{"diffoscope-json-version": 1, "source1": "/srv/reproducible-results/rbuild-debian/r-b-build.R1chK8H3/b1/binutils-mingw-w64_12_i386.changes", "source2": "/srv/reproducible-results/rbuild-debian/r-b-build.R1chK8H3/b2/binutils-mingw-w64_12_i386.changes", "unified_diff": null, "details": [{"source1": "Files", "source2": "Files", "unified_diff": "@@ -1,8 +1,8 @@\n \n f55d2758268170187e71172cb8df2aa1 24141444 debug optional binutils-mingw-w64-i686-dbgsym_2.44-2+12_i386.deb\n- c4a76a253d41d76a4dc49ad9d93f4551 2883508 devel optional binutils-mingw-w64-i686_2.44-2+12_i386.deb\n+ a18b61fe0fb23f23fcbe33f477ebc1c3 2883640 devel optional binutils-mingw-w64-i686_2.44-2+12_i386.deb\n 29a0d24b1288c91803a8342c6763e48d 26533200 debug optional binutils-mingw-w64-ucrt64-dbgsym_2.44-2+12_i386.deb\n- 4b3783fc52f1d72242e8862af5531351 3624108 devel optional binutils-mingw-w64-ucrt64_2.44-2+12_i386.deb\n+ a183b758fad3409e99735a4579994f45 3624236 devel optional binutils-mingw-w64-ucrt64_2.44-2+12_i386.deb\n 808dc0ec4956297947efd2776eb34df3 27004304 debug optional binutils-mingw-w64-x86-64-dbgsym_2.44-2+12_i386.deb\n- e9fc1ea34563f50c09bf43d234f590df 3622088 devel optional binutils-mingw-w64-x86-64_2.44-2+12_i386.deb\n+ a10b8b73b94f5fd5fe576ece6933e178 3622232 devel optional binutils-mingw-w64-x86-64_2.44-2+12_i386.deb\n 7c287b3633826f2385af78190c46604a 149224 devel optional binutils-mingw-w64_2.44-2+12_all.deb\n"}, {"source1": "binutils-mingw-w64-i686_2.44-2+12_i386.deb", "source2": "binutils-mingw-w64-i686_2.44-2+12_i386.deb", "unified_diff": null, "details": [{"source1": "file list", "source2": "file list", "unified_diff": "@@ -1,3 +1,3 @@\n -rw-r--r-- 0 0 0 4 2024-10-26 18:03:55.000000 debian-binary\n -rw-r--r-- 0 0 0 2008 2024-10-26 18:03:55.000000 control.tar.xz\n--rw-r--r-- 0 0 0 2881308 2024-10-26 18:03:55.000000 data.tar.xz\n+-rw-r--r-- 0 0 0 2881440 2024-10-26 18:03:55.000000 data.tar.xz\n"}, {"source1": "control.tar.xz", "source2": "control.tar.xz", "unified_diff": null, "details": [{"source1": "control.tar", "source2": "control.tar", "unified_diff": null, "details": [{"source1": "./md5sums", "source2": "./md5sums", "unified_diff": null, "details": [{"source1": "./md5sums", "source2": "./md5sums", "comments": ["Files differ"], "unified_diff": null}]}]}]}, {"source1": "data.tar.xz", "source2": "data.tar.xz", "unified_diff": null, "details": [{"source1": "data.tar", "source2": "data.tar", "unified_diff": null, "details": [{"source1": "file list", "source2": "file list", "unified_diff": "@@ -49,15 +49,15 @@\n -rw-r--r-- 0 root (0) root (0) 7222 2024-10-26 18:03:55.000000 ./usr/share/man/man1/i686-w64-mingw32-ar.1.gz\n -rw-r--r-- 0 root (0) root (0) 29664 2024-10-26 18:03:55.000000 ./usr/share/man/man1/i686-w64-mingw32-as.1.gz\n -rw-r--r-- 0 root (0) root (0) 3765 2024-10-26 18:03:55.000000 ./usr/share/man/man1/i686-w64-mingw32-c++filt.1.gz\n -rw-r--r-- 0 root (0) root (0) 5720 2024-10-26 18:03:55.000000 ./usr/share/man/man1/i686-w64-mingw32-dlltool.1.gz\n -rw-r--r-- 0 root (0) root (0) 311 2024-10-26 18:03:55.000000 ./usr/share/man/man1/i686-w64-mingw32-dllwrap.1.gz\n -rw-r--r-- 0 root (0) root (0) 2557 2024-10-26 18:03:55.000000 ./usr/share/man/man1/i686-w64-mingw32-elfedit.1.gz\n -rw-r--r-- 0 root (0) root (0) 7943 2024-10-26 18:03:55.000000 ./usr/share/man/man1/i686-w64-mingw32-gprof.1.gz\n--rw-r--r-- 0 root (0) root (0) 47205 2024-10-26 18:03:55.000000 ./usr/share/man/man1/i686-w64-mingw32-ld.1.gz\n+-rw-r--r-- 0 root (0) root (0) 47328 2024-10-26 18:03:55.000000 ./usr/share/man/man1/i686-w64-mingw32-ld.1.gz\n -rw-r--r-- 0 root (0) root (0) 7084 2024-10-26 18:03:55.000000 ./usr/share/man/man1/i686-w64-mingw32-nm.1.gz\n -rw-r--r-- 0 root (0) root (0) 14200 2024-10-26 18:03:55.000000 ./usr/share/man/man1/i686-w64-mingw32-objcopy.1.gz\n -rw-r--r-- 0 root (0) root (0) 14901 2024-10-26 18:03:55.000000 ./usr/share/man/man1/i686-w64-mingw32-objdump.1.gz\n -rw-r--r-- 0 root (0) root (0) 2144 2024-10-26 18:03:55.000000 ./usr/share/man/man1/i686-w64-mingw32-ranlib.1.gz\n -rw-r--r-- 0 root (0) root (0) 8507 2024-10-26 18:03:55.000000 ./usr/share/man/man1/i686-w64-mingw32-readelf.1.gz\n -rw-r--r-- 0 root (0) root (0) 3015 2024-10-26 18:03:55.000000 ./usr/share/man/man1/i686-w64-mingw32-size.1.gz\n -rw-r--r-- 0 root (0) root (0) 3806 2024-10-26 18:03:55.000000 ./usr/share/man/man1/i686-w64-mingw32-strings.1.gz\n"}, {"source1": "./usr/share/man/man1/i686-w64-mingw32-ld.1.gz", "source2": "./usr/share/man/man1/i686-w64-mingw32-ld.1.gz", "unified_diff": null, "details": [{"source1": "i686-w64-mingw32-ld.1", "source2": "i686-w64-mingw32-ld.1", "unified_diff": "@@ -1,9 +1,9 @@\n .\\\" -*- mode: troff; coding: utf-8 -*-\n-.\\\" Automatically generated by Pod::Man 5.0102 (Pod::Simple 3.45)\n+.\\\" Automatically generated by Pod::Man v6.0.2 (Pod::Simple 3.45)\n .\\\"\n .\\\" Standard preamble:\n .\\\" ========================================================================\n .de Sp \\\" Vertical space (when we can't use .PP)\n .if t .sp .5v\n .if n .sp\n ..\n@@ -48,18 +48,21 @@\n . if !\\nF==2 \\{\\\n . nr % 0\n . nr F 2\n . \\}\n . \\}\n .\\}\n .rr rF\n+.\\\"\n+.\\\" Required to disable full justification in groff 1.23.0.\n+.if n .ds AD l\n .\\\" ========================================================================\n .\\\"\n .IX Title \"LD 1\"\n-.TH LD 1 2024-10-26 binutils-2.44 \"GNU Development Tools\"\n+.TH LD 1 2025-02-02 binutils-2.44 \"GNU Development Tools\"\n .\\\" For nroff, turn off justification. Always turn off hyphenation; it makes\n .\\\" way too many mistakes in technical documents.\n .if n .ad l\n .nh\n .SH NAME\n ld \\- The GNU linker\n .SH SYNOPSIS\n@@ -68,15 +71,15 @@\n .SH DESCRIPTION\n .IX Header \"DESCRIPTION\"\n \\&\\fBld\\fR combines a number of object and archive files, relocates\n their data and ties up symbol references. Usually the last step in\n compiling a program is to run \\fBld\\fR.\n .PP\n \\&\\fBld\\fR accepts Linker Command Language files written in\n-a superset of AT&T's Link Editor Command Language syntax,\n+a superset of AT&T\\*(Aqs Link Editor Command Language syntax,\n to provide explicit and total control over the linking process.\n .PP\n This man page does not describe the command language; see the\n \\&\\fBld\\fR entry in \\f(CW\\*(C`info\\*(C'\\fR for full details on the command\n language and on other aspects of the GNU linker.\n .PP\n This version of \\fBld\\fR uses the general purpose BFD libraries\n@@ -92,41 +95,41 @@\n (or, in some cases, to get an output file in spite of the error).\n .PP\n The GNU linker \\fBld\\fR is meant to cover a broad range of situations,\n and to be as compatible as possible with other linkers. As a result,\n you have many choices to control its behavior.\n .SH OPTIONS\n .IX Header \"OPTIONS\"\n-The linker supports a plethora of command-line options, but in actual\n+The linker supports a plethora of command\\-line options, but in actual\n practice few of them are used in any particular context.\n For instance, a frequent use of \\fBld\\fR is to link standard Unix\n object files on a standard, supported Unix system. On such a system, to\n link a file \\f(CW\\*(C`hello.o\\*(C'\\fR:\n .PP\n .Vb 1\n \\& ld \\-o /lib/crt0.o hello.o \\-lc\n .Ve\n .PP\n This tells \\fBld\\fR to produce a file called \\fIoutput\\fR as the\n result of linking the file \\f(CW\\*(C`/lib/crt0.o\\*(C'\\fR with \\f(CW\\*(C`hello.o\\*(C'\\fR and\n the library \\f(CW\\*(C`libc.a\\*(C'\\fR, which will come from the standard search\n directories. (See the discussion of the \\fB\\-l\\fR option below.)\n .PP\n-Some of the command-line options to \\fBld\\fR may be specified at any\n+Some of the command\\-line options to \\fBld\\fR may be specified at any\n point in the command line. However, options which refer to files, such\n as \\fB\\-l\\fR or \\fB\\-T\\fR, cause the file to be read at the point at\n which the option appears in the command line, relative to the object\n-files and other file options. Repeating non-file options with a\n+files and other file options. Repeating non\\-file options with a\n different argument will either have no further effect, or override prior\n occurrences (those further to the left on the command line) of that\n option. Options which may be meaningfully specified more than once are\n noted in the descriptions below.\n .PP\n-Non-option arguments are object files or archives which are to be linked\n-together. They may follow, precede, or be mixed in with command-line\n+Non\\-option arguments are object files or archives which are to be linked\n+together. They may follow, precede, or be mixed in with command\\-line\n options, except that an object file argument may not be placed between\n an option and its argument.\n .PP\n Usually the linker is invoked with at least one object file, but you can\n specify other forms of binary input files using \\fB\\-l\\fR, \\fB\\-R\\fR,\n and the script command language. If \\fIno\\fR binary input files at all\n are specified, the linker does not produce any output, and issues the\n@@ -148,53 +151,53 @@\n option arguments must either follow the option letter without intervening\n whitespace, or be given as separate arguments immediately following the\n option that requires them.\n .PP\n For options whose names are multiple letters, either one dash or two can\n precede the option name; for example, \\fB\\-trace\\-symbol\\fR and\n \\&\\fB\\-\\-trace\\-symbol\\fR are equivalent. Note\\-\\-\\-there is one exception to\n-this rule. Multiple letter options that start with a lower case 'o' can\n+this rule. Multiple letter options that start with a lower case \\*(Aqo\\*(Aq can\n only be preceded by two dashes. This is to reduce confusion with the\n \\&\\fB\\-o\\fR option. So for example \\fB\\-omagic\\fR sets the output file\n name to \\fBmagic\\fR whereas \\fB\\-\\-omagic\\fR sets the NMAGIC flag on the\n output.\n .PP\n-Arguments to multiple-letter options must either be separated from the\n+Arguments to multiple\\-letter options must either be separated from the\n option name by an equals sign, or be given as separate arguments\n immediately following the option that requires them. For example,\n \\&\\fB\\-\\-trace\\-symbol foo\\fR and \\fB\\-\\-trace\\-symbol=foo\\fR are equivalent.\n-Unique abbreviations of the names of multiple-letter options are\n+Unique abbreviations of the names of multiple\\-letter options are\n accepted.\n .PP\n Note\\-\\-\\-if the linker is being invoked indirectly, via a compiler driver\n-(e.g. \\fBgcc\\fR) then all the linker command-line options should be\n+(e.g. \\fBgcc\\fR) then all the linker command\\-line options should be\n prefixed by \\fB\\-Wl,\\fR (or whatever is appropriate for the particular\n compiler driver) like this:\n .PP\n .Vb 1\n \\& gcc \\-Wl,\\-\\-start\\-group foo.o bar.o \\-Wl,\\-\\-end\\-group\n .Ve\n .PP\n This is important, because otherwise the compiler driver program may\n silently drop the linker options, resulting in a bad link. Confusion\n may also arise when passing options that require values through a\n driver, as the use of a space between option and argument acts as\n a separator, and causes the driver to pass only the option to the linker\n and the argument to the compiler. In this case, it is simplest to use\n-the joined forms of both single\\- and multiple-letter options, such as:\n+the joined forms of both single\\- and multiple\\-letter options, such as:\n .PP\n .Vb 1\n \\& gcc foo.o bar.o \\-Wl,\\-eENTRY \\-Wl,\\-Map=a.map\n .Ve\n .PP\n-Here is a table of the generic command-line switches accepted by the GNU\n+Here is a table of the generic command\\-line switches accepted by the GNU\n linker:\n .IP \\fB@\\fR\\fIfile\\fR 4\n .IX Item \"@file\"\n-Read command-line options from \\fIfile\\fR. The options read are\n+Read command\\-line options from \\fIfile\\fR. The options read are\n inserted in place of the original @\\fIfile\\fR option. If \\fIfile\\fR\n does not exist, or cannot be read, then the option will be treated\n literally, and not removed.\n .Sp\n Options in \\fIfile\\fR are separated by whitespace. A whitespace\n character may be included in an option by surrounding the entire\n option in either single or double quotes. Any character (including a\n@@ -212,57 +215,57 @@\n .IX Item \"--audit AUDITLIB\"\n Adds \\fIAUDITLIB\\fR to the \\f(CW\\*(C`DT_AUDIT\\*(C'\\fR entry of the dynamic section.\n \\&\\fIAUDITLIB\\fR is not checked for existence, nor will it use the DT_SONAME\n specified in the library. If specified multiple times \\f(CW\\*(C`DT_AUDIT\\*(C'\\fR\n will contain a colon separated list of audit interfaces to use. If the linker\n finds an object with an audit entry while searching for shared libraries,\n it will add a corresponding \\f(CW\\*(C`DT_DEPAUDIT\\*(C'\\fR entry in the output file.\n-This option is only meaningful on ELF platforms supporting the rtld-audit\n+This option is only meaningful on ELF platforms supporting the rtld\\-audit\n interface.\n-.IP \"\\fB\\-b\\fR \\fIinput-format\\fR\" 4\n+.IP \"\\fB\\-b\\fR \\fIinput\\-format\\fR\" 4\n .IX Item \"-b input-format\"\n .PD 0\n-.IP \\fB\\-\\-format=\\fR\\fIinput-format\\fR 4\n+.IP \\fB\\-\\-format=\\fR\\fIinput\\-format\\fR 4\n .IX Item \"--format=input-format\"\n .PD\n \\&\\fBld\\fR may be configured to support more than one kind of object\n file. If your \\fBld\\fR is configured this way, you can use the\n \\&\\fB\\-b\\fR option to specify the binary format for input object files\n that follow this option on the command line. Even when \\fBld\\fR is\n-configured to support alternative object formats, you don't usually need\n+configured to support alternative object formats, you don\\*(Aqt usually need\n to specify this, as \\fBld\\fR should be configured to expect as a\n default input format the most usual format on each machine.\n-\\&\\fIinput-format\\fR is a text string, the name of a particular format\n+\\&\\fIinput\\-format\\fR is a text string, the name of a particular format\n supported by the BFD libraries. (You can list the available binary\n formats with \\fBobjdump \\-i\\fR.)\n .Sp\n You may want to use this option if you are linking files with an unusual\n binary format. You can also use \\fB\\-b\\fR to switch formats explicitly (when\n linking object files of different formats), by including\n-\\&\\fB\\-b\\fR \\fIinput-format\\fR before each group of object files in a\n+\\&\\fB\\-b\\fR \\fIinput\\-format\\fR before each group of object files in a\n particular format.\n .Sp\n The default format is taken from the environment variable\n \\&\\f(CW\\*(C`GNUTARGET\\*(C'\\fR.\n .Sp\n You can also define the input format from a script, using the command\n \\&\\f(CW\\*(C`TARGET\\*(C'\\fR;\n-.IP \"\\fB\\-c\\fR \\fIMRI-commandfile\\fR\" 4\n+.IP \"\\fB\\-c\\fR \\fIMRI\\-commandfile\\fR\" 4\n .IX Item \"-c MRI-commandfile\"\n .PD 0\n-.IP \\fB\\-\\-mri\\-script=\\fR\\fIMRI-commandfile\\fR 4\n+.IP \\fB\\-\\-mri\\-script=\\fR\\fIMRI\\-commandfile\\fR 4\n .IX Item \"--mri-script=MRI-commandfile\"\n .PD\n For compatibility with linkers produced by MRI, \\fBld\\fR accepts script\n files written in an alternate, restricted command language, described in\n the MRI Compatible Script Files section of GNU ld documentation.\n Introduce MRI script files with\n the option \\fB\\-c\\fR; use the \\fB\\-T\\fR option to run linker\n-scripts written in the general-purpose \\fBld\\fR scripting language.\n-If \\fIMRI-cmdfile\\fR does not exist, \\fBld\\fR looks for it in the directories\n+scripts written in the general\\-purpose \\fBld\\fR scripting language.\n+If \\fIMRI\\-cmdfile\\fR does not exist, \\fBld\\fR looks for it in the directories\n specified by any \\fB\\-L\\fR options.\n .IP \\fB\\-d\\fR 4\n .IX Item \"-d\"\n .PD 0\n .IP \\fB\\-dc\\fR 4\n .IX Item \"-dc\"\n .IP \\fB\\-dp\\fR 4\n@@ -278,15 +281,15 @@\n .IP \"\\fB\\-P\\fR \\fIAUDITLIB\\fR\" 4\n .IX Item \"-P AUDITLIB\"\n .PD\n Adds \\fIAUDITLIB\\fR to the \\f(CW\\*(C`DT_DEPAUDIT\\*(C'\\fR entry of the dynamic section.\n \\&\\fIAUDITLIB\\fR is not checked for existence, nor will it use the DT_SONAME\n specified in the library. If specified multiple times \\f(CW\\*(C`DT_DEPAUDIT\\*(C'\\fR\n will contain a colon separated list of audit interfaces to use. This\n-option is only meaningful on ELF platforms supporting the rtld-audit interface.