sprintf (out, "\033[0m"); /* Default colour. */ ++ break; ++ ++ default: ++ /* URG */ ++ break; ++ } ++ ++ * consumed = nchars; ++ return out - orig_out; ++ ++ invalid: ++ /* Not a valid UTF-8 sequence. */ ++ *out = *in; ++ * consumed = 1; ++ return 1; ++} ++ ++/* Convert any UTF-8 encoded characters in NAME into the form specified by ++ unicode_display. Also converts control characters. Returns a static ++ buffer if conversion was necessary. ++ Code stolen from objdump.c:sanitize_string(). */ ++ ++static const char * ++convert_utf8 (const char * in) ++{ ++ static char * buffer = NULL; ++ static size_t buffer_len = 0; ++ const char * original = in; ++ char * out; ++ ++ /* Paranoia. */ ++ if (in == NULL) ++ return ""; ++ ++ /* See if any conversion is necessary. ++ In the majority of cases it will not be needed. */ ++ do ++ { ++ unsigned char c = *in++; ++ ++ if (c == 0) ++ return original; ++ ++ if (ISCNTRL (c)) ++ break; ++ ++ if (unicode_display != unicode_default && c >= 0xc0) ++ break; ++ } ++ while (1); ++ ++ /* Copy the input, translating as needed. */ ++ in = original; ++ if (buffer_len < (strlen (in) * 9)) ++ { ++ free ((void *) buffer); ++ buffer_len = strlen (in) * 9; ++ buffer = xmalloc (buffer_len + 1); ++ } ++ ++ out = buffer; ++ do ++ { ++ unsigned char c = *in++; ++ ++ if (c == 0) ++ break; ++ ++ if (ISCNTRL (c)) ++ { ++ *out++ = '^'; ++ *out++ = c + 0x40; ++ } ++ else if (unicode_display != unicode_default && c >= 0xc0) ++ { ++ unsigned int num_consumed; ++ ++ out += display_utf8 ((const unsigned char *)(in - 1), out, & num_consumed); ++ in += num_consumed - 1; ++ } ++ else ++ *out++ = c; ++ } ++ while (1); ++ ++ *out = 0; ++ return buffer; ++} ++ + /* Print symbol name NAME, read from ABFD, with printf format FORM, + demangling it if requested. */ + +@@ -417,6 +618,9 @@ print_symname (const char *form, const c + } + } + ++ if (unicode_display != unicode_default) ++ name = convert_utf8 (name); ++ + printf (form, name); + } + +@@ -1657,7 +1861,7 @@ main (int argc, char **argv) + bfd_init (); + set_default_bfd_target (); + +- while ((c = getopt_long (argc, argv, "aABCDef:gHhlnopPrSst:uvVvX:", ++ while ((c = getopt_long (argc, argv, "aABCDef:gHhlnopPrSst:uU:vVvX:", + long_options, (int *) 0)) != EOF) + { + switch (c) +@@ -1738,6 +1942,24 @@ main (int argc, char **argv) + case 'u': + undefined_only = 1; + break; ++ ++ case 'U': ++ if (streq (optarg, "default") || streq (optarg, "d")) ++ unicode_display = unicode_default; ++ else if (streq (optarg, "locale") || streq (optarg, "l")) ++ unicode_display = unicode_locale; ++ else if (streq (optarg, "escape") || streq (optarg, "e")) ++ unicode_display = unicode_escape; ++ else if (streq (optarg, "invalid") || streq (optarg, "i")) ++ unicode_display = unicode_invalid; ++ else if (streq (optarg, "hex") || streq (optarg, "x")) ++ unicode_display = unicode_hex; ++ else if (streq (optarg, "highlight") || streq (optarg, "h")) ++ unicode_display = unicode_highlight; ++ else ++ fatal (_("invalid argument to -U/--unicode: %s"), optarg); ++ break; ++ + case 'V': + show_version = 1; + break; +diff -rup binutils.orig/binutils/objdump.c binutils-2.30/binutils/objdump.c +--- binutils.orig/binutils/objdump.c 2021-10-20 12:39:41.677008510 +0100 ++++ binutils-2.30/binutils/objdump.c 2021-10-20 13:24:11.009293189 +0100 +@@ -192,6 +192,18 @@ static const struct objdump_private_desc + OBJDUMP_PRIVATE_VECTORS + NULL + }; ++ ++typedef enum unicode_display_type ++{ ++ unicode_default = 0, ++ unicode_locale, ++ unicode_escape, ++ unicode_hex, ++ unicode_highlight, ++ unicode_invalid ++} unicode_display_type; ++ ++static unicode_display_type unicode_display = unicode_default; + + static void usage (FILE *, int) ATTRIBUTE_NORETURN; + static void +@@ -225,6 +237,9 @@ usage (FILE *stream, int status) + -r, --reloc Display the relocation entries in the file\n\ + -R, --dynamic-reloc Display the dynamic relocation entries in the file\n\ + @ Read options from \n\ ++ -U[d|l|i|x|e|h] Controls the display of UTF-8 unicode characters\n\ ++ --unicode=[default|locale|invalid|hex|escape|highlight]\n")); ++ fprintf (stream, _("\ + -v, --version Display this program's version number\n\ + -i, --info List object formats and architectures supported\n\ + -H, --help Display this information\n\ +@@ -343,6 +358,7 @@ static struct option long_options[]= + {"stop-address", required_argument, NULL, OPTION_STOP_ADDRESS}, + {"syms", no_argument, NULL, 't'}, + {"target", required_argument, NULL, 'b'}, ++ {"unicode", required_argument, NULL, 'U'}, + {"version", no_argument, NULL, 'V'}, + {"wide", no_argument, NULL, 'w'}, + {"prefix", required_argument, NULL, OPTION_PREFIX}, +@@ -362,6 +378,191 @@ nonfatal (const char *msg) + exit_status = 1; + } + ++/* Convert a potential UTF-8 encoded sequence in IN into characters in OUT. ++ The conversion format is controlled by the unicode_display variable. ++ Returns the number of characters added to OUT. ++ Returns the number of bytes consumed from IN in CONSUMED. ++ Always consumes at least one byte and displays at least one character. */ ++ ++static unsigned int ++display_utf8 (const unsigned char * in, char * out, unsigned int * consumed) ++{ ++ char * orig_out = out; ++ unsigned int nchars = 0; ++ ++ if (unicode_display == unicode_default) ++ goto invalid; ++ ++ if (in[0] < 0xc0) ++ goto invalid; ++ ++ if ((in[1] & 0xc0) != 0x80) ++ goto invalid; ++ ++ if ((in[0] & 