\n+option is only meaningful on ELF platforms supporting the rtld\\-audit interface.\n The \\-P option is provided for Solaris compatibility.\n .IP \\fB\\-\\-enable\\-linker\\-version\\fR 4\n .IX Item \"--enable-linker-version\"\n Enables the \\f(CW\\*(C`LINKER_VERSION\\*(C'\\fR linker script directive, described\n in \\fBOutput Section Data\\fR. If this directive is used in a linker\n script and this option has been enabled then a string containing the\n linker version will be inserted at the current point.\n@@ -300,15 +303,15 @@\n does not insert a version string. This is the default.\n .IP \\fB\\-\\-enable\\-non\\-contiguous\\-regions\\fR 4\n .IX Item \"--enable-non-contiguous-regions\"\n This option avoids generating an error if an input section does not\n fit a matching output section. The linker tries to allocate the input\n section to subseque nt matching output sections, and generates an\n error only if no output section is large enough. This is useful when\n-several non-contiguous memory regions are available and the input\n+several non\\-contiguous memory regions are available and the input\n section does not require a particular one. The order in which input\n sections are evaluated does not change, for instance:\n .Sp\n .Vb 10\n \\& MEMORY {\n \\& MEM1 (rwx) : ORIGIN = 0x1000, LENGTH = 0x14\n \\& MEM2 (rwx) : ORIGIN = 0x1000, LENGTH = 0x40\n@@ -364,16 +367,16 @@\n .IX Item \"--exclude-modules-for-implib module,module,...\"\n Specifies a list of object files or archive members, from which symbols\n should not be automatically exported, but which should be copied wholesale\n into the import library being generated during the link. The module names\n may be delimited by commas or colons, and must match exactly the filenames\n used by \\fBld\\fR to open the files; for archive members, this is simply\n the member name, but for object files the name listed must include and\n-match precisely any path used to specify the input file on the linker's\n-command-line. This option is available only for the i386 PE targeted port\n+match precisely any path used to specify the input file on the linker\\*(Aqs\n+command\\-line. This option is available only for the i386 PE targeted port\n of the linker. Symbols explicitly listed in a .def file are still exported,\n regardless of this option.\n .IP \\fB\\-E\\fR 4\n .IX Item \"-E\"\n .PD 0\n .IP \\fB\\-\\-export\\-dynamic\\fR 4\n .IX Item \"--export-dynamic\"\n@@ -404,28 +407,28 @@\n the description of \\fB\\-\\-export\\-all\\-symbols\\fR below.\n .IP \\fB\\-\\-export\\-dynamic\\-symbol=\\fR\\fIglob\\fR 4\n .IX Item \"--export-dynamic-symbol=glob\"\n When creating a dynamically linked executable, symbols matching\n \\&\\fIglob\\fR will be added to the dynamic symbol table. When creating a\n shared library, references to symbols matching \\fIglob\\fR will not be\n bound to the definitions within the shared library. This option is a\n-no-op when creating a shared library and \\fB\\-Bsymbolic\\fR or\n+no\\-op when creating a shared library and \\fB\\-Bsymbolic\\fR or\n \\&\\fB\\-\\-dynamic\\-list\\fR are not specified. This option is only meaningful\n on ELF platforms which support shared libraries.\n .IP \\fB\\-\\-export\\-dynamic\\-symbol\\-list=\\fR\\fIfile\\fR 4\n .IX Item \"--export-dynamic-symbol-list=file\"\n Specify a \\fB\\-\\-export\\-dynamic\\-symbol\\fR for each pattern in the file.\n The format of the file is the same as the version node without\n scope and node name. See \\fBVERSION\\fR for more information.\n .IP \\fB\\-EB\\fR 4\n .IX Item \"-EB\"\n-Link big-endian objects. This affects the default output format.\n+Link big\\-endian objects. This affects the default output format.\n .IP \\fB\\-EL\\fR 4\n .IX Item \"-EL\"\n-Link little-endian objects. This affects the default output format.\n+Link little\\-endian objects. This affects the default output format.\n .IP \"\\fB\\-f\\fR \\fIname\\fR\" 4\n .IX Item \"-f name\"\n .PD 0\n .IP \\fB\\-\\-auxiliary=\\fR\\fIname\\fR 4\n .IX Item \"--auxiliary=name\"\n .PD\n When creating an ELF shared object, set the internal DT_AUXILIARY field\n@@ -437,15 +440,15 @@\n run the program, the dynamic linker will see the DT_AUXILIARY field. If\n the dynamic linker resolves any symbols from the filter object, it will\n first check whether there is a definition in the shared object\n \\&\\fIname\\fR. If there is one, it will be used instead of the definition\n in the filter object. The shared object \\fIname\\fR need not exist.\n Thus the shared object \\fIname\\fR may be used to provide an alternative\n implementation of certain functions, perhaps for debugging or for\n-machine-specific performance.\n+machine\\-specific performance.\n .Sp\n This option may be specified more than once. The DT_AUXILIARY entries\n will be created in the order in which they appear on the command line.\n .IP \"\\fB\\-F\\fR \\fIname\\fR\" 4\n .IX Item \"-F name\"\n .PD 0\n .IP \\fB\\-\\-filter=\\fR\\fIname\\fR 4\n@@ -461,15 +464,15 @@\n dynamic linker will resolve symbols according to the symbol table of the\n filter object as usual, but it will actually link to the definitions\n found in the shared object \\fIname\\fR. Thus the filter object can be\n used to select a subset of the symbols provided by the object\n \\&\\fIname\\fR.\n .Sp\n Some older linkers used the \\fB\\-F\\fR option throughout a compilation\n-toolchain for specifying object-file format for both input and output\n+toolchain for specifying object\\-file format for both input and output\n object files.\n The GNU linker uses other mechanisms for this purpose: the\n \\&\\fB\\-b\\fR, \\fB\\-\\-format\\fR, \\fB\\-\\-oformat\\fR options, the\n \\&\\f(CW\\*(C`TARGET\\*(C'\\fR command in linker scripts, and the \\f(CW\\*(C`GNUTARGET\\*(C'\\fR\n environment variable.\n The GNU linker will ignore the \\fB\\-F\\fR option when not\n creating an ELF shared object.\n@@ -698,15 +701,15 @@\n .Sp\n .Vb 1\n \\& Removed property 0xc0000002 to merge foo.o (0x1) and bar.o (not found)\n .Ve\n .Sp\n This indicates that property 0xc0000002 is removed from output when\n merging properties in \\fIfoo.o\\fR, whose property 0xc0000002 value\n-is 0x1, and \\fIbar.o\\fR, which doesn't have property 0xc0000002.\n+is 0x1, and \\fIbar.o\\fR, which doesn\\*(Aqt have property 0xc0000002.\n .Sp\n .Vb 1\n \\& Updated property 0xc0010001 (0x1) to merge foo.o (0x1) and bar.o (0x1)\n .Ve\n .Sp\n This indicates that property 0xc0010001 value is updated to 0x1 in output\n when merging properties in \\fIfoo.o\\fR, whose 0xc0010001 property value\n@@ -754,24 +757,24 @@\n .IP \\fB\\-N\\fR 4\n .IX Item \"-N\"\n .PD 0\n .IP \\fB\\-\\-omagic\\fR 4\n .IX Item \"--omagic\"\n .PD\n Set the text and data sections to be readable and writable. Also, do\n-not page-align the data segment, and disable linking against shared\n+not page\\-align the data segment, and disable linking against shared\n libraries. If the output format supports Unix style magic numbers,\n mark the output as \\f(CW\\*(C`OMAGIC\\*(C'\\fR. Note: Although a writable text section\n-is allowed for PE-COFF targets, it does not conform to the format\n+is allowed for PE\\-COFF targets, it does not conform to the format\n specification published by Microsoft.\n .IP \\fB\\-\\-no\\-omagic\\fR 4\n .IX Item \"--no-omagic\"\n This option negates most of the effects of the \\fB\\-N\\fR option. It\n-sets the text section to be read-only, and forces the data segment to\n-be page-aligned. Note \\- this option does not enable linking against\n+sets the text section to be read\\-only, and forces the data segment to\n+be page\\-aligned. Note \\- this option does not enable linking against\n shared libraries. Use \\fB\\-Bdynamic\\fR for this.\n .IP \"\\fB\\-o\\fR \\fIoutput\\fR\" 4\n .IX Item \"-o output\"\n .PD 0\n .IP \\fB\\-\\-output=\\fR\\fIoutput\\fR 4\n .IX Item \"--output=output\"\n .PD\n@@ -800,29 +803,29 @@\n files (the first one in the \\fBOBJS\\fR list) whereas the second\n version of the linker command will generate an error message and not\n delete anything.\n .IP \\fB\\-\\-dependency\\-file=\\fR\\fIdepfile\\fR 4\n .IX Item \"--dependency-file=depfile\"\n Write a \\fIdependency file\\fR to \\fIdepfile\\fR. This file contains a rule\n suitable for \\f(CW\\*(C`make\\*(C'\\fR describing the output file and all the input files\n-that were read to produce it. The output is similar to the compiler's\n-output with \\fB\\-M \\-MP\\fR. Note that there is no option like the compiler's \\fB\\-MM\\fR,\n-to exclude \"system files\" (which is not a well-specified concept in the\n+that were read to produce it. The output is similar to the compiler\\*(Aqs\n+output with \\fB\\-M \\-MP\\fR. Note that there is no option like the compiler\\*(Aqs \\fB\\-MM\\fR,\n+to exclude \"system files\" (which is not a well\\-specified concept in the\n linker, unlike \"system headers\" in the compiler). So the output from\n \\&\\fB\\-\\-dependency\\-file\\fR is always specific to the exact state of the\n installation where it was produced, and should not be copied into\n distributed makefiles without careful editing.\n .IP \"\\fB\\-O\\fR \\fIlevel\\fR\" 4\n .IX Item \"-O level\"\n If \\fIlevel\\fR is a numeric values greater than zero \\fBld\\fR optimizes\n the output. This might take significantly longer and therefore probably\n should only be enabled for the final binary. At the moment this\n option only affects ELF shared library generation. Future releases of\n the linker may make more use of this option. Also currently there is\n-no difference in the linker's behaviour for different non-zero values\n+no difference in the linker\\*(Aqs behaviour for different non\\-zero values\n of this option. Again this may change with future releases.\n .IP \"\\fB\\-plugin\\fR \\fIname\\fR\" 4\n .IX Item \"-plugin name\"\n Involve a plugin in the linking process. The \\fIname\\fR parameter is\n the absolute filename of the plugin. Usually this parameter is\n automatically added by the compiler, when using link time\n optimization, but users can also add their own plugins if they so\n@@ -844,15 +847,15 @@\n The option which are covered are: \\fB\\-Bdynamic\\fR, \\fB\\-Bstatic\\fR,\n \\&\\fB\\-dn\\fR, \\fB\\-dy\\fR, \\fB\\-call_shared\\fR, \\fB\\-non_shared\\fR,\n \\&\\fB\\-static\\fR, \\fB\\-N\\fR, \\fB\\-n\\fR, \\fB\\-\\-whole\\-archive\\fR,\n \\&\\fB\\-\\-no\\-whole\\-archive\\fR, \\fB\\-r\\fR, \\fB\\-Ur\\fR,\n \\&\\fB\\-\\-copy\\-dt\\-needed\\-entries\\fR, \\fB\\-\\-no\\-copy\\-dt\\-needed\\-entries\\fR,\n \\&\\fB\\-\\-as\\-needed\\fR, \\fB\\-\\-no\\-as\\-needed\\fR, and \\fB\\-a\\fR.\n .Sp\n-One target for this option are specifications for \\fIpkg-config\\fR. When\n+One target for this option are specifications for \\fIpkg\\-config\\fR. When\n used with the \\fB\\-\\-libs\\fR option all possibly needed libraries are\n listed and then possibly linked with all the time. It is better to return\n something as follows:\n .Sp\n .Vb 1\n \\& \\-Wl,\\-\\-push\\-state,\\-\\-as\\-needed \\-libone \\-libtwo \\-Wl,\\-\\-pop\\-state\n .Ve\n@@ -881,31 +884,31 @@\n .PD 0\n .IP \\fB\\-\\-relocatable\\fR 4\n .IX Item \"--relocatable\"\n .PD\n Generate relocatable output\\-\\-\\-i.e., generate an output file that can in\n turn serve as input to \\fBld\\fR. This is often called \\fIpartial\n linking\\fR. As a side effect, in environments that support standard Unix\n-magic numbers, this option also sets the output file's magic number to\n+magic numbers, this option also sets the output file\\*(Aqs magic number to\n \\&\\f(CW\\*(C`OMAGIC\\*(C'\\fR.\n If this option is not specified, an absolute file is produced. When\n linking C++ programs, this option \\fIwill not\\fR resolve references to\n constructors; to do that, use \\fB\\-Ur\\fR.\n .Sp\n When an input file does not have the same format as the output file,\n partial linking is only supported if that input file does not contain any\n relocations. Different output formats can have further restrictions; for\n example some \\f(CW\\*(C`a.out\\*(C'\\fR\\-based formats do not support partial linking\n with input files in other formats at all.\n .Sp\n When the relocatable output contains both contents which require\n-link-time optimization (LTO) and contents which don't require LTO,\n+link\\-time optimization (LTO) and contents which don\\*(Aqt require LTO,\n a .gnu_object_only section will be created to contain a relocatable\n object file, as if \\fB\\-r\\fR is applied to all relocatable inputs\n-which don't require LTO. When processing a relocatable input with\n+which don\\*(Aqt require LTO. When processing a relocatable input with\n a .gnu_object_only section, the linker will extract the .gnu_object_only\n section as a separate input.\n .Sp\n Note that since \\fB\\-r\\fR groups some sections from different input files\n together, there may be negative impacts on code size and locality in\n final executable or shared library.\n .Sp\n@@ -926,20 +929,20 @@\n the \\fB\\-rpath\\fR option.\n .IP \\fB\\-\\-rosegment\\fR 4\n .IX Item \"--rosegment\"\n .PD 0\n .IP \\fB\\-\\-no\\-rosegment\\fR 4\n .IX Item \"--no-rosegment\"\n .PD\n-Attempt to ensure that only a single read-only, non-code segment is\n+Attempt to ensure that only a single read\\-only, non\\-code segment is\n created. Only useful when used in conjunction with the \\fB\\-z\n-separate-code\\fR option. The resulting binaries should be smaller than\n-if \\fB\\-z separate-code\\fR is used on its own. Without this option,\n-or if \\fB\\-\\-no\\-rosegment\\fR is specified, the \\fB\\-z separate-code\\fR\n-option will create two read-only segments, one before the code segment\n+separate\\-code\\fR option. The resulting binaries should be smaller than\n+if \\fB\\-z separate\\-code\\fR is used on its own. Without this option,\n+or if \\fB\\-\\-no\\-rosegment\\fR is specified, the \\fB\\-z separate\\-code\\fR\n+option will create two read\\-only segments, one before the code segment\n and one after it.\n .Sp\n The name of the options are misleading, but they have been chosen in\n order for the linker to be compatible with the LLD and GOLD linkers.\n .Sp\n Thse options are only supported by ELF targets.\n .IP \\fB\\-s\\fR 4\n@@ -981,28 +984,28 @@\n .IP \"\\fB\\-T\\fR \\fIscriptfile\\fR\" 4\n .IX Item \"-T scriptfile\"\n .PD 0\n .IP \\fB\\-\\-script=\\fR\\fIscriptfile\\fR 4\n .IX Item \"--script=scriptfile\"\n .PD\n Use \\fIscriptfile\\fR as the linker script. This script replaces\n-\\&\\fBld\\fR's default linker script (rather than adding to it),\n+\\&\\fBld\\fR\\*(Aqs default linker script (rather than adding to it),\n unless the script contains \\f(CW\\*(C`INSERT\\*(C'\\fR, so \\fIcommandfile\\fR must\n specify everything necessary to describe the output file.\n .Sp\n If \\fIscriptfile\\fR does not exist in the current directory, \\f(CW\\*(C`ld\\*(C'\\fR\n looks for it in the directories specified by any preceding \\fB\\-L\\fR\n options.\n .Sp\n Command line options that appear before the \\fB\\-T\\fR option can\n affect the script, but command line options that appear after it do\n not.\n .Sp\n Multiple \\fB\\-T\\fR options will accumulate if they are augmenting the\n-current script, otherwise the last, non-augmenting, \\fB\\-T\\fR option\n+current script, otherwise the last, non\\-augmenting, \\fB\\-T\\fR option\n will be used.\n .Sp\n There are other ways of specifying linker scripts. See\n .IP \"\\fB\\-dT\\fR \\fIscriptfile\\fR\" 4\n .IX Item \"-dT scriptfile\"\n .PD 0\n .IP \\fB\\-\\-default\\-script=\\fR\\fIscriptfile\\fR 4\n@@ -1119,28 +1122,28 @@\n .IP \\fB\\-X\\fR 4\n .IX Item \"-X\"\n .PD 0\n .IP \\fB\\-\\-discard\\-locals\\fR 4\n .IX Item \"--discard-locals\"\n .PD\n Delete all temporary local symbols. (These symbols start with\n-system-specific local label prefixes, typically \\fB.L\\fR for ELF systems\n+system\\-specific local label prefixes, typically \\fB.L\\fR for ELF systems\n or \\fBL\\fR for traditional a.out systems.)\n .IP \"\\fB\\-y\\fR \\fIsymbol\\fR\" 4\n .IX Item \"-y symbol\"\n .PD 0\n .IP \\fB\\-\\-trace\\-symbol=\\fR\\fIsymbol\\fR 4\n .IX Item \"--trace-symbol=symbol\"\n .PD\n Print the name of each linked file in which \\fIsymbol\\fR appears. This\n option may be given any number of times. On many systems it is necessary\n to prepend an underscore.\n .Sp\n This option is useful when you have an undefined symbol in your link but\n-don't know where the reference is coming from.\n+don\\*(Aqt know where the reference is coming from.\n .IP \"\\fB\\-Y\\fR \\fIpath\\fR\" 4\n .IX Item \"-Y path\"\n Add \\fIpath\\fR to the default library search path. This option exists\n for Solaris compatibility.\n .IP \"\\fB\\-z\\fR \\fIkeyword\\fR\" 4\n .IX Item \"-z keyword\"\n The recognized keywords are:\n@@ -1201,26 +1204,26 @@\n .IX Item \"common-page-size=value\"\n Set the page size most commonly used to \\fIvalue\\fR. Memory image\n layout will be optimized to minimize memory pages if the system is\n using pages of this size.\n .IP \\fBdefs\\fR 4\n .IX Item \"defs\"\n Report unresolved symbol references from regular object files. This\n-is done even if the linker is creating a non-symbolic shared library.\n+is done even if the linker is creating a non\\-symbolic shared library.\n This option is the inverse of \\fB\\-z undefs\\fR.\n-.IP \\fBdynamic-undefined-weak\\fR 4\n+.IP \\fBdynamic\\-undefined\\-weak\\fR 4\n .IX Item \"dynamic-undefined-weak\"\n .PD 0\n-.IP \\fBnodynamic-undefined-weak\\fR 4\n+.IP \\fBnodynamic\\-undefined\\-weak\\fR 4\n .IX Item \"nodynamic-undefined-weak\"\n .PD\n Make undefined weak symbols dynamic when building a dynamic object,\n if they are referenced from a regular object file and not forced local\n by symbol visibility or versioning. Do not make them dynamic if\n-\\&\\fBnodynamic-undefined-weak\\fR. If neither option is given, a target\n+\\&\\fBnodynamic\\-undefined\\-weak\\fR. If neither option is given, a target\n may default to either option being in force, or make some other\n selection of undefined weak symbols dynamic. Not all targets support\n these options.\n .IP \\fBexecstack\\fR 4\n .IX Item \"execstack\"\n Marks the object as requiring executable stack.\n .IP \\fBglobal\\fR 4\n@@ -1230,38 +1233,38 @@\n of subsequently loaded libraries.\n .IP \\fBglobalaudit\\fR 4\n .IX Item \"globalaudit\"\n This option is only meaningful when building a dynamic executable.\n This option marks the executable as requiring global auditing by\n setting the \\f(CW\\*(C`DF_1_GLOBAUDIT\\*(C'\\fR bit in the \\f(CW\\*(C`DT_FLAGS_1\\*(C'\\fR dynamic\n tag. Global auditing requires that any auditing library defined via\n-the \\fB\\-\\-depaudit\\fR or \\fB\\-P\\fR command-line options be run for\n+the \\fB\\-\\-depaudit\\fR or \\fB\\-P\\fR command\\-line options be run for\n all dynamic objects loaded by the application.\n .IP \\fBibtplt\\fR 4\n .IX Item \"ibtplt\"\n Generate Intel Indirect Branch Tracking (IBT) enabled PLT entries.\n Supported for Linux/i386 and Linux/x86_64.\n .IP \\fBibt\\fR 4\n .IX Item \"ibt\"\n Generate GNU_PROPERTY_X86_FEATURE_1_IBT in .note.gnu.property section\n to indicate compatibility with IBT. This also implies \\fBibtplt\\fR.\n Supported for Linux/i386 and Linux/x86_64.\n-.IP \\fBindirect-extern-access\\fR 4\n+.IP \\fBindirect\\-extern\\-access\\fR 4\n .IX Item \"indirect-extern-access\"\n .PD 0\n-.IP \\fBnoindirect-extern-access\\fR 4\n+.IP \\fBnoindirect\\-extern\\-access\\fR 4\n .IX Item \"noindirect-extern-access\"\n .PD\n Generate GNU_PROPERTY_1_NEEDED_INDIRECT_EXTERN_ACCESS in\n \\&.note.gnu.property section to indicate that object file requires\n canonical function pointers and cannot be used with copy relocation.\n-This option also implies \\fBnoextern-protected-data\\fR and\n+This option also implies \\fBnoextern\\-protected\\-data\\fR and\n \\&\\fBnocopyreloc\\fR. Supported for i386 and x86\\-64.\n .Sp\n-\\&\\fBnoindirect-extern-access\\fR removes\n+\\&\\fBnoindirect\\-extern\\-access\\fR removes\n GNU_PROPERTY_1_NEEDED_INDIRECT_EXTERN_ACCESS from .note.gnu.property\n section.\n .IP \\fBinitfirst\\fR 4\n .IX Item \"initfirst\"\n This option is only meaningful when building a shared object.\n It marks the object so that its runtime initialization will occur\n before the runtime initialization of any other objects brought into\n@@ -1346,72 +1349,72 @@\n .IX Item \"lazy\"\n When generating an executable or shared library, mark it to tell the\n dynamic linker to defer function call resolution to the point when\n the function is called (lazy binding), rather than at load time.\n Lazy binding is the default.\n .IP \\fBloadfltr\\fR 4\n .IX Item \"loadfltr\"\n-Specify that the object's filters be processed immediately at runtime.\n+Specify that the object\\*(Aqs filters be processed immediately at runtime.\n .IP \\fBmax\\-page\\-size=\\fR\\fIvalue\\fR 4\n .IX Item \"max-page-size=value\"\n Set the maximum memory page size supported to \\fIvalue\\fR.\n-.IP \\fBmark-plt\\fR 4\n+.IP \\fBmark\\-plt\\fR 4\n .IX Item \"mark-plt\"\n .PD 0\n-.IP \\fBnomark-plt\\fR 4\n+.IP \\fBnomark\\-plt\\fR 4\n .IX Item \"nomark-plt\"\n .PD\n Mark PLT entries with dynamic tags, DT_X86_64_PLT, DT_X86_64_PLTSZ and\n-DT_X86_64_PLTENT. Since this option stores a non-zero value in the\n+DT_X86_64_PLTENT. Since this option stores a non\\-zero value in the\n r_addend field of R_X86_64_JUMP_SLOT relocations, the resulting\n executables and shared libraries are incompatible with dynamic linkers,\n such as those in older versions of glibc without the change to ignore\n r_addend in R_X86_64_GLOB_DAT and R_X86_64_JUMP_SLOT relocations, which\n-don't ignore the r_addend field of R_X86_64_JUMP_SLOT relocations.\n+don\\*(Aqt ignore the r_addend field of R_X86_64_JUMP_SLOT relocations.\n Supported for x86_64.\n .IP \\fBmuldefs\\fR 4\n .IX Item \"muldefs\"\n Allow multiple definitions.\n .IP \\fBnocopyreloc\\fR 4\n .IX Item \"nocopyreloc\"\n Disable linker generated .dynbss variables used in place of variables\n defined in shared libraries. May result in dynamic text relocations.\n .IP \\fBnodefaultlib\\fR 4\n .IX Item \"nodefaultlib\"\n Specify that the dynamic loader search for dependencies of this object\n should ignore any default library search paths.\n .IP \\fBnodelete\\fR 4\n .IX Item \"nodelete\"\n-Specify that the object shouldn't be unloaded at runtime.\n+Specify that the object shouldn\\*(Aqt be unloaded at runtime.\n .IP \\fBnodlopen\\fR 4\n .IX Item \"nodlopen\"\n Specify that the object is not available to \\f(CW\\*(C`dlopen\\*(C'\\fR.\n .IP \\fBnodump\\fR 4\n .IX Item \"nodump\"\n Specify that the object can not be dumped by \\f(CW\\*(C`dldump\\*(C'\\fR.\n .IP \\fBnoexecstack\\fR 4\n .IX Item \"noexecstack\"\n Marks the object as not requiring executable stack.\n-.IP \\fBnoextern-protected-data\\fR 4\n+.IP \\fBnoextern\\-protected\\-data\\fR 4\n .IX Item \"noextern-protected-data\"\n-Don't treat protected data symbols as external when building a shared\n+Don\\*(Aqt treat protected data symbols as external when building a shared\n library. This option overrides the linker backend default. It can be\n used to work around incorrect relocations against protected data symbols\n generated by compiler. Updates on protected data symbols by another\n-module aren't visible to the resulting shared library. Supported for\n+module aren\\*(Aqt visible to the resulting shared library. Supported for\n i386 and x86\\-64.\n-.IP \\fBnoreloc-overflow\\fR 4\n+.IP \\fBnoreloc\\-overflow\\fR 4\n .IX Item \"noreloc-overflow\"\n Disable relocation overflow check. This can be used to disable\n relocation overflow check if there will be no dynamic relocation\n-overflow at run-time. Supported for x86_64.\n-.IP \\fBmemory-seal\\fR 4\n+overflow at run\\-time. Supported for x86_64.\n+.IP \\fBmemory\\-seal\\fR 4\n .IX Item \"memory-seal\"\n .PD 0\n-.IP \\fBnomemory-seal\\fR 4\n+.IP \\fBnomemory\\-seal\\fR 4\n .IX Item \"nomemory-seal\"\n .PD\n Instruct the executable or shared library that the all PT_LOAD segments\n should be sealed to avoid further manipulation (such as changing the\n protection flags, the segment size, or remove the mapping).\n This is a security hardening that requires system support. This\n generates GNU_PROPERTY_MEMORY_SEAL in .note.gnu.property section\n@@ -1421,125 +1424,125 @@\n dynamic linker to resolve all symbols when the program is started, or\n when the shared library is loaded by dlopen, instead of deferring\n function call resolution to the point when the function is first\n called.\n .IP \\fBorigin\\fR 4\n .IX Item \"origin\"\n Specify that the object requires \\fR\\f(CB$ORIGIN\\fR\\fB\\fR handling in paths.\n-.IP \\fBpack-relative-relocs\\fR 4\n+.IP \\fBpack\\-relative\\-relocs\\fR 4\n .IX Item \"pack-relative-relocs\"\n .PD 0\n-.IP \\fBnopack-relative-relocs\\fR 4\n+.IP \\fBnopack\\-relative\\-relocs\\fR 4\n .IX Item \"nopack-relative-relocs\"\n .PD\n-Generate compact relative relocation in position-independent executable\n+Generate compact relative relocation in position\\-independent executable\n and shared library. It adds \\f(CW\\*(C`DT_RELR\\*(C'\\fR, \\f(CW\\*(C`DT_RELRSZ\\*(C'\\fR and\n \\&\\f(CW\\*(C`DT_RELRENT\\*(C'\\fR entries to the dynamic section. It is ignored when\n-building position-dependent executable and relocatable output.\n-\\&\\fBnopack-relative-relocs\\fR is the default, which disables compact\n+building position\\-dependent executable and relocatable output.\n+\\&\\fBnopack\\-relative\\-relocs\\fR is the default, which disables compact\n relative relocation. When linked against the GNU C Library, a\n GLIBC_ABI_DT_RELR symbol version dependency on the shared C Library is\n added to the output. Supported for i386 and x86\\-64.\n .IP \\fBrelro\\fR 4\n .IX Item \"relro\"\n .PD 0\n .IP \\fBnorelro\\fR 4\n .IX Item \"norelro\"\n .PD\n Create an ELF \\f(CW\\*(C`PT_GNU_RELRO\\*(C'\\fR segment header in the object. This\n-specifies a memory segment that should be made read-only after\n-relocation, if supported. Specifying \\fBcommon-page-size\\fR smaller\n+specifies a memory segment that should be made read\\-only after\n+relocation, if supported. Specifying \\fBcommon\\-page\\-size\\fR smaller\n than the system page size will render this protection ineffective.\n-Don't create an ELF \\f(CW\\*(C`PT_GNU_RELRO\\*(C'\\fR segment if \\fBnorelro\\fR.\n-.IP \\fBreport-relative-reloc\\fR 4\n+Don\\*(Aqt create an ELF \\f(CW\\*(C`PT_GNU_RELRO\\*(C'\\fR segment if \\fBnorelro\\fR.\n+.IP \\fBreport\\-relative\\-reloc\\fR 4\n .IX Item \"report-relative-reloc\"\n Report dynamic relative relocations generated by linker. Supported for\n Linux/i386 and Linux/x86_64.\n .IP \\fBsectionheader\\fR 4\n .IX Item \"sectionheader\"\n .PD 0\n .IP \\fBnosectionheader\\fR 4\n .IX Item \"nosectionheader\"\n .PD\n-Generate section header. Don't generate section header if\n+Generate section header. Don\\*(Aqt generate section header if\n \\&\\fBnosectionheader\\fR is used. \\fBsectionheader\\fR is the default.\n-.IP \\fBseparate-code\\fR 4\n+.IP \\fBseparate\\-code\\fR 4\n .IX Item \"separate-code\"\n .PD 0\n-.IP \\fBnoseparate-code\\fR 4\n+.IP \\fBnoseparate\\-code\\fR 4\n .IX Item \"noseparate-code\"\n .PD\n Create separate code \\f(CW\\*(C`PT_LOAD\\*(C'\\fR segment header in the object. This\n specifies a memory segment that should contain only instructions and must\n-be in wholly disjoint pages from any other data. Don't create separate\n-code \\f(CW\\*(C`PT_LOAD\\*(C'\\fR segment if \\fBnoseparate-code\\fR is used.\n+be in wholly disjoint pages from any other data. Don\\*(Aqt create separate\n+code \\f(CW\\*(C`PT_LOAD\\*(C'\\fR segment if \\fBnoseparate\\-code\\fR is used.\n .IP \\fBshstk\\fR 4\n .IX Item \"shstk\"\n Generate GNU_PROPERTY_X86_FEATURE_1_SHSTK in .note.gnu.property section\n to indicate compatibility with Intel Shadow Stack. Supported for\n Linux/i386 and Linux/x86_64.\n .IP \\fBstack\\-size=\\fR\\fIvalue\\fR 4\n .IX Item \"stack-size=value\"\n Specify a stack size for an ELF \\f(CW\\*(C`PT_GNU_STACK\\*(C'\\fR segment.\n-Specifying zero will override any default non-zero sized\n+Specifying zero will override any default non\\-zero sized\n \\&\\f(CW\\*(C`PT_GNU_STACK\\*(C'\\fR segment creation.\n-.IP \\fBstart-stop-gc\\fR 4\n+.IP \\fBstart\\-stop\\-gc\\fR 4\n .IX Item \"start-stop-gc\"\n .PD 0\n-.IP \\fBnostart-stop-gc\\fR 4\n+.IP \\fBnostart\\-stop\\-gc\\fR 4\n .IX Item \"nostart-stop-gc\"\n .PD\n When \\fB\\-\\-gc\\-sections\\fR is in effect, a reference from a retained\n section to \\f(CW\\*(C`_\\|_start_SECNAME\\*(C'\\fR or \\f(CW\\*(C`_\\|_stop_SECNAME\\*(C'\\fR causes all\n input sections named \\f(CW\\*(C`SECNAME\\*(C'\\fR to also be retained, if\n \\&\\f(CW\\*(C`SECNAME\\*(C'\\fR is representable as a C identifier and either\n \\&\\f(CW\\*(C`_\\|_start_SECNAME\\*(C'\\fR or \\f(CW\\*(C`_\\|_stop_SECNAME\\*(C'\\fR is synthesized by the\n-linker. \\fB\\-z start-stop-gc\\fR disables this effect, allowing\n+linker. \\fB\\-z start\\-stop\\-gc\\fR disables this effect, allowing\n sections to be garbage collected as if the special synthesized symbols\n-were not defined. \\fB\\-z start-stop-gc\\fR has no effect on a\n+were not defined. \\fB\\-z start\\-stop\\-gc\\fR has no effect on a\n definition of \\f(CW\\*(C`_\\|_start_SECNAME\\*(C'\\fR or \\f(CW\\*(C`_\\|_stop_SECNAME\\*(C'\\fR in an\n object file or linker script. Such a definition will prevent the\n linker providing a synthesized \\f(CW\\*(C`_\\|_start_SECNAME\\*(C'\\fR or\n \\&\\f(CW\\*(C`_\\|_stop_SECNAME\\*(C'\\fR respectively, and therefore the special\n treatment by garbage collection for those references.\n .IP \\fBstart\\-stop\\-visibility=\\fR\\fIvalue\\fR 4\n .IX Item \"start-stop-visibility=value\"\n Specify the ELF symbol visibility for synthesized\n \\&\\f(CW\\*(C`_\\|_start_SECNAME\\*(C'\\fR and \\f(CW\\*(C`_\\|_stop_SECNAME\\*(C'\\fR symbols. \\fIvalue\\fR must be exactly \\fBdefault\\fR,\n \\&\\fBinternal\\fR, \\fBhidden\\fR, or \\fBprotected\\fR. If no \\fB\\-z\n-start-stop-visibility\\fR option is given, \\fBprotected\\fR is used for\n-compatibility with historical practice. However, it's highly\n+start\\-stop\\-visibility\\fR option is given, \\fBprotected\\fR is used for\n+compatibility with historical practice. However, it\\*(Aqs highly\n recommended to use \\fB\\-z start\\-stop\\-visibility=hidden\\fR in new\n programs and shared libraries so that these symbols are not exported\n-between shared objects, which is not usually what's intended.\n+between shared objects, which is not usually what\\*(Aqs intended.\n .IP \\fBtext\\fR 4\n .IX Item \"text\"\n .PD 0\n .IP \\fBnotext\\fR 4\n .IX Item \"notext\"\n .IP \\fBtextoff\\fR 4\n .IX Item \"textoff\"\n .PD\n-Report an error if DT_TEXTREL is set, i.e., if the position-independent\n-or shared object has dynamic relocations in read-only sections. Don't\n+Report an error if DT_TEXTREL is set, i.e., if the position\\-independent\n+or shared object has dynamic relocations in read\\-only sections. Don\\*(Aqt\n report an error if \\fBnotext\\fR or \\fBtextoff\\fR.\n .IP \\fBundefs\\fR 4\n .IX Item \"undefs\"\n Do not report unresolved symbol references from regular object files,\n either when creating an executable, or when creating a shared library.\n This option is the inverse of \\fB\\-z defs\\fR.\n-.IP \\fBunique-symbol\\fR 4\n+.IP \\fBunique\\-symbol\\fR 4\n .IX Item \"unique-symbol\"\n .PD 0\n-.IP \\fBnounique-symbol\\fR 4\n+.IP \\fBnounique\\-symbol\\fR 4\n .IX Item \"nounique-symbol\"\n .PD\n Avoid duplicated local symbol names in the symbol string table. Append\n-\".\\f(CW\\*(C`number\\*(C'\\fR\" to duplicated local symbol names if \\fBunique-symbol\\fR\n-is used. \\fBnounique-symbol\\fR is the default.\n+\".\\f(CW\\*(C`number\\*(C'\\fR\" to duplicated local symbol names if \\fBunique\\-symbol\\fR\n+is used. \\fBnounique\\-symbol\\fR is the default.\n .IP \\fBx86\\-64\\-baseline\\fR 4\n .IX Item \"x86-64-baseline\"\n .PD 0\n .IP \\fBx86\\-64\\-v2\\fR 4\n .IX Item \"x86-64-v2\"\n .IP \\fBx86\\-64\\-v3\\fR 4\n .IX Item \"x86-64-v3\"\n@@ -1619,17 +1622,17 @@\n .PD\n This option affects ELF DT_NEEDED tags for dynamic libraries mentioned\n on the command line after the \\fB\\-\\-as\\-needed\\fR option. Normally\n the linker will add a DT_NEEDED tag for each dynamic library mentioned\n on the command line, regardless of whether the library is actually\n needed or not. \\fB\\-\\-as\\-needed\\fR causes a DT_NEEDED tag to only be\n emitted for a library that \\fIat that point in the link\\fR satisfies a\n-non-weak undefined symbol reference from a regular object file or, if\n+non\\-weak undefined symbol reference from a regular object file or, if\n the library is not found in the DT_NEEDED lists of other needed libraries, a\n-non-weak undefined symbol reference from another needed dynamic library.