0x20) == 0) ++ { ++ nchars = 2; ++ goto valid; ++ } ++ ++ if ((in[2] & 0xc0) != 0x80) ++ goto invalid; ++ ++ if ((in[0] & 0x10) == 0) ++ { ++ nchars = 3; ++ goto valid; ++ } ++ ++ if ((in[3] & 0xc0) != 0x80) ++ goto invalid; ++ ++ nchars = 4; ++ ++ valid: ++ switch (unicode_display) ++ { ++ case unicode_locale: ++ /* Copy the bytes into the output buffer as is. */ ++ memcpy (out, in, nchars); ++ out += nchars; ++ break; ++ ++ case unicode_invalid: ++ case unicode_hex: ++ { ++ unsigned int j; ++ ++ out += sprintf (out, "%c", unicode_display == unicode_hex ? '<' : '{'); ++ for (j = 0; j < nchars; j++) ++ out += sprintf (out, "%02x", in [j]); ++ out += sprintf (out, "%c", unicode_display == unicode_hex ? '>' : '}'); ++ } ++ break; ++ ++ case unicode_highlight: ++ if (isatty (1)) ++ out += sprintf (out, "\x1B[31;47m"); /* Red. */ ++ /* Fall through. */ ++ case unicode_escape: ++ switch (nchars) ++ { ++ case 2: ++ out += sprintf (out, "\\u%02x%02x", ++ ((in[0] & 0x1c) >> 2), ++ ((in[0] & 0x03) << 6) | (in[1] & 0x3f)); ++ break; ++ ++ case 3: ++ out += sprintf (out, "\\u%02x%02x", ++ ((in[0] & 0x0f) << 4) | ((in[1] & 0x3c) >> 2), ++ ((in[1] & 0x03) << 6) | ((in[2] & 0x3f))); ++ break; ++ ++ case 4: ++ out += sprintf (out, "\\u%02x%02x%02x", ++ ((in[0] & 0x07) << 6) | ((in[1] & 0x3c) >> 2), ++ ((in[1] & 0x03) << 6) | ((in[2] & 0x3c) >> 2), ++ ((in[2] & 0x03) << 6) | ((in[3] & 0x3f))); ++ break; ++ default: ++ /* URG. */ ++ break; ++ } ++ ++ if (unicode_display == unicode_highlight && isatty (1)) ++ out += sprintf (out, "\033[0m"); /* Default colour. */ ++ break; ++ ++ default: ++ /* URG */ ++ break; ++ } ++ ++ * consumed = nchars; ++ return out - orig_out; ++ ++ invalid: ++ /* Not a valid UTF-8 sequence. */ ++ *out = *in; ++ * consumed = 1; ++ return 1; ++} ++ ++/* Returns a version of IN with any control characters ++ replaced by escape sequences. Uses a static buffer ++ if necessary. ++ ++ If unicode display is enabled, then also handles the ++ conversion of unicode characters. */ ++ ++static const char * ++sanitize_string (const char * in) ++{ ++ static char * buffer = NULL; ++ static size_t buffer_len = 0; ++ const char * original = in; ++ char * out; ++ ++ /* Paranoia. */ ++ if (in == NULL) ++ return ""; ++ ++ /* See if any conversion is necessary. In the majority ++ of cases it will not be needed. */ ++ do ++ { ++ unsigned char c = *in++; ++ ++ if (c == 0) ++ return original; ++ ++ if (ISCNTRL (c)) ++ break; ++ ++ if (unicode_display != unicode_default && c >= 0xc0) ++ break; ++ } ++ while (1); ++ ++ /* Copy the input, translating as needed. */ ++ in = original; ++ if (buffer_len < (strlen (in) * 9)) ++ { ++ free ((void *) buffer); ++ buffer_len = strlen (in) * 9; ++ buffer = xmalloc (buffer_len + 1); ++ } ++ ++ out = buffer; ++ do ++ { ++ unsigned char c = *in++; ++ ++ if (c == 0) ++ break; ++ ++ if (ISCNTRL (c)) ++ { ++ *out++ = '^'; ++ *out++ = c + 0x40; ++ } ++ else if (unicode_display != unicode_default && c >= 0xc0) ++ { ++ unsigned int num_consumed; ++ ++ out += display_utf8 ((const unsigned char *)(in - 1), out, & num_consumed); ++ in += num_consumed - 1; ++ } ++ else ++ *out++ = c; ++ } ++ while (1); ++ ++ *out = 0; ++ return buffer; ++} ++ + /* Returns TRUE if the specified section should be dumped. */ + + static bfd_boolean +@@ -458,8 +657,8 @@ dump_section_header (bfd *abfd, asection + return; + + printf ("%3d %-*s %08lx ", section->index, longest_section_name, +- bfd_get_section_name (abfd, section), +- (unsigned long) bfd_section_size (abfd, section) / opb); ++ sanitize_string (bfd_section_name (abfd, section)), ++ (unsigned long) bfd_section_size (abfd, section) / opb); + bfd_printf_vma (abfd, bfd_get_section_vma (abfd, section)); + printf (" "); + bfd_printf_vma (abfd, section->lma); +@@ -892,6 +1091,8 @@ objdump_print_symname (bfd *abfd, struct + if (bfd_is_und_section (bfd_get_section (sym))) + hidden = TRUE; + ++ name = sanitize_string (name); ++ + if (inf != NULL) + { + (*inf->fprintf_func) (inf->stream, "%s", name); +@@ -1141,7 +1342,7 @@ objdump_print_addr_with_sym (bfd *abfd, + bfd_vma secaddr; + + (*inf->fprintf_func) (inf->stream, " <%s", +- bfd_get_section_name (abfd, sec)); ++ sanitize_string (bfd_section_name (abfd, sec))); + secaddr = bfd_get_section_vma (abfd, sec); + if (vma < secaddr) + { +@@ -1563,7 +1764,7 @@ show_line (bfd *abfd, asection *section, + && (prev_functionname == NULL + || strcmp (functionname, prev_functionname) != 0)) + { +- printf ("%s():\n", functionname); ++ printf ("%s():\n", sanitize_string (functionname)); + prev_line = -1; + } + if (linenumber > 0 +@@ -1572,10 +1773,10 @@ show_line (bfd *abfd, asection *section, + { + if (discriminator > 0) + printf ("%s:%u (discriminator %u)\n", +- filename == NULL ? "???" : filename, ++ filename == NULL ? "???" : sanitize_string (filename), + linenumber, discriminator); + else +- printf ("%s:%u\n", filename == NULL ? "???" : filename, ++ printf ("%s:%u\n", filename == NULL ? "???" : sanitize_string (filename), + linenumber); + } + if (unwind_inlines) +@@ -1586,8 +1787,10 @@ show_line (bfd *abfd, asection *section, + + while (bfd_find_inliner_info (abfd, &filename2, &functionname2, + &line2)) +- printf ("inlined by %s:%u (%s)\n", filename2, line2, +- functionname2); ++ { ++ printf ("inlined by %s:%u", sanitize_string (filename2), line2); ++ printf (" (%s)\n", sanitize_string (functionname2)); ++ } + } + } + +@@ -2081,7 +2284,7 @@ disassemble_bytes (struct disassemble_in + sym_name = bfd_get_section_name (aux->abfd, sym_sec); + if (sym_name == NULL || *sym_name == '\0') + sym_name = "*unknown*"; +- printf ("%s", sym_name); ++ printf ("%s", sanitize_string (sym_name)); + } + } + +@@ -2228,7 +2431,7 @@ disassemble_section (bfd *abfd, asection + && (*rel_pp)->address < rel_offset + addr_offset) + ++rel_pp; + +- printf (_("\nDisassembly of section %s:\n"), section->name); ++ printf (_("\nDisassembly of section %s:\n"), sanitize_string (section->name)); + + /* Find the nearest symbol forwards from our current position. */ + paux->require_sec = TRUE; +@@ -2512,7 +2715,7 @@ load_specific_debug_section (enum dwarf_ + { + free_debug_section (debug); + printf (_("\nCan't get contents for section '%s'.