\n+non\\-weak undefined symbol reference from another needed dynamic library.\n Object files or libraries appearing on the command line \\fIafter\\fR\n the library in question do not affect whether the library is seen as\n needed. This is similar to the rules for extraction of object files\n from archives. \\fB\\-\\-no\\-as\\-needed\\fR restores the default behaviour.\n .Sp\n Note: On Linux based systems the \\fB\\-\\-as\\-needed\\fR option also has\n an affect on the behaviour of the \\fB\\-\\-rpath\\fR and\n@@ -1682,15 +1685,15 @@\n Do not link against shared libraries. This is only meaningful on\n platforms for which shared libraries are supported. The different\n variants of this option are for compatibility with various systems. You\n may use this option multiple times on the command line: it affects\n library searching for \\fB\\-l\\fR options which follow it. This\n option also implies \\fB\\-\\-unresolved\\-symbols=report\\-all\\fR. This\n option can be used with \\fB\\-shared\\fR. Doing so means that a\n-shared library is being created but that all of the library's external\n+shared library is being created but that all of the library\\*(Aqs external\n references must be resolved by pulling in entries from static\n libraries.\n .IP \\fB\\-Bsymbolic\\fR 4\n .IX Item \"-Bsymbolic\"\n When creating a shared library, bind references to global symbols to the\n definition within the shared library, if any. Normally, it is possible\n for a program linked against a shared library to override the definition\n@@ -1702,19 +1705,19 @@\n symbols to the definition within the shared library, if any.\n This option is only meaningful on ELF platforms which support shared\n libraries.\n .IP \\fB\\-Bno\\-symbolic\\fR 4\n .IX Item \"-Bno-symbolic\"\n This option can cancel previously specified \\fB\\-Bsymbolic\\fR and\n \\&\\fB\\-Bsymbolic\\-functions\\fR.\n-.IP \\fB\\-\\-dynamic\\-list=\\fR\\fIdynamic-list-file\\fR 4\n+.IP \\fB\\-\\-dynamic\\-list=\\fR\\fIdynamic\\-list\\-file\\fR 4\n .IX Item \"--dynamic-list=dynamic-list-file\"\n Specify the name of a dynamic list file to the linker. This is\n typically used when creating shared libraries to specify a list of\n-global symbols whose references shouldn't be bound to the definition\n+global symbols whose references shouldn\\*(Aqt be bound to the definition\n within the shared library, or creating dynamically linked executables\n to specify a list of symbols which should be added to the symbol table\n in the executable. This option is only meaningful on ELF platforms\n which support shared libraries.\n .Sp\n The format of the dynamic list is the same as the version node without\n scope and node name. See \\fBVERSION\\fR for more information.\n@@ -1735,27 +1738,27 @@\n .IX Item \"--no-check-sections\"\n .PD\n Asks the linker \\fInot\\fR to check section addresses after they have\n been assigned to see if there are any overlaps. Normally the linker will\n perform this check, and if it finds any overlaps it will produce\n suitable error messages. The linker does know about, and does make\n allowances for sections in overlays. The default behaviour can be\n-restored by using the command-line switch \\fB\\-\\-check\\-sections\\fR.\n+restored by using the command\\-line switch \\fB\\-\\-check\\-sections\\fR.\n Section overlap is not usually checked for relocatable links. You can\n force checking in that case by using the \\fB\\-\\-check\\-sections\\fR\n option.\n .IP \\fB\\-\\-copy\\-dt\\-needed\\-entries\\fR 4\n .IX Item \"--copy-dt-needed-entries\"\n .PD 0\n .IP \\fB\\-\\-no\\-copy\\-dt\\-needed\\-entries\\fR 4\n .IX Item \"--no-copy-dt-needed-entries\"\n .PD\n This option affects the treatment of dynamic libraries referred to\n by DT_NEEDED tags \\fIinside\\fR ELF dynamic libraries mentioned on the\n-command line. Normally the linker won't add a DT_NEEDED tag to the\n+command line. Normally the linker won\\*(Aqt add a DT_NEEDED tag to the\n output binary for each library mentioned in a DT_NEEDED tag in an\n input dynamic library. With \\fB\\-\\-copy\\-dt\\-needed\\-entries\\fR\n specified on the command line however any dynamic libraries that\n follow it will have their DT_NEEDED entries added. The default\n behaviour can be restored with \\fB\\-\\-no\\-copy\\-dt\\-needed\\-entries\\fR.\n .Sp\n This option also has an effect on the resolution of symbols in dynamic\n@@ -1792,39 +1795,39 @@\n usually wasted: the types are usually small but the names are often not.\n \\&\\fB\\-\\-ctf\\-variables\\fR causes the generation of such a section.\n The default behaviour can be restored with \\fB\\-\\-no\\-ctf\\-variables\\fR.\n .IP \\fB\\-\\-ctf\\-share\\-types=\\fR\\fImethod\\fR 4\n .IX Item \"--ctf-share-types=method\"\n Adjust the method used to share types between translation units in CTF.\n .RS 4\n-.IP \\fBshare-unconflicted\\fR 4\n+.IP \\fBshare\\-unconflicted\\fR 4\n .IX Item \"share-unconflicted\"\n Put all types that do not have ambiguous definitions into the shared dictionary,\n where debuggers can easily access them, even if they only occur in one\n translation unit. This is the default.\n-.IP \\fBshare-duplicated\\fR 4\n+.IP \\fBshare\\-duplicated\\fR 4\n .IX Item \"share-duplicated\"\n Put only types that occur in multiple translation units into the shared\n-dictionary: types with only one definition go into per-translation-unit\n+dictionary: types with only one definition go into per\\-translation\\-unit\n dictionaries. Types with ambiguous definitions in multiple translation units\n-always go into per-translation-unit dictionaries. This tends to make the CTF\n+always go into per\\-translation\\-unit dictionaries. This tends to make the CTF\n larger, but may reduce the amount of CTF in the shared dictionary. For very\n large projects this may speed up opening the CTF and save memory in the CTF\n consumer at runtime.\n .RE\n .RS 4\n .RE\n .IP \\fB\\-\\-no\\-define\\-common\\fR 4\n .IX Item \"--no-define-common\"\n This option inhibits the assignment of addresses to common symbols.\n The script command \\f(CW\\*(C`INHIBIT_COMMON_ALLOCATION\\*(C'\\fR has the same effect.\n .Sp\n The \\fB\\-\\-no\\-define\\-common\\fR option allows decoupling\n the decision to assign addresses to Common symbols from the choice\n-of the output file type; otherwise a non-Relocatable output type\n+of the output file type; otherwise a non\\-Relocatable output type\n forces assigning addresses to Common symbols.\n Using \\fB\\-\\-no\\-define\\-common\\fR allows Common symbols that are referenced\n from a shared library to be assigned addresses only in the main program.\n This eliminates the unused duplicate space in the shared library,\n and also prevents any possible confusion over resolving to the wrong\n duplicate when there are many dynamic modules with specialized search\n paths for runtime symbol resolution.\n@@ -1876,26 +1879,26 @@\n .IX Item \"-Ifile\"\n .PD 0\n .IP \\fB\\-\\-dynamic\\-linker=\\fR\\fIfile\\fR 4\n .IX Item \"--dynamic-linker=file\"\n .PD\n Set the name of the dynamic linker. This is only meaningful when\n generating dynamically linked ELF executables. The default dynamic\n-linker is normally correct; don't use this unless you know what you are\n+linker is normally correct; don\\*(Aqt use this unless you know what you are\n doing.\n .IP \\fB\\-\\-no\\-dynamic\\-linker\\fR 4\n .IX Item \"--no-dynamic-linker\"\n When producing an executable file, omit the request for a dynamic\n-linker to be used at load-time. This is only meaningful for ELF\n+linker to be used at load\\-time. This is only meaningful for ELF\n executables that contain dynamic relocations, and usually requires\n entry point code that is capable of processing these relocations.\n .IP \\fB\\-\\-embedded\\-relocs\\fR 4\n .IX Item \"--embedded-relocs\"\n This option is similar to the \\fB\\-\\-emit\\-relocs\\fR option except\n-that the relocs are stored in a target-specific section. This option\n+that the relocs are stored in a target\\-specific section. This option\n is only supported by the \\fBBFIN\\fR, \\fBCR16\\fR and \\fIM68K\\fR\n targets.\n .IP \\fB\\-\\-disable\\-multiple\\-abs\\-defs\\fR 4\n .IX Item \"--disable-multiple-abs-defs\"\n Do not allow multiple definitions with symbols included\n in filename invoked by \\-R or \\-\\-just\\-symbols\n .IP \\fB\\-\\-fatal\\-warnings\\fR 4\n@@ -1920,15 +1923,15 @@\n .IX Item \"--force-exe-suffix\"\n Make sure that an output file has a .exe suffix.\n .Sp\n If a successfully built fully linked output file does not have a\n \\&\\f(CW\\*(C`.exe\\*(C'\\fR or \\f(CW\\*(C`.dll\\*(C'\\fR suffix, this option forces the linker to copy\n the output file to one of the same name with a \\f(CW\\*(C`.exe\\*(C'\\fR suffix. This\n option is useful when using unmodified Unix makefiles on a Microsoft\n-Windows host, since some versions of Windows won't run an image unless\n+Windows host, since some versions of Windows won\\*(Aqt run an image unless\n it ends in a \\f(CW\\*(C`.exe\\*(C'\\fR suffix.\n .IP \\fB\\-\\-gc\\-sections\\fR 4\n .IX Item \"--gc-sections\"\n .PD 0\n .IP \\fB\\-\\-no\\-gc\\-sections\\fR 4\n .IX Item \"--no-gc-sections\"\n .PD\n@@ -1938,15 +1941,15 @@\n \\&\\fB\\-\\-no\\-gc\\-sections\\fR on the command line. Note that garbage\n collection for COFF and PE format targets is supported, but the\n implementation is currently considered to be experimental.\n .Sp\n \\&\\fB\\-\\-gc\\-sections\\fR decides which input sections are used by\n examining symbols and relocations. The section containing the entry\n symbol and all sections containing symbols undefined on the\n-command-line will be kept, as will sections containing symbols\n+command\\-line will be kept, as will sections containing symbols\n referenced by dynamic objects. Note that when building shared\n libraries, the linker must assume that any visible symbol is\n referenced. Once this initial set of sections has been determined,\n the linker recursively marks as used any section referenced by their\n relocations. See \\fB\\-\\-entry\\fR, \\fB\\-\\-undefined\\fR, and\n \\&\\fB\\-\\-gc\\-keep\\-exported\\fR.\n .Sp\n@@ -1979,15 +1982,15 @@\n collected regardless of the external visibility of contained symbols.\n Note that this option has no effect when linking shared objects since\n it is already the default behaviour. This option is only supported for\n ELF format targets.\n .IP \\fB\\-\\-print\\-output\\-format\\fR 4\n .IX Item \"--print-output-format\"\n Print the name of the default output format (perhaps influenced by\n-other command-line options). This is the string that would appear\n+other command\\-line options). This is the string that would appear\n in an \\f(CW\\*(C`OUTPUT_FORMAT\\*(C'\\fR linker script command.\n .IP \\fB\\-\\-print\\-memory\\-usage\\fR 4\n .IX Item \"--print-memory-usage\"\n Print used size, total size and used size of memory regions created with\n the \\fBMEMORY\\fR command. This is useful on embedded targets to have a\n quick view of amount of free memory. The format of the output has one\n headline and one line per region. It is both human readable and easily\n@@ -1996,18 +1999,18 @@\n .Vb 3\n \\& Memory region Used Size Region Size %age Used\n \\& ROM: 256 KB 1 MB 25.00%\n \\& RAM: 32 B 2 GB 0.00%\n .Ve\n .IP \\fB\\-\\-help\\fR 4\n .IX Item \"--help\"\n-Print a summary of the command-line options on the standard output and exit.\n+Print a summary of the command\\-line options on the standard output and exit.\n .IP \\fB\\-\\-target\\-help\\fR 4\n .IX Item \"--target-help\"\n-Print a summary of all target-specific options on the standard output and exit.\n+Print a summary of all target\\-specific options on the standard output and exit.\n .IP \\fB\\-Map=\\fR\\fImapfile\\fR 4\n .IX Item \"-Map=mapfile\"\n Print a link map to the file \\fImapfile\\fR. See the description of the\n \\&\\fB\\-M\\fR option, above. If \\fImapfile\\fR is just the character\n \\&\\f(CW\\*(C`\\-\\*(C'\\fR then the map will be written to stdout.\n .Sp\n Specifying a directory as \\fImapfile\\fR causes the linker map to be\n@@ -2045,15 +2048,15 @@\n .IP \\fB\\-\\-no\\-undefined\\fR 4\n .IX Item \"--no-undefined\"\n .PD 0\n .IP \"\\fB\\-z defs\\fR\" 4\n .IX Item \"-z defs\"\n .PD\n Report unresolved symbol references from regular object files. This\n-is done even if the linker is creating a non-symbolic shared library.\n+is done even if the linker is creating a non\\-symbolic shared library.\n The switch \\fB\\-\\-[no\\-]allow\\-shlib\\-undefined\\fR controls the\n behaviour for reporting unresolved references found in shared\n libraries being linked in.\n .Sp\n The effects of this option can be reverted by using \\f(CW\\*(C`\\-z undefs\\*(C'\\fR.\n .IP \\fB\\-\\-allow\\-multiple\\-definition\\fR 4\n .IX Item \"--allow-multiple-definition\"\n@@ -2101,15 +2104,15 @@\n .RE\n .IP \\fB\\-\\-error\\-handling\\-script=\\fR\\fIscriptname\\fR 4\n .IX Item \"--error-handling-script=scriptname\"\n If this option is provided then the linker will invoke\n \\&\\fIscriptname\\fR whenever an error is encountered. Currently however\n only two kinds of error are supported: missing symbols and missing\n libraries. Two arguments will be passed to script: the keyword\n-\"undefined-symbol\" or `missing\\-lib\" and the \\fIname\\fR of the\n+\"undefined\\-symbol\" or \\`missing\\-lib\" and the \\fIname\\fR of the\n undefined symbol or missing library. The intention is that the script\n will provide suggestions to the user as to where the symbol or library\n might be found. After the script has finished then the normal linker\n error message will be displayed.\n .Sp\n The availability of this option is controlled by a configure time\n switch, so it may not be present in specific implementations.\n@@ -2150,23 +2153,23 @@\n errors during the link process; it exits without writing an output file\n when it issues any error whatsoever.\n .IP \\fB\\-nostdlib\\fR 4\n .IX Item \"-nostdlib\"\n Only search library directories explicitly specified on the\n command line. Library directories specified in linker scripts\n (including linker scripts specified on the command line) are ignored.\n-.IP \\fB\\-\\-oformat=\\fR\\fIoutput-format\\fR 4\n+.IP \\fB\\-\\-oformat=\\fR\\fIoutput\\-format\\fR 4\n .IX Item \"--oformat=output-format\"\n \\&\\fBld\\fR may be configured to support more than one kind of object\n file. If your \\fBld\\fR is configured this way, you can use the\n \\&\\fB\\-\\-oformat\\fR option to specify the binary format for the output\n object file. Even when \\fBld\\fR is configured to support alternative\n-object formats, you don't usually need to specify this, as \\fBld\\fR\n+object formats, you don\\*(Aqt usually need to specify this, as \\fBld\\fR\n should be configured to produce as a default output format the most\n-usual format on each machine. \\fIoutput-format\\fR is a text string, the\n+usual format on each machine. \\fIoutput\\-format\\fR is a text string, the\n name of a particular format supported by the BFD libraries. (You can\n list the available binary formats with \\fBobjdump \\-i\\fR.) The script\n command \\f(CW\\*(C`OUTPUT_FORMAT\\*(C'\\fR can also specify the output format, but\n this option overrides it.\n .IP \"\\fB\\-\\-out\\-implib\\fR \\fIfile\\fR\" 4\n .IX Item \"--out-implib file\"\n Create an import library in \\fIfile\\fR corresponding to the executable\n@@ -2226,15 +2229,15 @@\n .IP \\fB\\-\\-retain\\-symbols\\-file=\\fR\\fIfilename\\fR 4\n .IX Item \"--retain-symbols-file=filename\"\n Retain \\fIonly\\fR the symbols listed in the file \\fIfilename\\fR,\n discarding all others. \\fIfilename\\fR is simply a flat file, with one\n symbol name per line. This option is especially useful in environments\n (such as VxWorks)\n where a large global symbol table is accumulated gradually, to conserve\n-run-time memory.\n+run\\-time memory.\n .Sp\n \\&\\fB\\-\\-retain\\-symbols\\-file\\fR does \\fInot\\fR discard undefined symbols,\n or symbols needed for relocations.\n .Sp\n You may only specify \\fB\\-\\-retain\\-symbols\\-file\\fR once in the command\n line. It overrides \\fB\\-s\\fR and \\fB\\-S\\fR.\n .IP \\fB\\-rpath=\\fR\\fIdir\\fR 4\n@@ -2267,16 +2270,16 @@\n the \\fB\\-rpath\\fR option.