\n"), +- section->name); ++ sanitize_string (section->name)); + return FALSE; + } + /* Ensure any string section has a terminating NUL. */ +@@ -2534,7 +2737,7 @@ load_specific_debug_section (enum dwarf_ + { + free_debug_section (debug); + printf (_("\nCan't get contents for section '%s'.\n"), +- section->name); ++ sanitize_string (section->name)); + return FALSE; + } + +@@ -2780,14 +2983,14 @@ read_section_stabs (bfd *abfd, const cha + stabsect = bfd_get_section_by_name (abfd, sect_name); + if (stabsect == NULL) + { +- printf (_("No %s section present\n\n"), sect_name); ++ printf (_("No %s section present\n\n"), sanitize_string (sect_name)); + return FALSE; + } + + if (!bfd_malloc_and_get_section (abfd, stabsect, &contents)) + { + non_fatal (_("reading %s section of %s failed: %s"), +- sect_name, bfd_get_filename (abfd), ++ sanitize_string (sect_name), bfd_get_filename (abfd), + bfd_errmsg (bfd_get_error ())); + exit_status = 1; + free (contents); +@@ -2830,7 +3033,7 @@ print_section_stabs (bfd *abfd, + stabp = stabs; + stabs_end = stabp + stab_size; + +- printf (_("Contents of %s section:\n\n"), stabsect_name); ++ printf (_("Contents of %s section:\n\n"), sanitize_string (stabsect_name)); + printf ("Symnum n_type n_othr n_desc n_value n_strx String\n"); + + /* Loop through all symbols and print them. +@@ -3093,7 +3296,7 @@ dump_section (bfd *abfd, asection *secti + if (start_offset >= stop_offset) + return; + +- printf (_("Contents of section %s:"), section->name); ++ printf (_("Contents of section %s:"), sanitize_string (section->name)); + if (display_file_offsets) + printf (_(" (Starting at file offset: 0x%lx)"), + (unsigned long) (section->filepos + start_offset)); +@@ -3102,7 +3305,7 @@ dump_section (bfd *abfd, asection *secti + if (!bfd_get_full_section_contents (abfd, section, &data)) + { + non_fatal (_("Reading section %s failed because: %s"), +- section->name, bfd_errmsg (bfd_get_error ())); ++ sanitize_string (section->name), bfd_errmsg (bfd_get_error ())); + return; + } + +@@ -3242,6 +3445,24 @@ dump_symbols (bfd *abfd ATTRIBUTE_UNUSED + free (alloc); + } + } ++ else if (unicode_display != unicode_default ++ && name != NULL && *name != '\0') ++ { ++ const char * sanitized_name; ++ ++ /* If we want to sanitize the name, we do it here, and ++ temporarily clobber it while calling bfd_print_symbol. ++ FIXME: This is a gross hack. */ ++ sanitized_name = sanitize_string (name); ++ if (sanitized_name != name) ++ (*current)->name = sanitized_name; ++ else ++ sanitized_name = NULL; ++ bfd_print_symbol (cur_bfd, stdout, *current, ++ bfd_print_symbol_all); ++ if (sanitized_name != NULL) ++ (*current)->name = name; ++ } + else + bfd_print_symbol (cur_bfd, stdout, *current, + bfd_print_symbol_all); +@@ -3307,7 +3528,7 @@ dump_reloc_set (bfd *abfd, asection *sec + && (last_functionname == NULL + || strcmp (functionname, last_functionname) != 0)) + { +- printf ("%s():\n", functionname); ++ printf ("%s():\n", sanitize_string (functionname)); + if (last_functionname != NULL) + free (last_functionname); + last_functionname = xstrdup (functionname); +@@ -3321,9 +3542,9 @@ dump_reloc_set (bfd *abfd, asection *sec + || (discriminator != last_discriminator))) + { + if (discriminator > 0) +- printf ("%s:%u\n", filename == NULL ? "???" : filename, linenumber); ++ printf ("%s:%u\n", filename == NULL ? "???" : sanitize_string (filename), linenumber); + else +- printf ("%s:%u (discriminator %u)\n", filename == NULL ? "???" : filename, ++ printf ("%s:%u (discriminator %u)\n", filename == NULL ? "???" : sanitize_string (filename), + linenumber, discriminator); + last_line = linenumber; + last_discriminator = discriminator; +@@ -3393,7 +3614,7 @@ dump_reloc_set (bfd *abfd, asection *sec + { + if (section_name == NULL) + section_name = "*unknown*"; +- printf ("[%s]", section_name); ++ printf ("[%s]", sanitize_string (section_name)); + } + + if (q->addend) +@@ -3443,7 +3664,7 @@ dump_relocs_in_section (bfd *abfd, + if (relsize < 0) + bfd_fatal (bfd_get_filename (abfd)); + +- printf ("RELOCATION RECORDS FOR [%s]:", section->name); ++ printf ("RELOCATION RECORDS FOR [%s]:", sanitize_string (section->name)); + + if (relsize == 0) + { +@@ -3835,7 +4056,7 @@ main (int argc, char **argv) + set_default_bfd_target (); + + while ((c = getopt_long (argc, argv, +- "pP:ib:m:M:VvCdDlfFaHhrRtTxsSI:j:wE:zgeGW::", ++ "pP:ib:m:M:VvCdDlfFaHhrRtTxsSI:j:wE:zgeGW::U:", + long_options, (int *) 0)) + != EOF) + { +@@ -4069,6 +4290,23 @@ main (int argc, char **argv) + seenflag = TRUE; + break; + ++ case 'U': ++ if (streq (optarg, "default") || streq (optarg, "d")) ++ unicode_display = unicode_default; ++ else if (streq (optarg, "locale") || streq (optarg, "l")) ++ unicode_display = unicode_locale; ++ else if (streq (optarg, "escape") || streq (optarg, "e")) ++ unicode_display = unicode_escape; ++ else if (streq (optarg, "invalid") || streq (optarg, "i")) ++ unicode_display = unicode_invalid; ++ else if (streq (optarg, "hex") || streq (optarg, "x")) ++ unicode_display = unicode_hex; ++ else if (streq (optarg, "highlight") || streq (optarg, "h")) ++ unicode_display = unicode_highlight; ++ else ++ fatal (_("invalid argument to -U/--unicode: %s"), optarg); ++ break; ++ + case 'H': + usage (stdout, 0); + /* No need to set seenflag or to break - usage() does not return. */ +diff -rup binutils.orig/binutils/strings.c binutils-2.30/binutils/strings.c +--- binutils.orig/binutils/strings.c 2021-10-20 12:39:41.686008454 +0100 ++++ binutils-2.30/binutils/strings.c 2021-10-20 13:32:22.029182908 +0100 +@@ -55,6 +55,19 @@ + -T {bfdname} + Specify a non-default object file format. + ++ --unicode={default|locale|invalid|hex|escape|highlight} ++ -U {d|l|i|x|e|h} ++ Determine how to handle UTF-8 unicode characters. The default ++ is no special treatment. All other versions of this option ++ only apply if the encoding is valid and enabling the option ++ implies --encoding=S. ++ The 'locale' option displays the characters according to the ++ current locale. The 'invalid' option treats them as ++ non-string characters. The 'hex' option displays them as hex ++ byte sequences. The 'escape' option displays them as escape ++ sequences and the 'highlight' option displays them as ++ coloured escape sequences. ++ + --output-separator=sep_string + -s sep_string String used to separate parsed strings in output. + Default is newline. +@@ -76,6 +89,22 @@ + #include "safe-ctype.h" + #include "bucomm.h" + ++#ifndef streq ++#define streq(a,b) (strcmp ((a),(b)) == 0) ++#endif ++ ++typedef enum unicode_display_type ++{ ++ unicode_default = 0, ++ unicode_locale, ++ unicode_escape, ++ unicode_hex, ++ unicode_highlight, ++ unicode_invalid ++} unicode_display_type; ++ ++static unicode_display_type unicode_display = unicode_default; ++ + #define STRING_ISGRAPHIC(c) \ + ( (c) >= 0 \ + && (c) <= 255 \ +@@ -94,7 +123,7 @@ extern int errno; + static int address_radix; + + /* Minimum length of sequence of graphic chars to trigger output. */ +-static int string_min; ++static uint string_min; + + /* Whether or not we include all whitespace as a graphic char. */ + static bfd_boolean include_all_whitespace; +@@ -130,6 +159,7 @@ static struct option long_options[] = + {"target", required_argument, NULL, 'T'}, + {"output-separator", required_argument, NULL, 's'}, + {"help", no_argument, NULL, 'h'}, ++ {"unicode", required_argument, NULL, 'U'}, + {"version", no_argument, NULL, 'v'}, + {NULL, 0, NULL, 0} + }; +@@ -173,7 +203,7 @@ main (int argc, char **argv) + encoding = 's'; + output_separator = NULL; + +- while ((optc = getopt_long (argc, argv, "adfhHn:wot:e:T:s:Vv0123456789", ++ while ((optc = getopt_long (argc, argv, "adfhHn:wot:e:T:s:U:Vv0123456789", + long_options, (int *) 0)) != EOF) + { + switch (optc) +@@ -246,6 +276,23 @@ main (int argc, char **argv) + output_separator = optarg; + break; + ++ case 'U': ++ if (streq (optarg, "default") || streq (optarg, "d")) ++ unicode_display = unicode_default; ++ else if (streq (optarg, "locale") || streq (optarg, "l")) ++ unicode_display = unicode_locale; ++ else if (streq (optarg, "escape") || streq (optarg, "e")) ++ unicode_display = unicode_escape; ++ else if (streq (optarg, "invalid") || streq (optarg, "i")) ++ unicode_display = unicode_invalid; ++ else if (streq (optarg, "hex") || streq (optarg, "x")) ++ unicode_display = unicode_hex; ++ else if (streq (optarg, "highlight") || streq (optarg, "h")) ++ unicode_display = unicode_highlight; ++ else ++ fatal (_("invalid argument to -U/--unicode: %s"), optarg); ++ break; ++ + case 'V': + case 'v': + print_version ("strings"); +@@ -260,6 +307,9 @@ main (int argc, char **argv) + } + } + ++ if (unicode_display != unicode_default) ++ encoding = 'S'; ++ + if (numeric_opt != 0) + { + string_min = (int) strtoul (argv[numeric_opt - 1] + 1, &s, 0); +@@ -501,11 +551,629 @@ get_char (FILE *stream, file_ptr *addres + + return r; + } ++ ++static void ++print_filename_and_address (const char * filename, file_ptr address) ++{ ++ if (print_filenames) ++ printf ("%s: ", filename); ++ ++ if (! print_addresses) ++ return; ++ ++ switch (address_radix) ++ { ++ case 8: ++ if (sizeof (address) > sizeof (long)) ++ { ++#ifndef __MSVCRT__ ++ printf ("%7llo ", (unsigned long long) address); ++#else ++ printf ("%7I64o ", (unsigned long long) address); ++#endif ++ } ++ else ++ printf ("%7lo ", (unsigned long) address); ++ break; ++ ++ case 10: ++ if (sizeof (address) > sizeof (long)) ++ { ++#ifndef __MSVCRT__ ++ printf ("%7llu ", (unsigned long long) address); ++#else ++ printf ("%7I64d ", (unsigned long long) address); ++#endif ++ } ++ else ++ printf ("%7ld ", (long) address); ++ break; ++ ++ case 16: ++ if (sizeof (address) > sizeof (long)) ++ { ++#ifndef __MSVCRT__ ++ printf ("%7llx ", (unsigned long long) address); ++#else ++ printf ("%7I64x ", (unsigned long long) address); ++#endif ++ } ++ else ++ printf ("%7lx ", (unsigned long) address); ++ break; ++ } ++} ++ ++/* Return non-zero if the bytes starting at BUFFER form a valid UTF-8 encoding. ++ If the encoding is valid then returns the number of bytes it uses. */ ++ ++static unsigned int ++is_valid_utf8 (const unsigned char * buffer, unsigned long buflen) ++{ ++ if (buffer[0] < 0xc0) ++ return 0; ++ ++ if (buflen < 2) ++ return 0; ++ ++ if ((buffer[1] & 0xc0) != 0x80) ++ return 0; ++ ++ if ((buffer[0] & 0x20) == 0) ++ return 2; ++ ++ if (buflen < 3) ++ return 0; ++ ++ if ((buffer[2] & 0xc0) != 0x80) ++ return 0; ++ ++ if ((buffer[0] & 0x10) == 0) ++ return 3; ++ ++ if (buflen < 4) ++ return 0; ++ ++ if ((buffer[3] & 0xc0) != 0x80) ++ return 0; ++ ++ return 4; ++} ++ ++/* Display a UTF-8 encoded character in BUFFER according to the setting ++ of unicode_display. The character is known to be valid. ++ Returns the number of bytes consumed. */ ++ ++static unsigned int ++display_utf8_char (const unsigned char * buffer) ++{ ++ unsigned int j; ++ unsigned int utf8_len; ++ ++ switch (buffer[0] & 0x30) ++ { ++ case 0x00: ++ case 0x10: ++ utf8_len = 2; ++ break; ++ case 0x20: ++ utf8_len = 3; ++ break; ++ default: ++ utf8_len = 4; ++ } ++ ++ switch (unicode_display) ++ { ++ default: ++ fprintf (stderr, "ICE: unexpected unicode display type\n"); ++ break; ++ ++ case unicode_escape: ++ case unicode_highlight: ++ if (unicode_display == unicode_highlight && isatty (1)) ++ printf ("\x1B[31;47m"); /* Red. */ ++ ++ switch (utf8_len) ++ { ++ case 2: ++ printf ("\\u%02x%02x", ++ ((buffer[0] & 0x1c) >> 2), ++ ((buffer[0] & 0x03) << 6) | (buffer[1] & 0x3f)); ++ break; ++ ++ case 3: ++ printf ("\\u%02x%02x", ++ ((buffer[0] & 0x0f) << 4) | ((buffer[1] & 0x3c) >> 2), ++ ((buffer[1] & 0x03) << 6) | ((buffer[2] & 0x3f))); ++ break; ++ ++ case 4: ++ printf ("\\u%02x%02x%02x", ++ ((buffer[0] & 0x07) << 6) | ((buffer[1] & 0x3c) >> 2), ++ ((buffer[1] & 0x03) << 6) | ((buffer[2] & 0x3c) >> 2), ++ ((buffer[2] & 0x03) << 6) | ((buffer[3] & 0x3f))); ++ break; ++ default: ++ /* URG. */ ++ break; ++ } ++ ++ if (unicode_display == unicode_highlight && isatty (1)) ++ printf ("\033[0m"); /* Default colour. */ ++ break; ++ ++ case unicode_hex: ++ putchar ('<'); ++ for (j = 0; j < utf8_len; j++) ++ printf ("%02x", buffer [j]); ++ putchar ('>'); ++ break; ++ ++ case unicode_locale: ++ printf ("%.1s", buffer); ++ break; ++ } ++ ++ return utf8_len; ++} ++ ++/* Display strings in BUFFER. Treat any UTF-8 encoded characters encountered ++ according to the setting of the unicode_display variable. The buffer ++ contains BUFLEN bytes. ++ ++ Display the characters as if they started at ADDRESS and are contained in ++ FILENAME. */ ++ ++static void ++print_unicode_buffer (const char * filename, ++ file_ptr address, ++ const unsigned char * buffer, ++ unsigned long buflen) ++{ ++ /* Paranoia checks... */ ++ if (filename == NULL ++ || buffer == NULL ++ || unicode_display == unicode_default ++ || encoding != 'S' ++ || encoding_bytes != 1) ++ { ++ fprintf (stderr, "ICE: bad arguments to print_unicode_buffer\n"); ++ return; ++ } ++ ++ if (buflen == 0) ++ return; ++ ++ /* We must only display strings that are at least string_min *characters* ++ long. So we scan the buffer in two stages. First we locate the start ++ of a potential string. Then we walk along it until we have found ++ string_min characters. Then we go back to the start point and start ++ displaying characters according to the unicode_display setting. */ ++ ++ unsigned long start_point = 0; ++ unsigned long i = 0; ++ unsigned int char_len = 1; ++ unsigned int num_found = 0; ++ ++ for (i = 0; i < buflen; i += char_len) ++ { ++ int c = buffer[i]; ++ ++ char_len = 1; ++ ++ /* Find the first potential character of a string. */ ++ if (! STRING_ISGRAPHIC (c)) ++ { ++ num_found = 0; ++ continue; ++ } ++ ++ if (c > 126) ++ { ++ if (c < 0xc0) ++ { ++ num_found = 0; ++ continue; ++ } ++ ++ if ((char_len = is_valid_utf8 (buffer + i, buflen - i)) == 0) ++ { ++ char_len = 1; ++ num_found = 0; ++ continue; ++ } ++ ++ if (unicode_display == unicode_invalid) ++ { ++ /* We have found a valid UTF-8 character, but we treat it as non-graphic. */ ++ num_found = 0; ++ continue; ++ } ++ } ++ ++ if (num_found == 0) ++ /* We have found a potential starting point for a string. */ ++ start_point = i; ++ ++ ++ num_found; ++ ++ if (num_found >= string_min) ++ break; ++ } ++ ++ if (num_found < string_min) ++ return; ++ ++ print_filename_and_address (filename, address + start_point); ++ ++ /* We have found string_min characters. Display them and any ++ more that follow. */ ++ for (i = start_point; i < buflen; i += char_len) ++ { ++ int c = buffer[i]; ++ ++ char_len = 1; ++ ++ if (! STRING_ISGRAPHIC (c)) ++ break; ++ else if (c < 127) ++ putchar (c); ++ else if (! is_valid_utf8 (buffer + i, buflen - i)) ++ break; ++ else if (unicode_display == unicode_invalid) ++ break; ++ else ++ char_len = display_utf8_char (buffer + i); ++ } ++ ++ if (output_separator) ++ fputs (output_separator, stdout); ++ else ++ putchar ('\n'); ++ ++ /* FIXME: Using tail recursion here is lazy programming... */ ++ print_unicode_buffer (filename, address + i, buffer + i, buflen - i); ++} ++ ++static int ++get_unicode_byte (FILE * stream, ++ unsigned char * putback, ++ unsigned int * num_putback, ++ unsigned int * num_read) ++{ ++ if (* num_putback > 0) ++ { ++ * num_putback = * num_putback - 1; ++ return putback [* num_putback]; ++ } ++ ++ * num_read = * num_read + 1; ++ ++#if defined(HAVE_GETC_UNLOCKED) && HAVE_DECL_GETC_UNLOCKED ++ return getc_unlocked (stream); ++#else ++ return