\n .IP \\fB\\-rpath\\-link=\\fR\\fIdir\\fR 4\n .IX Item \"-rpath-link=dir\"\n When using ELF or SunOS, one shared library may require another. This\n happens when an \\f(CW\\*(C`ld \\-shared\\*(C'\\fR link includes a shared library as one\n of the input files.\n .Sp\n-When the linker encounters such a dependency when doing a non-shared,\n-non-relocatable link, it will automatically try to locate the required\n+When the linker encounters such a dependency when doing a non\\-shared,\n+non\\-relocatable link, it will automatically try to locate the required\n shared library and include it in the link, if it is not included\n explicitly. In such a case, several directories are searched as\n described below. The \\fB\\-rpath\\-link\\fR option specifies the first\n set of directories to search. This option may specify a sequence of\n directory names either by providing a list of names separated by\n colons, or by appearing multiple times.\n .Sp\n@@ -2424,16 +2427,16 @@\n .IP \\fB\\-\\-sort\\-common=ascending\\fR 4\n .IX Item \"--sort-common=ascending\"\n .IP \\fB\\-\\-sort\\-common=descending\\fR 4\n .IX Item \"--sort-common=descending\"\n .PD\n This option tells \\fBld\\fR to sort the common symbols by alignment in\n ascending or descending order when it places them in the appropriate output\n-sections. The symbol alignments considered are sixteen-byte or larger,\n-eight-byte, four-byte, two-byte, and one-byte. This is to prevent gaps\n+sections. The symbol alignments considered are sixteen\\-byte or larger,\n+eight\\-byte, four\\-byte, two\\-byte, and one\\-byte. This is to prevent gaps\n between symbols due to alignment constraints. If no sorting order is\n specified, then descending order is assumed.\n .IP \\fB\\-\\-sort\\-section=name\\fR 4\n .IX Item \"--sort-section=name\"\n This option will apply \\f(CW\\*(C`SORT_BY_NAME\\*(C'\\fR to all wildcard section\n patterns in the linker script.\n .IP \\fB\\-\\-sort\\-section=alignment\\fR 4\n@@ -2465,19 +2468,19 @@\n .IP \\fB\\-\\-stats\\fR 4\n .IX Item \"--stats\"\n Compute and display statistics about the operation of the linker, such\n as execution time and memory usage.\n .IP \\fB\\-\\-sysroot=\\fR\\fIdirectory\\fR 4\n .IX Item \"--sysroot=directory\"\n Use \\fIdirectory\\fR as the location of the sysroot, overriding the\n-configure-time default. This option is only supported by linkers\n+configure\\-time default. This option is only supported by linkers\n that were configured using \\fB\\-\\-with\\-sysroot\\fR.\n .IP \\fB\\-\\-task\\-link\\fR 4\n .IX Item \"--task-link\"\n-This is used by COFF/PE based targets to create a task-linked object\n+This is used by COFF/PE based targets to create a task\\-linked object\n file where all of the global symbols have been converted to statics.\n .IP \\fB\\-\\-traditional\\-format\\fR 4\n .IX Item \"--traditional-format\"\n For some targets, the output of \\fBld\\fR is different in some ways from\n the output of some existing linker. This switch requests \\fBld\\fR to\n use the traditional format instead.\n .Sp\n@@ -2525,41 +2528,41 @@\n byte of the first segment. Note that when \\fB\\-pie\\fR is used with\n \\&\\fB\\-Ttext\\-segment=\\fR\\fIorg\\fR, the output executable is marked\n ET_EXEC so that the address of the first byte of the text segment will\n be guaranteed to be \\fIorg\\fR at run time.\n .IP \\fB\\-Trodata\\-segment=\\fR\\fIorg\\fR 4\n .IX Item \"-Trodata-segment=org\"\n When creating an ELF executable or shared object for a target where\n-the read-only data is in its own segment separate from the executable\n-text, it will set the address of the first byte of the read-only data segment.\n+the read\\-only data is in its own segment separate from the executable\n+text, it will set the address of the first byte of the read\\-only data segment.\n .IP \\fB\\-Tldata\\-segment=\\fR\\fIorg\\fR 4\n .IX Item \"-Tldata-segment=org\"\n When creating an ELF executable or shared object for x86\\-64 medium memory\n model, it will set the address of the first byte of the ldata segment.\n .IP \\fB\\-\\-unresolved\\-symbols=\\fR\\fImethod\\fR 4\n .IX Item \"--unresolved-symbols=method\"\n Determine how to handle unresolved symbols. There are four possible\n values for \\fBmethod\\fR:\n .RS 4\n-.IP \\fBignore-all\\fR 4\n+.IP \\fBignore\\-all\\fR 4\n .IX Item \"ignore-all\"\n Do not report any unresolved symbols.\n-.IP \\fBreport-all\\fR 4\n+.IP \\fBreport\\-all\\fR 4\n .IX Item \"report-all\"\n Report all unresolved symbols. This is the default.\n-.IP \\fBignore-in-object-files\\fR 4\n+.IP \\fBignore\\-in\\-object\\-files\\fR 4\n .IX Item \"ignore-in-object-files\"\n Report unresolved symbols that are contained in shared libraries, but\n ignore them if they come from regular object files.\n-.IP \\fBignore-in-shared-libs\\fR 4\n+.IP \\fBignore\\-in\\-shared\\-libs\\fR 4\n .IX Item \"ignore-in-shared-libs\"\n Report unresolved symbols that come from regular object files, but\n ignore them if they come from shared libraries. This can be useful\n when creating a dynamic binary and it is known that all the shared\n-libraries that it should be referencing are included on the linker's\n+libraries that it should be referencing are included on the linker\\*(Aqs\n command line.\n .RE\n .RS 4\n .Sp\n The behaviour for shared libraries on their own can also be controlled\n by the \\fB\\-\\-[no\\-]allow\\-shlib\\-undefined\\fR option.\n .Sp\n@@ -2573,22 +2576,22 @@\n .IP \\fB\\-\\-verbose[=\\fR\\fINUMBER\\fR\\fB]\\fR 4\n .IX Item \"--verbose[=NUMBER]\"\n .PD\n Display the version number for \\fBld\\fR and list the linker emulations\n supported. Display which input files can and cannot be opened. Display\n the linker script being used by the linker. If the optional \\fINUMBER\\fR\n argument > 1, plugin symbol status will also be displayed.\n-.IP \\fB\\-\\-version\\-script=\\fR\\fIversion-scriptfile\\fR 4\n+.IP \\fB\\-\\-version\\-script=\\fR\\fIversion\\-scriptfile\\fR 4\n .IX Item \"--version-script=version-scriptfile\"\n Specify the name of a version script to the linker. This is typically\n used when creating shared libraries to specify additional information\n about the version hierarchy for the library being created. This option\n is only fully supported on ELF platforms which support shared libraries;\n see \\fBVERSION\\fR. It is partially supported on PE platforms, which can\n-use version scripts to filter symbol visibility in auto-export mode: any\n+use version scripts to filter symbol visibility in auto\\-export mode: any\n symbols marked \\fBlocal\\fR in the version script will not be exported.\n .IP \\fB\\-\\-warn\\-common\\fR 4\n .IX Item \"--warn-common\"\n Warn when a common symbol is combined with another common symbol or with\n a symbol definition. Unix linkers allow this somewhat sloppy practice,\n but linkers on some other operating systems do not. This option allows\n you to find potential problems from combining global symbols.\n@@ -2638,15 +2641,15 @@\n .Sp\n .Vb 3\n \\& (
): warning: definition of \\`\\*(Aq\n \\& overriding common\n \\& (
): warning: common is here\n .Ve\n .IP 3. 4\n-Merging a common symbol with a previous same-sized common symbol.\n+Merging a common symbol with a previous same\\-sized common symbol.\n .Sp\n .Vb 3\n \\& (
): warning: multiple common\n \\& of \\`\\*(Aq\n \\& (
): warning: previous common is here\n .Ve\n .IP 4. 4\n@@ -2696,18 +2699,18 @@\n Only generate a warning if an object file requests an executable\n stack, but not if the \\fB\\-z execstack\\fR option is used.\n .RE\n .RS 4\n .Sp\n The default state depends upon how the linker was configured when it\n was built. The \\fB\\-\\-no\\-warn\\-execstack\\fR option always puts the\n-linker into the no-warnings state. The \\fB\\-\\-warn\\-execstack\\fR\n-option puts the linker into the warn-always state. The\n+linker into the no\\-warnings state. The \\fB\\-\\-warn\\-execstack\\fR\n+option puts the linker into the warn\\-always state. The\n \\&\\fB\\-\\-warn\\-execstack\\-objects\\fR option puts the linker into the\n-warn-for-object-files-only state.\n+warn\\-for\\-object\\-files\\-only state.\n .Sp\n Note: ELF format input files can specify that they need an executable\n stack by having a \\fI.note.GNU\\-stack\\fR section with the executable\n bit set in its section flags. They can specify that they do not need\n an executable stack by having the same section, but without the\n executable flag bit set. If an input file does not have a\n \\&\\fI.note.GNU\\-stack\\fR section then the default behaviour is target\n@@ -2720,52 +2723,52 @@\n .PD 0\n .IP \\fB\\-\\-no\\-error\\-execstack\\fR 4\n .IX Item \"--no-error-execstack\"\n .PD\n If the linker is going to generate a warning message about an\n executable stack then the \\fB\\-\\-error\\-execstack\\fR option will\n instead change that warning into an error. Note \\- this option does\n-not change the linker's execstack warning generation state. Use\n+not change the linker\\*(Aqs execstack warning generation state. Use\n \\&\\fB\\-\\-warn\\-execstack\\fR or \\fB\\-\\-warn\\-execstack\\-objects\\fR to set\n a specific warning state.\n .Sp\n The \\fB\\-\\-no\\-error\\-execstack\\fR option will restore the default\n behaviour of generating warning messages.\n .IP \\fB\\-\\-warn\\-multiple\\-gp\\fR 4\n .IX Item \"--warn-multiple-gp\"\n Warn if multiple global pointer values are required in the output file.\n This is only meaningful for certain processors, such as the Alpha.\n-Specifically, some processors put large-valued constants in a special\n+Specifically, some processors put large\\-valued constants in a special\n section. A special register (the global pointer) points into the middle\n of this section, so that constants can be loaded efficiently via a\n-base-register relative addressing mode. Since the offset in\n-base-register relative mode is fixed and relatively small (e.g., 16\n+base\\-register relative addressing mode. Since the offset in\n+base\\-register relative mode is fixed and relatively small (e.g., 16\n bits), this limits the maximum size of the constant pool. Thus, in\n large programs, it is often necessary to use multiple global pointer\n values in order to be able to address all possible constants. This\n option causes a warning to be issued whenever this case occurs.\n .IP \\fB\\-\\-warn\\-once\\fR 4\n .IX Item \"--warn-once\"\n Only warn once for each undefined symbol, rather than once per module\n which refers to it.\n .IP \\fB\\-\\-warn\\-rwx\\-segments\\fR 4\n .IX Item \"--warn-rwx-segments\"\n .PD 0\n .IP \\fB\\-\\-no\\-warn\\-rwx\\-segments\\fR 4\n .IX Item \"--no-warn-rwx-segments\"\n .PD\n-Warn if the linker creates a loadable, non-zero sized segment that has\n+Warn if the linker creates a loadable, non\\-zero sized segment that has\n all three of the read, write and execute permission flags set. Such a\n segment represents a potential security vulnerability. In addition\n warnings will be generated if a thread local storage segment is\n created with the execute permission flag set, regardless of whether or\n not it has the read and/or write flags set.\n .Sp\n These warnings are enabled by default. They can be disabled via the\n-\\&\\fB\\-\\-no\\-warn\\-rwx\\-segments\\fR option and re-enabled via the\n+\\&\\fB\\-\\-no\\-warn\\-rwx\\-segments\\fR option and re\\-enabled via the\n \\&\\fB\\-\\-warn\\-rwx\\-segments\\fR option.\n .IP \\fB\\-\\-error\\-rwx\\-segments\\fR 4\n .IX Item \"--error-rwx-segments\"\n .PD 0\n .IP \\fB\\-\\-no\\-error\\-rwx\\-segments\\fR 4\n .IX Item \"--no-error-rwx-segments\"\n .PD\n@@ -2784,40 +2787,40 @@\n Warn if the address of an output section is changed because of\n alignment. Typically, the alignment will be set by an input section.\n The address will only be changed if it not explicitly specified; that\n is, if the \\f(CW\\*(C`SECTIONS\\*(C'\\fR command does not specify a start address for\n the section.\n .IP \\fB\\-\\-warn\\-textrel\\fR 4\n .IX Item \"--warn-textrel\"\n-Warn if the linker adds DT_TEXTREL to a position-independent executable\n+Warn if the linker adds DT_TEXTREL to a position\\-independent executable\n or shared object.\n .IP \\fB\\-\\-warn\\-alternate\\-em\\fR 4\n .IX Item \"--warn-alternate-em\"\n Warn if an object has alternate ELF machine code.\n .IP \\fB\\-\\-warn\\-unresolved\\-symbols\\fR 4\n .IX Item \"--warn-unresolved-symbols\"\n If the linker is going to report an unresolved symbol (see the option\n \\&\\fB\\-\\-unresolved\\-symbols\\fR) it will normally generate an error.\n This option makes it generate a warning instead.\n .IP \\fB\\-\\-error\\-unresolved\\-symbols\\fR 4\n .IX Item \"--error-unresolved-symbols\"\n-This restores the linker's default behaviour of generating errors when\n+This restores the linker\\*(Aqs default behaviour of generating errors when\n it is reporting unresolved symbols.\n .IP \\fB\\-\\-whole\\-archive\\fR 4\n .IX Item \"--whole-archive\"\n For each archive mentioned on the command line after the\n \\&\\fB\\-\\-whole\\-archive\\fR option, include every object file in the archive\n in the link, rather than searching the archive for the required object\n files. This is normally used to turn an archive file into a shared\n library, forcing every object to be included in the resulting shared\n library. This option may be used more than once.\n .Sp\n-Two notes when using this option from gcc: First, gcc doesn't know\n+Two notes when using this option from gcc: First, gcc doesn\\*(Aqt know\n about this option, so you have to use \\fB\\-Wl,\\-whole\\-archive\\fR.\n-Second, don't forget to use \\fB\\-Wl,\\-no\\-whole\\-archive\\fR after your\n+Second, don\\*(Aqt forget to use \\fB\\-Wl,\\-no\\-whole\\-archive\\fR after your\n list of archives, because gcc will add its own list of archives to\n your link and you may not want this flag to affect those as well.\n .IP \\fB\\-\\-wrap=\\fR\\fIsymbol\\fR 4\n .IX Item \"--wrap=symbol\"\n Use a wrapper function for \\fIsymbol\\fR. Any undefined reference to\n \\&\\fIsymbol\\fR will be resolved to \\f(CW\\*(C`_\\|_wrap_\\fR\\f(CIsymbol\\fR\\f(CW\\*(C'\\fR. Any\n undefined reference to \\f(CW\\*(C`_\\|_real_\\fR\\f(CIsymbol\\fR\\f(CW\\*(C'\\fR will be resolved to\n@@ -2895,22 +2898,22 @@\n \\&\\fB\\-\\-enable\\-new\\-dtags\\fR, the new dynamic tags will be created as needed\n and older dynamic tags will be omitted.\n If you specify \\fB\\-\\-disable\\-new\\-dtags\\fR, no new dynamic tags will be\n created. By default, the new dynamic tags are not created. Note that\n those options are only available for ELF systems.\n .IP \\fB\\-\\-hash\\-size=\\fR\\fInumber\\fR 4\n .IX Item \"--hash-size=number\"\n-Set the default size of the linker's hash tables to a prime number\n+Set the default size of the linker\\*(Aqs hash tables to a prime number\n close to \\fInumber\\fR. Increasing this value can reduce the length of\n time it takes the linker to perform its tasks, at the expense of\n-increasing the linker's memory requirements. Similarly reducing this\n+increasing the linker\\*(Aqs memory requirements. Similarly reducing this\n value can reduce the memory requirements at the expense of speed.\n .IP \\fB\\-\\-hash\\-style=\\fR\\fIstyle\\fR 4\n .IX Item \"--hash-style=style\"\n-Set the type of linker's hash table(s). \\fIstyle\\fR can be either\n+Set the type of linker\\*(Aqs hash table(s). \\fIstyle\\fR can be either\n \\&\\f(CW\\*(C`sysv\\*(C'\\fR for classic ELF \\f(CW\\*(C`.hash\\*(C'\\fR section, \\f(CW\\*(C`gnu\\*(C'\\fR for\n new style GNU \\f(CW\\*(C`.gnu.hash\\*(C'\\fR section or \\f(CW\\*(C`both\\*(C'\\fR for both\n the classic ELF \\f(CW\\*(C`.hash\\*(C'\\fR and new style GNU \\f(CW\\*(C`.gnu.hash\\*(C'\\fR\n hash tables. The default depends upon how the linker was configured,\n but for most Linux based systems it will be \\f(CW\\*(C`both\\*(C'\\fR.\n .IP \\fB\\-\\-compress\\-debug\\-sections=none\\fR 4\n .IX Item \"--compress-debug-sections=none\"\n@@ -2923,45 +2926,45 @@\n .IX Item \"--compress-debug-sections=zlib-gabi\"\n .IP \\fB\\-\\-compress\\-debug\\-sections=zstd\\fR 4\n .IX Item \"--compress-debug-sections=zstd\"\n .PD\n On ELF platforms, these options control how DWARF debug sections are\n compressed using zlib.