getc (stream); ++#endif ++} ++ ++/* Helper function for print_unicode_stream. */ ++ ++static void ++print_unicode_stream_body (const char * filename, ++ file_ptr address, ++ FILE * stream, ++ unsigned char * putback_buf, ++ unsigned int num_putback, ++ unsigned char * print_buf) ++{ ++ /* It would be nice if we could just read the stream into a buffer ++ and then process if with print_unicode_buffer. But the input ++ might be huge or it might time-locked (eg stdin). So instead ++ we go one byte at a time... */ ++ ++ file_ptr start_point = 0; ++ unsigned int num_read = 0; ++ unsigned int num_chars = 0; ++ unsigned int num_print = 0; ++ int c; ++ ++ /* Find a series of string_min characters. Put them into print_buf. */ ++ do ++ { ++ if (num_chars >= string_min) ++ break; ++ ++ c = get_unicode_byte (stream, putback_buf, & num_putback, & num_read); ++ if (c == EOF) ++ break; ++ ++ if (! STRING_ISGRAPHIC (c)) ++ { ++ num_chars = num_print = 0; ++ continue; ++ } ++ ++ if (num_chars == 0) ++ start_point = num_read - 1; ++ ++ if (c < 127) ++ { ++ print_buf[num_print] = c; ++ num_chars ++; ++ num_print ++; ++ continue; ++ } ++ ++ if (c < 0xc0) ++ { ++ num_chars = num_print = 0; ++ continue; ++ } ++ ++ /* We *might* have a UTF-8 sequence. Time to start peeking. */ ++ char utf8[4]; ++ ++ utf8[0] = c; ++ c = get_unicode_byte (stream, putback_buf, & num_putback, & num_read); ++ if (c == EOF) ++ break; ++ utf8[1] = c; ++ ++ if ((utf8[1] & 0xc0) != 0x80) ++ { ++ /* Invalid UTF-8. */ ++ putback_buf[num_putback++] = utf8[1]; ++ num_chars = num_print = 0; ++ continue; ++ } ++ else if ((utf8[0] & 0x20) == 0) ++ { ++ /* A valid 2-byte UTF-8 encoding. */ ++ if (unicode_display == unicode_invalid) ++ { ++ putback_buf[num_putback++] = utf8[1]; ++ num_chars = num_print = 0; ++ } ++ else ++ { ++ print_buf[num_print ++] = utf8[0]; ++ print_buf[num_print ++] = utf8[1]; ++ num_chars ++; ++ } ++ continue; ++ } ++ ++ c = get_unicode_byte (stream, putback_buf, & num_putback, & num_read); ++ if (c == EOF) ++ break; ++ utf8[2] = c; ++ ++ if ((utf8[2] & 0xc0) != 0x80) ++ { ++ /* Invalid UTF-8. */ ++ putback_buf[num_putback++] = utf8[2]; ++ putback_buf[num_putback++] = utf8[1]; ++ num_chars = num_print = 0; ++ continue; ++ } ++ else if ((utf8[0] & 0x10) == 0) ++ { ++ /* A valid 3-byte UTF-8 encoding. */ ++ if (unicode_display == unicode_invalid) ++ { ++ putback_buf[num_putback++] = utf8[2]; ++ putback_buf[num_putback++] = utf8[1]; ++ num_chars = num_print = 0; ++ } ++ else ++ { ++ print_buf[num_print ++] = utf8[0]; ++ print_buf[num_print ++] = utf8[1]; ++ print_buf[num_print ++] = utf8[2]; ++ num_chars ++; ++ } ++ continue; ++ } ++ ++ c = get_unicode_byte (stream, putback_buf, & num_putback, & num_read); ++ if (c == EOF) ++ break; ++ utf8[3] = c; ++ ++ if ((utf8[3] & 0xc0) != 0x80) ++ { ++ /* Invalid UTF-8. */ ++ putback_buf[num_putback++] = utf8[3]; ++ putback_buf[num_putback++] = utf8[2]; ++ putback_buf[num_putback++] = utf8[1]; ++ num_chars = num_print = 0; ++ } ++ /* We have a valid 4-byte UTF-8 encoding. */ ++ else if (unicode_display == unicode_invalid) ++ { ++ putback_buf[num_putback++] = utf8[3]; ++ putback_buf[num_putback++] = utf8[1]; ++ putback_buf[num_putback++] = utf8[2]; ++ num_chars = num_print = 0; ++ } ++ else ++ { ++ print_buf[num_print ++] = utf8[0]; ++ print_buf[num_print ++] = utf8[1]; ++ print_buf[num_print ++] = utf8[2]; ++ print_buf[num_print ++] = utf8[3]; ++ num_chars ++; ++ } ++ } ++ while (1); ++ ++ if (num_chars >= string_min) ++ { ++ /* We know that we have string_min valid characters in print_buf, ++ and there may be more to come in the stream. Start displaying ++ them. */ ++ ++ print_filename_and_address (filename, address + start_point); ++ ++ unsigned int i; ++ for (i = 0; i < num_print;) ++ { ++ if (print_buf[i] < 127) ++ putchar (print_buf[i++]); ++ else ++ i += display_utf8_char (print_buf + i); ++ } ++ ++ /* OK so now we have to start read unchecked bytes. */ ++ ++ /* Find a series of string_min characters. Put them into print_buf. */ ++ do ++ { ++ c = get_unicode_byte (stream, putback_buf, & num_putback, & num_read); ++ if (c == EOF) ++ break; ++ ++ if (! STRING_ISGRAPHIC (c)) ++ break; ++ ++ if (c < 127) ++ { ++ putchar (c); ++ continue; ++ } ++ ++ if (c < 0xc0) ++ break; ++ ++ /* We *might* have a UTF-8 sequence. Time to start peeking. */ ++ unsigned char utf8[4]; ++ ++ utf8[0] = c; ++ c = get_unicode_byte (stream, putback_buf, & num_putback, & num_read); ++ if (c == EOF) ++ break; ++ utf8[1] = c; ++ ++ if ((utf8[1] & 0xc0) != 0x80) ++ { ++ /* Invalid UTF-8. */ ++ putback_buf[num_putback++] = utf8[1]; ++ break; ++ } ++ else if ((utf8[0] & 0x20) == 0) ++ { ++ /* Valid 2-byte UTF-8. */ ++ if (unicode_display == unicode_invalid) ++ { ++ putback_buf[num_putback++] = utf8[1]; ++ break; ++ } ++ else ++ { ++ (void) display_utf8_char (utf8); ++ continue; ++ } ++ } ++ ++ c = get_unicode_byte (stream, putback_buf, & num_putback, & num_read); ++ if (c == EOF) ++ break; ++ utf8[2] = c; ++ ++ if ((utf8[2] & 0xc0) != 0x80) ++ { ++ /* Invalid UTF-8. */ ++ putback_buf[num_putback++] = utf8[2]; ++ putback_buf[num_putback++] = utf8[1]; ++ break; ++ } ++ else if ((utf8[0] & 0x10) == 0) ++ { ++ /* Valid 3-byte UTF-8. */ ++ if (unicode_display == unicode_invalid) ++ { ++ putback_buf[num_putback++] = utf8[2]; ++ putback_buf[num_putback++] = utf8[1]; ++ break; ++ } ++ else ++ { ++ (void) display_utf8_char (utf8); ++ continue; ++ } ++ } ++ ++ c = get_unicode_byte (stream, putback_buf, & num_putback, & num_read); ++ if (c == EOF) ++ break; ++ utf8[3] = c; ++ ++ if ((utf8[3] & 0xc0) != 0x80) ++ { ++ /* Invalid UTF-8. */ ++ putback_buf[num_putback++] = utf8[3]; ++ putback_buf[num_putback++] = utf8[2]; ++ putback_buf[num_putback++] = utf8[1]; ++ break; ++ } ++ else if (unicode_display == unicode_invalid) ++ { ++ putback_buf[num_putback++] = utf8[3]; ++ putback_buf[num_putback++] = utf8[2]; ++ putback_buf[num_putback++] = utf8[1]; ++ break; ++ } ++ else ++ /* A valid 4-byte UTF-8 encoding. */ ++ (void) display_utf8_char (utf8); ++ } ++ while (1); ++ ++ if (output_separator) ++ fputs (output_separator, stdout); ++ else ++ putchar ('\n'); ++ } ++ ++ if (c != EOF) ++ /* FIXME: Using tail recursion here is lazy, but it works. */ ++ print_unicode_stream_body (filename, address + num_read, stream, putback_buf, num_putback, print_buf); ++} ++ ++/* Display strings read in from STREAM. Treat any UTF-8 encoded characters ++ encountered according to the setting of the unicode_display variable. ++ The stream is positioned at ADDRESS and is attached to FILENAME. */ ++ ++static void ++print_unicode_stream (const char * filename, ++ file_ptr address, ++ FILE * stream) ++{ ++ /* Paranoia checks... */ ++ if (filename == NULL ++ || stream == NULL ++ || unicode_display == unicode_default ++ || encoding != 'S' ++ || encoding_bytes != 1) ++ { ++ fprintf (stderr, "ICE: bad arguments to print_unicode_stream\n"); ++ return; ++ } ++ ++ /* Allocate space for string_min 4-byte utf-8 characters. */ ++ unsigned char * print_buf = xmalloc ((4 * string_min) + 1); ++ /* We should never have to put back more than 4 bytes. */ ++ unsigned char putback_buf[5]; ++ unsigned int num_putback = 0; ++ ++ print_unicode_stream_body (filename, address, stream, putback_buf, num_putback, print_buf); ++ free (print_buf); ++} + + /* Find the strings in file FILENAME, read from STREAM. + Assume that STREAM is positioned so that the next byte read + is at address ADDRESS in the file. +- Stop reading at address STOP_POINT in the file, if nonzero. + + If STREAM is NULL, do not read from it. + The caller can supply a buffer of characters +@@ -516,20 +1184,29 @@ get_char (FILE *stream, file_ptr *addres + + static void + print_strings (const char *filename, FILE *stream, file_ptr address, +- int stop_point, int magiccount, char *magic) ++ int stop_point, int magiccount, char *magic) + { ++ if (unicode_display != unicode_default) ++ { ++ if (magic != NULL) ++ print_unicode_buffer (filename, address, ++ (const unsigned char *) magic, magiccount); ++ ++ if (stream != NULL) ++ print_unicode_stream (filename, address, stream); ++ return; ++ } ++ + char *buf = (char *) xmalloc (sizeof (char) * (string_min + 1)); + + while (1) + { + file_ptr start; +- int i; ++ unsigned int i; + long c; + + /* See if the next `string_min' chars are all graphic chars. */ + tryline: +- if (stop_point && address >= stop_point) +- break; + start = address; + for (i = 0; i < string_min; i++) + { +@@ -659,6 +1336,8 @@ usage (FILE *stream, int status) + -T --target= Specify the binary file format\n\ + -e --encoding={s,S,b,l,B,L} Select character size and endianness:\n\ + s = 7-bit, S = 8-bit, {b,l} = 16-bit, {B,L} = 32-bit\n\ ++ --unicode={default|show|invalid|hex|escape|highlight}\n\ ++ -U {d|s|i|x|e|h} Specify how to treat UTF-8 encoded unicode characters\n\ + -s --output-separator= String used to separate strings in output.\n\ + @ Read options from \n\ + -h --help Display this information\n\ +--- binutils.orig/binutils/readelf.c 2021-10-20 12:39:41.715008273 +0100 ++++ binutils-2.30/binutils/readelf.c 2021-10-20 14:05:42.242326079 +0100 +@@ -287,6 +287,18 @@ typedef enum print_mode + } + print_mode; + ++typedef enum unicode_display_type ++{ ++ unicode_locale, ++ unicode_escape, ++ unicode_hex, ++ unicode_highlight, ++ unicode_invalid ++} unicode_display_type; ++ ++static unicode_display_type unicode_display = unicode_locale; ++ ++ + /* Versioned symbol info. */ + enum versioned_symbol_info + { +@@ -538,11 +550,18 @@ print_symbol (signed int width, const ch + if (c == 0) + break; + +- /* Do not print control characters directly as they can affect terminal +- settings. Such characters usually appear in the names generated +- by the assembler for local labels. */ +- if (ISCNTRL (c)) ++ if (ISPRINT (c)) ++ { ++ putchar (c); ++ width_remaining --; ++ num_printed ++; ++ } ++ else if (ISCNTRL (c)) + { ++ /* Do not print control characters directly as they can affect terminal ++ settings. Such characters usually appear in the names generated ++ by the assembler for local labels. */ ++ + if (width_remaining < 2) + break; + +@@ -550,11 +569,135 @@ print_symbol (signed int width, const ch + width_remaining -= 2; + num_printed += 2; + } +- else if (ISPRINT (c)) ++ else if (c == 0x7f) + { +- putchar (c); +- width_remaining --; +- num_printed ++; ++ if (width_remaining < 5) ++ break; ++ printf (""); ++ width_remaining -= 5; ++ num_printed += 5; ++ } ++ else if (unicode_display != unicode_locale) ++ { ++ /* Display unicode characters as something else. */ ++ unsigned char bytes[4]; ++ bfd_boolean is_utf8; ++ unsigned int nbytes; ++ ++ bytes[0] = c; ++ ++ if (bytes[0] < 0xc0) ++ { ++ nbytes = 1; ++ is_utf8 = FALSE; ++ } ++ else ++ { ++ bytes[1] = *symbol++; ++ ++ if ((bytes[1] & 0xc0) != 0x80) ++ { ++ is_utf8 = FALSE; ++ /* Do not consume this character. It may only ++ be the first byte in the sequence that was ++ corrupt. */ ++ --symbol; ++ nbytes = 1; ++ } ++ else if ((bytes[0] & 0x20) == 0) ++ { ++ is_utf8 = TRUE; ++ nbytes = 2; ++ } ++ else ++ { ++ bytes[2] = *symbol++; ++ ++ if ((bytes[2] & 0xc0) != 0x80) ++ { ++ is_utf8 = FALSE; ++ symbol -= 2; ++ nbytes = 1; ++ } ++ else if ((bytes[0] & 0x10) == 0) ++ { ++ is_utf8 = TRUE; ++ nbytes = 3; ++ } ++ else ++ { ++ bytes[3] = *symbol++; ++ ++ nbytes = 4; ++ ++ if ((bytes[3] & 0xc0) != 0x80) ++ { ++ is_utf8 = FALSE; ++ symbol -= 3; ++ nbytes = 1; ++ } ++ else ++ is_utf8 = TRUE; ++ } ++ } ++ } ++ ++ if (unicode_display == unicode_invalid) ++ is_utf8 = FALSE; ++ ++ if (unicode_display == unicode_hex || ! is_utf8) ++ { ++ unsigned int i; ++ ++ if (width_remaining < (nbytes * 2) + 2) ++ break; ++ ++ putchar (is_utf8 ? '<' : '{'); ++ for (i = 0; i < nbytes; i++) ++ printf ("%02x", bytes[i]); ++ putchar (is_utf8 ? '>' : '}'); ++ } ++ else ++ { ++ if (unicode_display == unicode_highlight && isatty (1)) ++ printf ("\x1B[31;47m"); /* Red. */ ++ ++ switch (nbytes) ++ { ++ case 2: ++ if (width_remaining < 6) ++ break; ++ printf ("\\u%02x%02x", ++ (bytes[0] & 0x1c) >> 2, ++ ((bytes[0] & 0x03) << 6) | (bytes[1] & 0x3f)); ++ break; ++ case 3: ++ if (width_remaining < 6) ++ break; ++ printf ("\\u%02x%02x", ++ ((bytes[0] & 0x0f) << 4) | ((bytes[1] & 0x3c) >> 2), ++ ((bytes[1] & 0x03) << 6) | (bytes[2] & 0x3f)); ++ break; ++ case 4: ++ if (width_remaining < 8) ++ break; ++ printf ("\\u%02x%02x%02x", ++ ((bytes[0] & 0x07) << 6) | ((bytes[1] & 0x3c) >> 2), ++ ((bytes[1] & 0x03) << 6) | ((bytes[2] & 0x3c) >> 2), ++ ((bytes[2] & 0x03) << 6) | (bytes[3] & 0x3f)); ++ ++ break; ++ default: ++ /* URG. */ ++ break; ++ } ++ ++ if (unicode_display == unicode_highlight && isatty (1)) ++ printf ("\033[0m"); /* Default colour. */ ++ } ++ ++ if (bytes[nbytes - 1] == 0) ++ break; + } + else + { +@@ -4375,6 +4518,7 @@ static struct option options[] = + {"dwarf-start", required_argument, 0, OPTION_DWARF_START}, + {"dwarf-check", no_argument, 0, OPTION_DWARF_CHECK}, + ++ {"unicode", required_argument, 0, 'U'}, + {"version", no_argument, 0, 'v'}, + {"wide", no_argument, 0, 'W'}, + {"help", no_argument, 0, 'H'}, +@@ -4432,6 +4576,11 @@ usage (FILE * stream) + fprintf (stream, _("\ + -I --histogram Display histogram of bucket list lengths\n\ + -W --wide Allow output width to exceed 80 characters\n\ ++ -U --unicode=[locale|escape|hex|highlight|invalid]\n\ ++ Display unicode characters as determined by the current locale\n\ ++ (default), escape sequences, \"\", highlighted\n\ ++ escape sequences, or treat them as invalid and display as\n\ ++ \"{hex sequences}\"\n\ + @ Read options from \n\ + -H --help Display this information\n\ + -v --version Display the version number of readelf\n")); +@@ -4525,7 +4674,7 @@ parse_args (Filedata * filedata, int arg + usage (stderr); + + while ((c = getopt_long +- (argc, argv, "ADHINR:SVWacdeghi:lnp:rstuvw::x:z", options, NULL)) != EOF) ++ (argc, argv, "ADHINR:SU:VWacdeghi:lnp:rstuvw::x:z", options, NULL)) != EOF) + { + switch (c) + { +@@ -4659,6 +4808,25 @@ parse_args (Filedata * filedata, int arg + request_dump (filedata, DISASS_DUMP); + break; + #endif ++ case 'U': ++ if (optarg == NULL) ++ error (_("Missing arg to -U/--unicode")); /* Can this happen ? */ ++ else if (streq (optarg, "default") || streq (optarg, "d")) ++ unicode_display = unicode_locale; ++ else if (streq (optarg, "locale") || streq (optarg, "l")) ++ unicode_display = unicode_locale; ++ else if (streq (optarg, "escape") || streq (optarg, "e")) ++ unicode_display = unicode_escape; ++ else if (streq (optarg, "invalid") || streq (optarg, "i")) ++ unicode_display = unicode_invalid; ++ else if (streq (optarg, "hex") || streq (optarg, "x")) ++ unicode_display = unicode_hex; ++ else if (streq (optarg, "highlight") || streq (optarg, "h")) ++ unicode_display = unicode_highlight; ++ else ++ error (_("unknown argument to -U/--unicode: %s"), optarg); ++ break; ++ + case 'v': + print_version (program_name); + break; diff --git a/SPECS/binutils.spec b/SPECS/binutils.spec index d571952..106683a 100644 --- a/SPECS/binutils.spec +++ b/SPECS/binutils.spec @@ -43,7 +43,7 @@ Summary: A GNU collection of binary utilities Name: binutils%{?name_cross}%{?_with_debug:-debug} Version: 2.30 -Release: 108%{?dist} +Release: 108%{?dist}.1 License: GPLv3+ URL: https://sourceware.org/binutils @@ -591,6 +591,10 @@ Patch91: binutils-clearing-discarded-relocs.patch # Lifetime: Fixed in 2.37 (maybe) Patch92: binutils-gold-i386-gnu-property-notes.patch +# Purpose: Add options to control the display of multibyte characters. CVE 2021-42574 +# Lifetime: Fixed in 2.38 (maybe) +Patch93: binutils.unicode.patch + #---------------------------------------------------------------------------- Provides: bundled(libiberty) @@ -820,6 +824,7 @@ using libelf instead of BFD. %patch90 -p1 %patch91 -p1 %patch92 -p1 +%patch93 -p1 # We cannot run autotools as there is an exact requirement of autoconf-2.59. # FIXME - this is no longer true. Maybe try reinstating autotool use ? @@ -1269,6 +1274,9 @@ exit 0 #---------------------------------------------------------------------------- %changelog +* Wed Oct 20 2021 Nick Clifton - 2.30-108.1 +- Add ability to control the display of unicode characters. (#2009172) + * Wed Jul 07 2021 Nick Clifton - 2.30-108 - Fix thinko in previous delta. (#1970961)