\n .Sp\n-\\&\\fB\\-\\-compress\\-debug\\-sections=none\\fR doesn't compress DWARF debug\n+\\&\\fB\\-\\-compress\\-debug\\-sections=none\\fR doesn\\*(Aqt compress DWARF debug\n sections. \\fB\\-\\-compress\\-debug\\-sections=zlib\\-gnu\\fR compresses\n DWARF debug sections and renames them to begin with \\fB.zdebug\\fR\n instead of \\fB.debug\\fR. \\fB\\-\\-compress\\-debug\\-sections=zlib\\-gabi\\fR\n also compresses DWARF debug sections, but rather than renaming them it\n-sets the SHF_COMPRESSED flag in the sections' headers.\n+sets the SHF_COMPRESSED flag in the sections\\*(Aq headers.\n .Sp\n The \\fB\\-\\-compress\\-debug\\-sections=zlib\\fR option is an alias for\n \\&\\fB\\-\\-compress\\-debug\\-sections=zlib\\-gabi\\fR.\n .Sp\n \\&\\fB\\-\\-compress\\-debug\\-sections=zstd\\fR compresses DWARF debug sections using\n zstd.\n .Sp\n Note that this option overrides any compression in input debug\n sections, so if a binary is linked with \\fB\\-\\-compress\\-debug\\-sections=none\\fR\n for example, then any compressed debug sections in input files will be\n uncompressed before they are copied into the output binary.\n .Sp\n The default compression behaviour varies depending upon the target\n involved and the configure options used to build the toolchain. The\n-default can be determined by examining the output from the linker's\n+default can be determined by examining the output from the linker\\*(Aqs\n \\&\\fB\\-\\-help\\fR option.\n .IP \\fB\\-\\-reduce\\-memory\\-overheads\\fR 4\n .IX Item \"--reduce-memory-overheads\"\n This option reduces memory requirements at ld runtime, at the expense of\n linking speed. This was introduced to select the old O(n^2) algorithm\n for link map file generation, rather than the new O(n) algorithm which uses\n about 40% more memory for symbol storage.\n .Sp\n Another effect of the switch is to set the default hash table size to\n-1021, which again saves memory at the cost of lengthening the linker's\n+1021, which again saves memory at the cost of lengthening the linker\\*(Aqs\n run time. This is not done however if the \\fB\\-\\-hash\\-size\\fR switch\n has been used.\n .Sp\n The \\fB\\-\\-reduce\\-memory\\-overheads\\fR switch may be also be used to\n enable other tradeoffs in future versions of the linker.\n .IP \\fB\\-\\-max\\-cache\\-size=\\fR\\fIsize\\fR 4\n .IX Item \"--max-cache-size=size\"\n@@ -2984,28 +2987,28 @@\n specified as an even number of hexadecimal digits (\\f(CW\\*(C`\\-\\*(C'\\fR and\n \\&\\f(CW\\*(C`:\\*(C'\\fR characters between digit pairs are ignored). If \\fIstyle\\fR\n is omitted, \\f(CW\\*(C`sha1\\*(C'\\fR is used.\n .Sp\n The \\f(CW\\*(C`md5\\*(C'\\fR, \\f(CW\\*(C`sha1\\*(C'\\fR, and \\f(CW\\*(C`xx\\*(C'\\fR styles produces an\n identifier that is always the same in an identical output file, but\n are almost certainly unique among all nonidentical output files. It\n-is not intended to be compared as a checksum for the file's contents.\n+is not intended to be compared as a checksum for the file\\*(Aqs contents.\n A linked file may be changed later by other tools, but the build ID\n bit string identifying the original linked file does not change.\n .Sp\n Passing \\f(CW\\*(C`none\\*(C'\\fR for \\fIstyle\\fR disables the setting from any\n \\&\\f(CW\\*(C`\\-\\-build\\-id\\*(C'\\fR options earlier on the command line.\n .IP \\fB\\-\\-package\\-metadata=\\fR\\fIJSON\\fR 4\n .IX Item \"--package-metadata=JSON\"\n Request the creation of a \\f(CW\\*(C`.note.package\\*(C'\\fR ELF note section. The\n contents of the note are in JSON format, as per the package metadata\n specification. For more information see:\n https://systemd.io/ELF_PACKAGE_METADATA/\n-The JSON argument support percent-encoding and following %[string]\n-(where string refers to the name in HTML's Named Character References)\n+The JSON argument support percent\\-encoding and following %[string]\n+(where string refers to the name in HTML\\*(Aqs Named Character References)\n encoding: \\fB%[comma]\\fR for \\fB,\\fR, \\fB%[lbrace]\\fR for \\fB{\\fR,\n \\&\\fB%[quot]\\fR for \\fB\"\\fR, \\fB%[rbrace]\\fR for \\fB}\\fR, and\n \\&\\fB%[space]\\fR for space character.\n If the JSON argument is missing/empty then this will disable the\n creation of the metadata note, if one had been enabled by an earlier\n occurrence of the \\-\\-package\\-metadata option.\n If the linker has been built with libjansson, then the JSON string\n@@ -3017,21 +3020,21 @@\n use this option. In addition, the linker fully supports the standard\n \\&\\f(CW\\*(C`*.def\\*(C'\\fR files, which may be specified on the linker command line\n like an object file (in fact, it should precede archives it exports\n symbols from, to ensure that they get linked in, just like a normal\n object file).\n .PP\n In addition to the options common to all targets, the i386 PE linker\n-support additional command-line options that are specific to the i386\n+support additional command\\-line options that are specific to the i386\n PE target. Options that take values may be separated from their\n values by either a space or an equals sign.\n .IP \\fB\\-\\-add\\-stdcall\\-alias\\fR 4\n .IX Item \"--add-stdcall-alias\"\n If given, symbols with a stdcall suffix (@\\fInn\\fR) will be exported\n-as-is and also with the suffix stripped.\n+as\\-is and also with the suffix stripped.\n [This option is specific to the i386 PE targeted port of the linker]\n .IP \"\\fB\\-\\-base\\-file\\fR \\fIfile\\fR\" 4\n .IX Item \"--base-file file\"\n Use \\fIfile\\fR as the name of a file in which to save the base\n addresses of all the relocations needed for generating DLLs with\n \\&\\fIdlltool\\fR.\n [This is an i386 PE specific option]\n@@ -3046,24 +3049,24 @@\n .PD 0\n .IP \\fB\\-\\-disable\\-long\\-section\\-names\\fR 4\n .IX Item \"--disable-long-section-names\"\n .PD\n The PE variants of the COFF object format add an extension that permits\n the use of section names longer than eight characters, the normal limit\n for COFF. By default, these names are only allowed in object files, as\n-fully-linked executable images do not carry the COFF string table required\n+fully\\-linked executable images do not carry the COFF string table required\n to support the longer names. As a GNU extension, it is possible to\n allow their use in executable images as well, or to (probably pointlessly!)\n disallow it in object files, by using these two options. Executable images\n-generated with these long section names are slightly non-standard, carrying\n+generated with these long section names are slightly non\\-standard, carrying\n as they do a string table, and may generate confusing output when examined\n-with non-GNU PE-aware tools, such as file viewers and dumpers. However,\n+with non\\-GNU PE\\-aware tools, such as file viewers and dumpers. However,\n GDB relies on the use of PE long section names to find Dwarf\\-2 debug\n information sections in an executable image at runtime, and so if neither\n-option is specified on the command-line, \\fBld\\fR will enable long\n+option is specified on the command\\-line, \\fBld\\fR will enable long\n section names, overriding the default and technically correct behaviour,\n when it finds the presence of debug information while linking an executable\n image and not stripping symbols.\n [This option is valid for all PE targeted ports of the linker]\n .IP \\fB\\-\\-enable\\-stdcall\\-fixup\\fR 4\n .IX Item \"--enable-stdcall-fixup\"\n .PD 0\n@@ -3074,49 +3077,49 @@\n do \"fuzzy linking\" by looking for another defined symbol that differs\n only in the format of the symbol name (cdecl vs stdcall) and will\n resolve that symbol by linking to the match. For example, the\n undefined symbol \\f(CW\\*(C`_foo\\*(C'\\fR might be linked to the function\n \\&\\f(CW\\*(C`_foo@12\\*(C'\\fR, or the undefined symbol \\f(CW\\*(C`_bar@16\\*(C'\\fR might be linked\n to the function \\f(CW\\*(C`_bar\\*(C'\\fR. When the linker does this, it prints a\n warning, since it normally should have failed to link, but sometimes\n-import libraries generated from third-party dlls may need this feature\n+import libraries generated from third\\-party dlls may need this feature\n to be usable. If you specify \\fB\\-\\-enable\\-stdcall\\-fixup\\fR, this\n feature is fully enabled and warnings are not printed. If you specify\n \\&\\fB\\-\\-disable\\-stdcall\\-fixup\\fR, this feature is disabled and such\n mismatches are considered to be errors.\n [This option is specific to the i386 PE targeted port of the linker]\n .IP \\fB\\-\\-leading\\-underscore\\fR 4\n .IX Item \"--leading-underscore\"\n .PD 0\n .IP \\fB\\-\\-no\\-leading\\-underscore\\fR 4\n .IX Item \"--no-leading-underscore\"\n .PD\n-For most targets default symbol-prefix is an underscore and is defined\n-in target's description. By this option it is possible to\n-disable/enable the default underscore symbol-prefix.\n+For most targets default symbol\\-prefix is an underscore and is defined\n+in target\\*(Aqs description. By this option it is possible to\n+disable/enable the default underscore symbol\\-prefix.\n .IP \\fB\\-\\-export\\-all\\-symbols\\fR 4\n .IX Item \"--export-all-symbols\"\n If given, all global symbols in the objects used to build a DLL will\n be exported by the DLL. Note that this is the default if there\n-otherwise wouldn't be any exported symbols. When symbols are\n+otherwise wouldn\\*(Aqt be any exported symbols. When symbols are\n explicitly exported via DEF files or implicitly exported via function\n attributes, the default is to not export anything else unless this\n option is given. Note that the symbols \\f(CW\\*(C`DllMain@12\\*(C'\\fR,\n \\&\\f(CW\\*(C`DllEntryPoint@0\\*(C'\\fR, \\f(CW\\*(C`DllMainCRTStartup@12\\*(C'\\fR, and\n \\&\\f(CW\\*(C`impure_ptr\\*(C'\\fR will not be automatically\n exported. Also, symbols imported from other DLLs will not be\n-re-exported, nor will symbols specifying the DLL's internal layout\n+re\\-exported, nor will symbols specifying the DLL\\*(Aqs internal layout\n such as those beginning with \\f(CW\\*(C`_head_\\*(C'\\fR or ending with\n \\&\\f(CW\\*(C`_iname\\*(C'\\fR. In addition, no symbols from \\f(CW\\*(C`libgcc\\*(C'\\fR,\n \\&\\f(CW\\*(C`libstd++\\*(C'\\fR, \\f(CW\\*(C`libmingw32\\*(C'\\fR, or \\f(CW\\*(C`crtX.o\\*(C'\\fR will be exported.\n Symbols whose names begin with \\f(CW\\*(C`_\\|_rtti_\\*(C'\\fR or \\f(CW\\*(C`_\\|_builtin_\\*(C'\\fR will\n not be exported, to help with C++ DLLs. Finally, there is an\n-extensive list of cygwin-private symbols that are not exported\n+extensive list of cygwin\\-private symbols that are not exported\n (obviously, this applies on when building DLLs for cygwin targets).\n-These cygwin-excludes are: \\f(CW\\*(C`_cygwin_dll_entry@12\\*(C'\\fR,\n+These cygwin\\-excludes are: \\f(CW\\*(C`_cygwin_dll_entry@12\\*(C'\\fR,\n \\&\\f(CW\\*(C`_cygwin_crt0_common@8\\*(C'\\fR, \\f(CW\\*(C`_cygwin_noncygwin_dll_entry@12\\*(C'\\fR,\n \\&\\f(CW\\*(C`_fmode\\*(C'\\fR, \\f(CW\\*(C`_impure_ptr\\*(C'\\fR, \\f(CW\\*(C`cygwin_attach_dll\\*(C'\\fR,\n \\&\\f(CW\\*(C`cygwin_premain0\\*(C'\\fR, \\f(CW\\*(C`cygwin_premain1\\*(C'\\fR, \\f(CW\\*(C`cygwin_premain2\\*(C'\\fR,\n \\&\\f(CW\\*(C`cygwin_premain3\\*(C'\\fR, and \\f(CW\\*(C`environ\\*(C'\\fR.\n [This option is specific to the i386 PE targeted port of the linker]\n .IP \"\\fB\\-\\-exclude\\-symbols\\fR \\fIsymbol\\fR\\fB,\\fR\\fIsymbol\\fR\\fB,...\\fR\" 4\n .IX Item \"--exclude-symbols symbol,symbol,...\"\n@@ -3200,21 +3203,21 @@\n .PD 0\n .IP \\fB\\-\\-enable\\-auto\\-image\\-base=\\fR\\fIvalue\\fR 4\n .IX Item \"--enable-auto-image-base=value\"\n .PD\n Automatically choose the image base for DLLs, optionally starting with base\n \\&\\fIvalue\\fR, unless one is specified using the \\f(CW\\*(C`\\-\\-image\\-base\\*(C'\\fR argument.\n By using a hash generated from the dllname to create unique image bases\n-for each DLL, in-memory collisions and relocations which can delay program\n+for each DLL, in\\-memory collisions and relocations which can delay program\n execution are avoided.\n [This option is specific to the i386 PE targeted port of the linker]\n .IP \\fB\\-\\-disable\\-auto\\-image\\-base\\fR 4\n .IX Item \"--disable-auto-image-base\"\n Do not automatically generate a unique image base. If there is no\n-user-specified image base (\\f(CW\\*(C`\\-\\-image\\-base\\*(C'\\fR) then use the platform\n+user\\-specified image base (\\f(CW\\*(C`\\-\\-image\\-base\\*(C'\\fR) then use the platform\n default.\n [This option is specific to the i386 PE targeted port of the linker]\n .IP \"\\fB\\-\\-dll\\-search\\-prefix\\fR \\fIstring\\fR\" 4\n .IX Item \"--dll-search-prefix string\"\n When linking dynamically to a dll without an import library,\n search for \\f(CW\\*(C`.dll\\*(C'\\fR in preference to\n \\&\\f(CW\\*(C`lib.dll\\*(C'\\fR. This behaviour allows easy distinction\n@@ -3228,29 +3231,29 @@\n DATA imports from DLLs, thus making it possible to bypass the dllimport\n mechanism on the user side and to reference unmangled symbol names.\n [This option is specific to the i386 PE targeted port of the linker]\n .Sp\n The following remarks pertain to the original implementation of the\n feature and are obsolete nowadays for Cygwin and MinGW targets.\n .Sp\n-Note: Use of the 'auto\\-import' extension will cause the text section\n+Note: Use of the \\*(Aqauto\\-import\\*(Aq extension will cause the text section\n of the image file to be made writable. This does not conform to the\n-PE-COFF format specification published by Microsoft.\n+PE\\-COFF format specification published by Microsoft.\n .Sp\n-Note \\- use of the 'auto\\-import' extension will also cause read only\n+Note \\- use of the \\*(Aqauto\\-import\\*(Aq extension will also cause read only\n data which would normally be placed into the .rdata section to be\n placed into the .data section instead. This is in order to work\n around a problem with consts that is described here:\n http://www.cygwin.com/ml/cygwin/2004\\-09/msg01101.html\n .Sp\n-Using 'auto\\-import' generally will 'just work' \\-\\- but sometimes you may\n+Using \\*(Aqauto\\-import\\*(Aq generally will \\*(Aqjust work\\*(Aq \\-\\- but sometimes you may\n see this message:\n .Sp\n-\"variable '' can't be auto-imported. Please read the\n-documentation for ld's \\f(CW\\*(C`\\-\\-enable\\-auto\\-import\\*(C'\\fR for details.\"\n+\"variable \\*(Aq\\*(Aq can\\*(Aqt be auto\\-imported. Please read the\n+documentation for ld\\*(Aqs \\f(CW\\*(C`\\-\\-enable\\-auto\\-import\\*(C'\\fR for details.\"\n .Sp\n This message occurs when some (sub)expression accesses an address\n ultimately given by the sum of two constants (Win32 import tables only\n allow one). Instances where this may occur include accesses to member\n fields of struct variables imported from a DLL, as well as using a\n constant index into an array variable imported from a DLL. Any\n multiword variable (arrays, structs, long long, etc) may trigger\n@@ -3261,18 +3264,18 @@\n There are several ways to address this difficulty, regardless of the\n data type of the exported variable:\n .Sp\n One way is to use \\-\\-enable\\-runtime\\-pseudo\\-reloc switch. This leaves the task\n of adjusting references in your client code for runtime environment, so\n this method works only when runtime environment supports this feature.\n .Sp\n-A second solution is to force one of the 'constants' to be a variable \\-\\-\n-that is, unknown and un-optimizable at compile time. For arrays,\n-there are two possibilities: a) make the indexee (the array's address)\n-a variable, or b) make the 'constant' index a variable. Thus:\n+A second solution is to force one of the \\*(Aqconstants\\*(Aq to be a variable \\-\\-\n+that is, unknown and un\\-optimizable at compile time. For arrays,\n+there are two possibilities: a) make the indexee (the array\\*(Aqs address)\n+a variable, or b) make the \\*(Aqconstant\\*(Aq index a variable. Thus:\n .Sp\n .Vb 3\n \\& extern type extern_array[];\n \\& extern_array[1] \\-\\->\n \\& { volatile type *t=extern_array; t[1] }\n .Ve\n .Sp\n@@ -3298,21 +3301,21 @@\n .Vb 3\n \\& extern long long extern_ll;\n \\& extern_ll \\-\\->\n \\& { volatile long long * local_ll=&extern_ll; *local_ll }\n .Ve\n .Sp\n A third method of dealing with this difficulty is to abandon\n-\\&'auto\\-import' for the offending symbol and mark it with\n+\\&\\*(Aqauto\\-import\\*(Aq for the offending symbol and mark it with\n \\&\\f(CW\\*(C`_\\|_declspec(dllimport)\\*(C'\\fR. However, in practice that\n-requires using compile-time #defines to indicate whether you are\n+requires using compile\\-time #defines to indicate whether you are\n building a DLL, building client code that will link to the DLL, or\n merely building/linking to a static library. In making the choice\n-between the various methods of resolving the 'direct address with\n-constant offset' problem, you should consider typical real-world usage:\n+between the various methods of resolving the \\*(Aqdirect address with\n+constant offset\\*(Aq problem, you should consider typical real\\-world usage:\n .Sp\n Original:\n .Sp\n .Vb 7\n \\& \\-\\-foo.h\n \\& extern int arr[];\n \\& \\-\\-foo.c\n@@ -3351,37 +3354,37 @@\n \\& \\-\\-foo.c\n \\& #include \"foo.h\"\n \\& void main(int argc, char **argv){\n \\& printf(\"%d\\en\",arr[1]);\n \\& }\n .Ve\n .Sp\n-A fourth way to avoid this problem is to re-code your\n+A fourth way to avoid this problem is to re\\-code your\n library to use a functional interface rather than a data interface\n for the offending variables (e.g. \\fBset_foo()\\fR and \\fBget_foo()\\fR accessor\n functions).\n .IP \\fB\\-\\-disable\\-auto\\-import\\fR 4\n .IX Item \"--disable-auto-import\"\n Do not attempt to do sophisticated linking of \\f(CW\\*(C`_symbol\\*(C'\\fR to\n \\&\\f(CW\\*(C`_\\|_imp_\\|_symbol\\*(C'\\fR for DATA imports from DLLs.\n [This option is specific to the i386 PE targeted port of the linker]\n .IP \\fB\\-\\-enable\\-runtime\\-pseudo\\-reloc\\fR 4\n .IX Item \"--enable-runtime-pseudo-reloc\"\n If your code contains expressions described in \\-\\-enable\\-auto\\-import section,\n-that is, DATA imports from DLL with non-zero offset, this switch will create\n-a vector of 'runtime pseudo relocations' which can be used by runtime\n+that is, DATA imports from DLL with non\\-zero offset, this switch will create\n+a vector of \\*(Aqruntime pseudo relocations\\*(Aq which can be used by runtime\n environment to adjust references to such data in your client code.\n [This option is specific to the i386 PE targeted port of the linker]\n .IP \\fB\\-\\-disable\\-runtime\\-pseudo\\-reloc\\fR 4\n .IX Item \"--disable-runtime-pseudo-reloc\"\n-Do not create pseudo relocations for non-zero offset DATA imports from DLLs.\n+Do not create pseudo relocations for non\\-zero offset DATA imports from DLLs.\n [This option is specific to the i386 PE targeted port of the linker]\n .IP \\fB\\-\\-enable\\-extra\\-pe\\-debug\\fR 4\n .IX Item \"--enable-extra-pe-debug\"\n-Show additional debug info related to auto-import symbol thunking.\n+Show additional debug info related to auto\\-import symbol thunking.\n [This option is specific to the i386 PE targeted port of the linker]\n .IP \\fB\\-\\-section\\-alignment\\fR 4\n .IX Item \"--section-alignment\"\n Sets the section alignment. Sections in memory will always begin at\n addresses which are a multiple of this number. Defaults to 0x1000.\n [This option is specific to the i386 PE targeted port of the linker]\n .IP \"\\fB\\-\\-stack\\fR \\fIreserve\\fR\" 4\n@@ -3624,71 +3627,71 @@\n .IP \\fB\\-N\\fR 4\n .IX Item \"-N\"\n .PD 0\n .IP \\fB\\-\\-omagic\\fR 4\n .IX Item \"--omagic\"\n .PD\n Mark the output as \\f(CW\\*(C`OMAGIC\\*(C'\\fR (0407) in the \\fIa.out\\fR header to\n-indicate that the text segment is not to be write-protected and\n+indicate that the text segment is not to be write\\-protected and\n shared. Since the text and data sections are both readable and\n writable, the data section is allocated immediately contiguous after\n the text segment. This is the oldest format for PDP11 executable\n programs and is the default for \\fBld\\fR on PDP11 Unix systems\n from the beginning through 2.11BSD.\n .IP \\fB\\-n\\fR 4\n .IX Item \"-n\"\n .PD 0\n .IP \\fB\\-\\-nmagic\\fR 4\n .IX Item \"--nmagic\"\n .PD\n Mark the output as \\f(CW\\*(C`NMAGIC\\*(C'\\fR (0410) in the \\fIa.out\\fR header to\n indicate that when the output file is executed, the text portion will\n-be read-only and shareable among all processes executing the same\n+be read\\-only and shareable among all processes executing the same\n file. This involves moving the data areas up to the first possible 8K\n byte page boundary following the end of the text. This option creates\n a \\fIpure executable\\fR format.\n .IP \\fB\\-z\\fR 4\n .IX Item \"-z\"\n .PD 0\n .IP \\fB\\-\\-imagic\\fR 4\n .IX Item \"--imagic\"\n .PD\n Mark the output as \\f(CW\\*(C`IMAGIC\\*(C'\\fR (0411) in the \\fIa.out\\fR header to\n indicate that when the output file is executed, the program text and\n data areas will be loaded into separate address spaces using the split\n instruction and data space feature of the memory management unit in\n larger models of the PDP11. This doubles the address space available\n-to the program. The text segment is again pure, write-protected, and\n+to the program. The text segment is again pure, write\\-protected, and\n shareable. The only difference in the output format between this\n option and the others, besides the magic number, is that both the text\n and data sections start at location 0. The \\fB\\-z\\fR option selected\n this format in 2.11BSD. This option creates a \\fIseparate\n executable\\fR format.\n .IP \\fB\\-\\-no\\-omagic\\fR 4\n .IX Item \"--no-omagic\"\n Equivalent to \\fB\\-\\-nmagic\\fR for pdp11\\-aout.\n .SH ENVIRONMENT\n .IX Header \"ENVIRONMENT\"\n You can change the behaviour of \\fBld\\fR with the environment variables\n \\&\\f(CW\\*(C`GNUTARGET\\*(C'\\fR,\n \\&\\f(CW\\*(C`LDEMULATION\\*(C'\\fR and \\f(CW\\*(C`COLLECT_NO_DEMANGLE\\*(C'\\fR.\n .PP\n-\\&\\f(CW\\*(C`GNUTARGET\\*(C'\\fR determines the input-file object format if you don't\n+\\&\\f(CW\\*(C`GNUTARGET\\*(C'\\fR determines the input\\-file object format if you don\\*(Aqt\n use \\fB\\-b\\fR (or its synonym \\fB\\-\\-format\\fR). Its value should be one\n of the BFD names for an input format. If there is no\n \\&\\f(CW\\*(C`GNUTARGET\\*(C'\\fR in the environment, \\fBld\\fR uses the natural format\n of the target. If \\f(CW\\*(C`GNUTARGET\\*(C'\\fR is set to \\f(CW\\*(C`default\\*(C'\\fR then BFD\n attempts to discover the input format by examining binary input files;\n this method often succeeds, but there are potential ambiguities, since\n there is no method of ensuring that the magic number used to specify\n-object-file formats is unique. However, the configuration procedure for\n+object\\-file formats is unique. However, the configuration procedure for\n BFD on each system places the conventional format for that system first\n-in the search-list, so ambiguities are resolved in favor of convention.\n+in the search\\-list, so ambiguities are resolved in favor of convention.\n .PP\n-\\&\\f(CW\\*(C`LDEMULATION\\*(C'\\fR determines the default emulation if you don't use the\n+\\&\\f(CW\\*(C`LDEMULATION\\*(C'\\fR determines the default emulation if you don\\*(Aqt use the\n \\&\\fB\\-m\\fR option. The emulation can affect various aspects of linker\n behaviour, particularly the default linker script. You can list the\n available emulations with the \\fB\\-\\-verbose\\fR or \\fB\\-V\\fR options. If\n the \\fB\\-m\\fR option is not used, and the \\f(CW\\*(C`LDEMULATION\\*(C'\\fR environment\n variable is not defined, the default emulation depends upon how the\n linker was configured.\n .PP\n@@ -3711,10 +3714,10 @@\n .SH COPYRIGHT\n .IX Header \"COPYRIGHT\"\n Copyright (c) 1991\\-2025 Free Software Foundation, Inc.\n .PP\n Permission is granted to copy, distribute and/or modify this document\n under the terms of the GNU Free Documentation License, Version 1.3\n or any later version published by the Free Software Foundation;\n-with no Invariant Sections, with no Front-Cover Texts, and with no\n-Back-Cover Texts. A copy of the license is included in the\n+with no Invariant Sections, with no Front\\-Cover Texts, and with no\n+Back\\-Cover Texts. A copy of the license is included in the\n section entitled \"GNU Free Documentation License\".\n"}]}]}]}]}, {"source1": "binutils-mingw-w64-ucrt64_2.44-2+12_i386.deb", "source2": "binutils-mingw-w64-ucrt64_2.44-2+12_i386.deb", "unified_diff": null, "details": [{"source1": "file list", "source2": "file list", "unified_diff": "@@ -1,3 +1,3 @@\n -rw-r--r-- 0 0 0 4 2024-10-26 18:03:55.000000 debian-binary\n -rw-r--r-- 0 0 0 2052 2024-10-26 18:03:55.000000 control.tar.xz\n--rw-r--r-- 0 0 0 3621864 2024-10-26 18:03:55.000000 data.tar.xz\n+-rw-r--r-- 0 0 0 3621992 2024-10-26 18:03:55.000000 data.tar.xz\n"}, {"source1": "control.tar.xz", "source2": "control.tar.xz", "unified_diff": null, "details": [{"source1": "control.tar", "source2": "control.tar", "unified_diff": null, "details": [{"source1": "./md5sums", "source2": "./md5sums", "unified_diff": null, "details": [{"source1": "./md5sums", "source2": "./md5sums", "comments": ["Files differ"], "unified_diff": null}]}]}]}, {"source1": "data.tar.xz", "source2": "data.tar.xz", "unified_diff": null, "details": [{"source1": "data.tar", "source2": "data.tar", "unified_diff": null, "details": [{"source1": "file list", "source2": "file list", "unified_diff": "@@ -35,15 +35,15 @@\n -rw-r--r-- 0 root (0) root (0) 7222 2024-10-26 18:03:55.000000 ./usr/share/man/man1/x86_64-w64-mingw32ucrt-ar.1.gz\n -rw-r--r-- 0 root (0) root (0) 29664 2024-10-26 18:03:55.000000 ./usr/share/man/man1/x86_64-w64-mingw32ucrt-as.1.gz\n -rw-r--r-- 0 root (0) root (0) 3765 2024-10-26 18:03:55.000000 ./usr/share/man/man1/x86_64-w64-mingw32ucrt-c++filt.1.gz\n -rw-r--r-- 0 root (0) root (0) 5720 2024-10-26 18:03:55.000000 ./usr/share/man/man1/x86_64-w64-mingw32ucrt-dlltool.1.gz\n -rw-r--r-- 0 root (0) root (0) 311 2024-10-26 18:03:55.000000 ./usr/share/man/man1/x86_64-w64-mingw32ucrt-dllwrap.1.gz\n -rw-r--r-- 0 root (0) root (0) 2557 2024-10-26 18:03:55.000000 ./usr/share/man/man1/x86_64-w64-mingw32ucrt-elfedit.1.gz\n -rw-r--r-- 0 root (0) root (0) 7943 2024-10-26 18:03:55.000000 ./usr/share/man/man1/x86_64-w64-mingw32ucrt-gprof.1.gz\n--rw-r--r-- 0 root (0) root (0) 47205 2024-10-26 18:03:55.000000 ./usr/share/man/man1/x86_64-w64-mingw32ucrt-ld.1.gz\n+-rw-r--r-- 0 root (0) root (0) 47328 2024-10-26 18:03:55.000000 ./usr/share/man/man1/x86_64-w64-mingw32ucrt-ld.1.gz\n -rw-r--r-- 0 root (0) root (0) 7084 2024-10-26 18:03:55.000000 ./usr/share/man/man1/x86_64-w64-mingw32ucrt-nm.1.gz\n -rw-r--r-- 0 root (0) root (0) 14200 2024-10-26 18:03:55.000000 ./usr/share/man/man1/x86_64-w64-mingw32ucrt-objcopy.1.gz\n -rw-r--r-- 0 root (0) root (0) 14901 2024-10-26 18:03:55.000000 ./usr/share/man/man1/x86_64-w64-mingw32ucrt-objdump.1.gz\n -rw-r--r-- 0 root (0) root (0) 2144 2024-10-26 18:03:55.000000 ./usr/share/man/man1/x86_64-w64-mingw32ucrt-ranlib.1.gz\n -rw-r--r-- 0 root (0) root (0) 8507 2024-10-26 18:03:55.000000 ./usr/share/man/man1/x86_64-w64-mingw32ucrt-readelf.1.gz\n -rw-r--r-- 0 root (0) root (0) 3015 2024-10-26 18:03:55.000000 ./usr/share/man/man1/x86_64-w64-mingw32ucrt-size.1.gz\n -rw-r--r-- 0 root (0) root (0) 3806 2024-10-26 18:03:55.000000 ./usr/share/man/man1/x86_64-w64-mingw32ucrt-strings.1.gz\n"}, {"source1": "./usr/share/man/man1/x86_64-w64-mingw32ucrt-ld.1.gz", "source2": "./usr/share/man/man1/x86_64-w64-mingw32ucrt-ld.1.gz", "unified_diff": null, "details": [{"source1": "x86_64-w64-mingw32ucrt-ld.1", "source2": "x86_64-w64-mingw32ucrt-ld.1", "unified_diff": "@@ -1,9 +1,9 @@\n .\\\" -*- mode: troff; coding: utf-8 -*-\n-.\\\" Automatically generated by Pod::Man 5.0102 (Pod::Simple 3.45)\n+.\\\" Automatically generated by Pod::Man v6.0.2 (Pod::Simple 3.45)\n .\\\"\n .\\\" Standard preamble:\n .\\\" ========================================================================\n .de Sp \\\" Vertical space (when we can't use .PP)\n .if t .sp .5v\n .if n .sp\n ..\n@@ -48,18 +48,21 @@\n . if !\\nF==2 \\{\\\n . nr % 0\n . nr F 2\n . \\}\n . \\}\n .\\}\n .rr rF\n+.\\\"\n+.\\\" Required to disable full justification in groff 1.23.0.\n+.if n .ds AD l\n .\\\" ========================================================================\n .\\\"\n .IX Title \"LD 1\"\n-.TH LD 1 2024-10-26 binutils-2.44 \"GNU Development Tools\"\n+.TH LD 1 2025-02-02 binutils-2.44 \"GNU Development Tools\"\n .\\\" For nroff, turn off justification. Always turn off hyphenation; it makes\n .\\\" way too many mistakes in technical documents.\n .if n .ad l\n .nh\n .SH NAME\n ld \\- The GNU linker\n .SH SYNOPSIS\n@@ -68,15 +71,15 @@\n .SH DESCRIPTION\n .IX Header \"DESCRIPTION\"\n \\&\\fBld\\fR combines a number of object and archive files, relocates\n their data and ties up symbol references. Usually the last step in\n compiling a program is to run \\fBld\\fR.\n .PP\n \\&\\fBld\\fR accepts Linker Command Language files written in\n-a superset of AT&T's Link Editor Command Language syntax,\n+a superset of AT&T\\*(Aqs Link Editor Command Language syntax,\n to provide explicit and total control over the linking process.\n .PP\n This man page does not describe the command language; see the\n \\&\\fBld\\fR entry in \\f(CW\\*(C`info\\*(C'\\fR for full details on the command\n language and on other aspects of the GNU linker.\n .PP\n This version of \\fBld\\fR uses the general purpose BFD libraries\n@@ -92,41 +95,41 @@\n (or, in some cases, to get an output file in spite of the error).\n .PP\n The GNU linker \\fBld\\fR is meant to cover a broad range of situations,\n and to be as compatible as possible with other linkers. As a result,\n you have many choices to control its behavior.\n .SH OPTIONS\n .IX Header \"OPTIONS\"\n-The linker supports a plethora of command-line options, but in actual\n+The linker supports a plethora of command\\-line options, but in actual\n practice few of them are used in any particular context.\n For instance, a frequent use of \\fBld\\fR is to link standard Unix\n object files on a standard, supported Unix system. On such a system, to\n link a file \\f(CW\\*(C`hello.o\\*(C'\\fR:\n .PP\n .Vb 1\n \\& ld \\-o /lib/crt0.o hello.o \\-lc\n .Ve\n .PP\n This tells \\fBld\\fR to produce a file called \\fIoutput\\fR as the\n result of linking the file \\f(CW\\*(C`/lib/crt0.o\\*(C'\\fR with \\f(CW\\*(C`hello.o\\*(C'\\fR and\n the library \\f(CW\\*(C`libc.a\\*(C'\\fR, which will come from the standard search\n directories. (See the discussion of the \\fB\\-l\\fR option below.)\n .PP\n-Some of the command-line options to \\fBld\\fR may be specified at any\n+Some of the command\\-line options to \\fBld\\fR may be specified at any\n point in the command line. However, options which refer to files, such\n as \\fB\\-l\\fR or \\fB\\-T\\fR, cause the file to be read at the point at\n which the option appears in the command line, relative to the object\n-files and other file options. Repeating non-file options with a\n+files and other file options. Repeating non\\-file options with a\n different argument will either have no further effect, or override prior\n occurrences (those further to the left on the command line) of that\n option. Options which may be meaningfully specified more than once are\n noted in the descriptions below.\n .PP\n-Non-option arguments are object files or archives which are to be linked\n-together. They may follow, precede, or be mixed in with command-line\n+Non\\-option arguments are object files or archives which are to be linked\n+together. They may follow, precede, or be mixed in with command\\-line\n options, except that an object file argument may not be placed between\n an option and its argument.\n .PP\n Usually the linker is invoked with at least one object file, but you can\n specify other forms of binary input files using \\fB\\-l\\fR, \\fB\\-R\\fR,\n and the script command language. If \\fIno\\fR binary input files at all\n are specified, the linker does not produce any output, and issues the\n@@ -148,53 +151,53 @@\n option arguments must either follow the option letter without intervening\n whitespace, or be given as separate arguments immediately following the\n option that requires them.\n .PP\n For options whose names are multiple letters, either one dash or two can\n precede the option name; for example, \\fB\\-trace\\-symbol\\fR and\n \\&\\fB\\-\\-trace\\-symbol\\fR are equivalent. Note\\-\\-\\-there is one exception to\n-this rule. Multiple letter options that start with a lower case 'o' can\n+this rule. Multiple letter options that start with a lower case \\*(Aqo\\*(Aq can\n only be preceded by two dashes. This is to reduce confusion with the\n \\&\\fB\\-o\\fR option. So for example \\fB\\-omagic\\fR sets the output file\n name to \\fBmagic\\fR whereas \\fB\\-\\-omagic\\fR sets the NMAGIC flag on the\n output.\n .PP\n-Arguments to multiple-letter options must either be separated from the\n+Arguments to multiple\\-letter options must either be separated from the\n option name by an equals sign, or be given as separate arguments\n immediately following the option that requires them. For example,\n \\&\\fB\\-\\-trace\\-symbol foo\\fR and \\fB\\-\\-trace\\-symbol=foo\\fR are equivalent.\n-Unique abbreviations of the names of multiple-letter options are\n+Unique abbreviations of the names of multiple\\-letter options are\n accepted.\n .PP\n Note\\-\\-\\-if the linker is being invoked indirectly, via a compiler driver\n-(e.g. \\fBgcc\\fR) then all the linker command-line options should be\n+(e.g. \\fBgcc\\fR) then all the linker command\\-line options should be\n prefixed by \\fB\\-Wl,\\fR (or whatever is appropriate for the particular\n compiler driver) like this:\n .PP\n .Vb 1\n \\& gcc \\-Wl,\\-\\-start\\-group foo.o bar.o \\-Wl,\\-\\-end\\-group\n .Ve\n .PP\n This is important, because otherwise the compiler driver program may\n silently drop the linker options, resulting in a bad link. Confusion\n may also arise when passing options that require values through a\n driver, as the use of a space between option and argument acts as\n a separator, and causes the driver to pass only the option to the linker\n and the argument to the compiler. In this case, it is simplest to use\n-the joined forms of both single\\- and multiple-letter options, such as:\n+the joined forms of both single\\- and multiple\\-letter options, such as:\n .PP\n .Vb 1\n \\& gcc foo.o bar.o \\-Wl,\\-eENTRY \\-Wl,\\-Map=a.map\n .Ve\n .PP\n-Here is a table of the generic command-line switches accepted by the GNU\n+Here is a table of the generic command\\-line switches accepted by the GNU\n linker:\n .IP \\fB@\\fR\\fIfile\\fR 4\n .IX Item \"@file\"\n-Read command-line options from \\fIfile\\fR. The options read are\n+Read command\\-line options from \\fIfile\\fR. The options read are\n inserted in place of the original @\\fIfile\\fR option. If \\fIfile\\fR\n does not exist, or cannot be read, then the option will be treated\n literally, and not removed.\n .Sp\n Options in \\fIfile\\fR are separated by whitespace. A whitespace\n character may be included in an option by surrounding the entire\n option in either single or double quotes. Any character (including a\n@@ -212,57 +215,57 @@\n .IX Item \"--audit AUDITLIB\"\n Adds \\fIAUDITLIB\\fR to the \\f(CW\\*(C`DT_AUDIT\\*(C'\\fR entry of the dynamic section.\n \\&\\fIAUDITLIB\\fR is not checked for existence, nor will it use the DT_SONAME\n specified in the library. If specified multiple times \\f(CW\\*(C`DT_AUDIT\\*(C'\\fR\n will contain a colon separated list of audit interfaces to use. If the linker\n finds an object with an audit entry while searching for shared libraries,\n it will add a corresponding \\f(CW\\*(C`DT_DEPAUDIT\\*(C'\\fR entry in the output file.\n-This option is only meaningful on ELF platforms supporting the rtld-audit\n+This option is only meaningful on ELF platforms supporting the rtld\\-audit\n interface.\n-.IP \"\\fB\\-b\\fR \\fIinput-format\\fR\" 4\n+.IP \"\\fB\\-b\\fR \\fIinput\\-format\\fR\" 4\n .IX Item \"-b input-format\"\n .PD 0\n-.IP \\fB\\-\\-format=\\fR\\fIinput-format\\fR 4\n+.IP \\fB\\-\\-format=\\fR\\fIinput\\-format\\fR 4\n .IX Item \"--format=input-format\"\n .PD\n \\&\\fBld\\fR may be configured to support more than one kind of object\n file. If your \\fBld\\fR is configured this way, you can use the\n \\&\\fB\\-b\\fR option to specify the binary format for input object files\n that follow this option on the command line. Even when \\fBld\\fR is\n-configured to support alternative object formats, you don't usually need\n+configured to support alternative object formats, you don\\*(Aqt usually need\n to specify this, as \\fBld\\fR should be configured to expect as a\n default input format the most usual format on each machine.\n-\\&\\fIinput-format\\fR is a text string, the name of a particular format\n+\\&\\fIinput\\-format\\fR is a text string, the name of a particular format\n supported by the BFD libraries. (You can list the available binary\n formats with \\fBobjdump \\-i\\fR.)\n .Sp\n You may want to use this option if you are linking files with an unusual\n binary format. You can also use \\fB\\-b\\fR to switch formats explicitly (when\n linking object files of different formats), by including\n-\\&\\fB\\-b\\fR \\fIinput-format\\fR before each group of object files in a\n+\\&\\fB\\-b\\fR \\fIinput\\-format\\fR before each group of object files in a\n particular format.\n .Sp\n The default format is taken from the environment variable\n \\&\\f(CW\\*(C`GNUTARGET\\*(C'\\fR.\n .Sp\n You can also define the input format from a script, using the command\n \\&\\f(CW\\*(C`TARGET\\*(C'\\fR;\n-.IP \"\\fB\\-c\\fR \\fIMRI-commandfile\\fR\" 4\n+.IP \"\\fB\\-c\\fR \\fIMRI\\-commandfile\\fR\" 4\n .IX Item \"-c MRI-commandfile\"\n .PD 0\n-.IP \\fB\\-\\-mri\\-script=\\fR\\fIMRI-commandfile\\fR 4\n+.IP \\fB\\-\\-mri\\-script=\\fR\\fIMRI\\-commandfile\\fR 4\n .IX Item \"--mri-script=MRI-commandfile\"\n .PD\n For compatibility with linkers produced by MRI, \\fBld\\fR accepts script\n files written in an alternate, restricted command language, described in\n the MRI Compatible Script Files section of GNU ld documentation.\n Introduce MRI script files with\n the option \\fB\\-c\\fR; use the \\fB\\-T\\fR option to run linker\n-scripts written in the general-purpose \\fBld\\fR scripting language.\n-If \\fIMRI-cmdfile\\fR does not exist, \\fBld\\fR looks for it in the directories\n+scripts written in the general\\-purpose \\fBld\\fR scripting language.\n+If \\fIMRI\\-cmdfile\\fR does not exist, \\fBld\\fR looks for it in the directories\n specified by any \\fB\\-L\\fR options.\n .IP \\fB\\-d\\fR 4\n .IX Item \"-d\"\n .PD 0\n .IP \\fB\\-dc\\fR 4\n .IX Item \"-dc\"\n .IP \\fB\\-dp\\fR 4\n@@ -278,15 +281,15 @@\n .IP \"\\fB\\-P\\fR \\fIAUDITLIB\\fR\" 4\n .IX Item \"-P AUDITLIB\"\n .PD\n Adds \\fIAUDITLIB\\fR to the \\f(CW\\*(C`DT_DEPAUDIT\\*(C'\\fR entry of the dynamic section.\n \\&\\fIAUDITLIB\\fR is not checked for existence, nor will it use the DT_SONAME\n specified in the library. If specified multiple times \\f(CW\\*(C`DT_DEPAUDIT\\*(C'\\fR\n will contain a colon separated list of audit interfaces to use. This\n-option is only meaningful on ELF platforms supporting the rtld-audit interface.\n+option is only meaningful on ELF platforms supporting the rtld\\-audit interface.\n The \\-P option is provided for Solaris compatibility.\n .IP \\fB\\-\\-enable\\-linker\\-version\\fR 4\n .IX Item \"--enable-linker-version\"\n Enables the \\f(CW\\*(C`LINKER_VERSION\\*(C'\\fR linker script directive, described\n in \\fBOutput Section Data\\fR. If this directive is used in a linker\n script and this option has been enabled then a string containing the\n linker version will be inserted at the current point.\n@@ -300,15 +303,15 @@\n does not insert a version string. This is the default.\n .IP \\fB\\-\\-enable\\-non\\-contiguous\\-regions\\fR 4\n .IX Item \"--enable-non-contiguous-regions\"\n This option avoids generating an error if an input section does not\n fit a matching output section. The linker tries to allocate the input\n section to subseque nt matching output sections, and generates an\n error only if no output section is large enough. This is useful when\n-several non-contiguous memory regions are available and the input\n+several non\\-contiguous memory regions are available and the input\n section does not require a particular one. The order in which input\n sections are evaluated does not change, for instance:\n .Sp\n .Vb 10\n \\& MEMORY {\n \\& MEM1 (rwx) : ORIGIN = 0x1000, LENGTH = 0x14\n \\& MEM2 (rwx) : ORIGIN = 0x1000, LENGTH = 0x40\n@@ -364,16 +367,16 @@\n .IX Item \"--exclude-modules-for-implib module,module,...\"\n Specifies a list of object files or archive members, from which symbols\n should not be automatically exported, but which should be copied wholesale\n into the import library being generated during the link. The module names\n may be delimited by commas or colons, and must match exactly the filenames\n used by \\fBld\\fR to open the files; for archive members, this is simply\n the member name, but for object files the name listed must include and\n-match precisely any path used to specify the input file on the linker's\n-command-line. This option is available only for the i386 PE targeted port\n+match precisely any path used to specify the input file on the linker\\*(Aqs\n+command\\-line. This option is available only for the i386 PE targeted port\n of the linker. Symbols explicitly listed in a .def file are still exported,\n regardless of this option.\n .IP \\fB\\-E\\fR 4\n .IX Item \"-E\"\n .PD 0\n .IP \\fB\\-\\-export\\-dynamic\\fR 4\n .IX Item \"--export-dynamic\"\n@@ -404,28 +407,28 @@\n the description of \\fB\\-\\-export\\-all\\-symbols\\fR below.\n .IP \\fB\\-\\-export\\-dynamic\\-symbol=\\fR\\fIglob\\fR 4\n .IX Item \"--export-dynamic-symbol=glob\"\n When creating a dynamically linked executable, symbols matching\n \\&\\fIglob\\fR will be added to the dynamic symbol table. When creating a\n shared library, references to symbols matching \\fIglob\\fR will not be\n bound to the definitions within the shared library. This option is a\n-no-op when creating a shared library and \\fB\\-Bsymbolic\\fR or\n+no\\-op when creating a shared library and \\fB\\-Bsymbolic\\fR or\n \\&\\fB\\-\\-dynamic\\-list\\fR are not specified. This option is only meaningful\n on ELF platforms which support shared libraries.\n .IP \\fB\\-\\-export\\-dynamic\\-symbol\\-list=\\fR\\fIfile\\fR 4\n .IX Item \"--export-dynamic-symbol-list=file\"\n Specify a \\fB\\-\\-export\\-dynamic\\-symbol\\fR for each pattern in the file.\n The format of the file is the same as the version node without\n scope and node name. See \\fBVERSION\\fR for more information.\n .IP \\fB\\-EB\\fR 4\n .IX Item \"-EB\"\n-Link big-endian objects. This affects the default output format.\n+Link big\\-endian objects. This affects the default output format.\n .IP \\fB\\-EL\\fR 4\n .IX Item \"-EL\"\n-Link little-endian objects. This affects the default output format.\n+Link little\\-endian objects. This affects the default output format.\n .IP \"\\fB\\-f\\fR \\fIname\\fR\" 4\n .IX Item \"-f name\"\n .PD 0\n .IP \\fB\\-\\-auxiliary=\\fR\\fIname\\fR 4\n .IX Item \"--auxiliary=name\"\n .PD\n When creating an ELF shared object, set the internal DT_AUXILIARY field\n@@ -437,15 +440,15 @@\n run the program, the dynamic linker will see the DT_AUXILIARY field. If\n the dynamic linker resolves any symbols from the filter object, it will\n first check whether there is a definition in the shared object\n \\&\\fIname\\fR. If there is one, it will be used instead of the definition\n in the filter object. The shared object \\fIname\\fR need not exist.\n Thus the shared object \\fIname\\fR may be used to provide an alternative\n implementation of certain functions, perhaps for debugging or for\n-machine-specific performance.\n+machine\\-specific performance.\n .Sp\n This option may be specified more than once. The DT_AUXILIARY entries\n will be created in the order in which they appear on the command line.\n .IP \"\\fB\\-F\\fR \\fIname\\fR\" 4\n .IX Item \"-F name\"\n .PD 0\n .IP \\fB\\-\\-filter=\\fR\\fIname\\fR 4\n@@ -461,15 +464,15 @@\n dynamic linker will resolve symbols according to the symbol table of the\n filter object as usual, but it will actually link to the definitions\n found in the shared object \\fIname\\fR. Thus the filter object can be\n used to select a subset of the symbols provided by the object\n \\&\\fIname\\fR.\n .Sp\n Some older linkers used the \\fB\\-F\\fR option throughout a compilation\n-toolchain for specifying object-file format for both input and output\n+toolchain for specifying object\\-file format for both input and output\n object files.\n The GNU linker uses other mechanisms for this purpose: the\n \\&\\fB\\-b\\fR, \\fB\\-\\-format\\fR, \\fB\\-\\-oformat\\fR options, the\n \\&\\f(CW\\*(C`TARGET\\*(C'\\fR command in linker scripts, and the \\f(CW\\*(C`GNUTARGET\\*(C'\\fR\n environment variable.\n The GNU linker will ignore the \\fB\\-F\\fR option when not\n creating an ELF shared object.\n@@ -698,15 +701,15 @@\n .Sp\n .Vb 1\n \\& Removed property 0xc0000002 to merge foo.o (0x1) and bar.o (not found)\n .Ve\n .Sp\n This indicates that property 0xc0000002 is removed from output when\n merging properties in \\fIfoo.o\\fR, whose property 0xc0000002 value\n-is 0x1, and \\fIbar.o\\fR, which doesn't have property 0xc0000002.\n+is 0x1, and \\fIbar.o\\fR, which doesn\\*(Aqt have property 0xc0000002.\n .Sp\n .Vb 1\n \\& Updated property 0xc0010001 (0x1) to merge foo.o (0x1) and bar.o (0x1)\n .Ve\n .Sp\n This indicates that property 0xc0010001 value is updated to 0x1 in output\n when merging properties in \\fIfoo.o\\fR, whose 0xc0010001 property value\n@@ -754,24 +757,24 @@\n .IP \\fB\\-N\\fR 4\n .IX Item \"-N\"\n .PD 0\n .IP \\fB\\-\\-omagic\\fR 4\n .IX Item \"--omagic\"\n .PD\n Set the text and data sections to be readable and writable. Also, do\n-not page-align the data segment, and disable linking against shared\n+not page\\-align the data segment, and disable linking against shared\n libraries. If the output format supports Unix style magic numbers,\n mark the output as \\f(CW\\*(C`OMAGIC\\*(C'\\fR. Note: Although a writable text section\n-is allowed for PE-COFF targets, it does not conform to the format\n+is allowed for PE\\-COFF targets, it does not conform to the format\n specification published by Microsoft.\n .IP \\fB\\-\\-no\\-omagic\\fR 4\n .IX Item \"--no-omagic\"\n This option negates most of the effects of the \\fB\\-N\\fR option. It\n-sets the text section to be read-only, and forces the data segment to\n-be page-aligned. Note \\- this option does not enable linking against\n+sets the text section to be read\\-only, and forces the data segment to\n+be page\\-aligned. Note \\- this option does not enable linking against\n shared libraries. Use \\fB\\-Bdynamic\\fR for this.\n .IP \"\\fB\\-o\\fR \\fIoutput\\fR\" 4\n .IX Item \"-o output\"\n .PD 0\n .IP \\fB\\-\\-output=\\fR\\fIoutput\\fR 4\n .IX Item \"--output=output\"\n .PD\n@@ -800,29 +803,29 @@\n files (the first one in the \\fBOBJS\\fR list) whereas the second\n version of the linker command will generate an error message and not\n delete anything.\n .IP \\fB\\-\\-dependency\\-file=\\fR\\fIdepfile\\fR 4\n .IX Item \"--dependency-file=depfile\"\n Write a \\fIdependency file\\fR to \\fIdepfile\\fR. This file contains a rule\n suitable for \\f(CW\\*(C`make\\*(C'\\fR describing the output file and all the input files\n-that were read to produce it. The output is similar to the compiler's\n-output with \\fB\\-M \\-MP\\fR. Note that there is no option like the compiler's \\fB\\-MM\\fR,\n-to exclude \"system files\" (which is not a well-specified concept in the\n+that were read to produce it. The output is similar to the compiler\\*(Aqs\n+output with \\fB\\-M \\-MP\\fR. Note that there is no option like the compiler\\*(Aqs \\fB\\-MM\\fR,\n+to exclude \"system files\" (which is not a well\\-specified concept in the\n linker, unlike \"system headers\" in the compiler). So the output from\n \\&\\fB\\-\\-dependency\\-file\\fR is always specific to the exact state of the\n installation where it was produced, and should not be copied into\n distributed makefiles without careful editing.\n .IP \"\\fB\\-O\\fR \\fIlevel\\fR\" 4\n .IX Item \"-O level\"\n If \\fIlevel\\fR is a numeric values greater than zero \\fBld\\fR optimizes\n the output. This might take significantly longer and therefore probably\n should only be enabled for the final binary. At the moment this\n option only affects ELF shared library generation. Future releases of\n the linker may make more use of this option. Also currently there is\n-no difference in the linker's behaviour for different non-zero values\n+no difference in the linker\\*(Aqs behaviour for different non\\-zero values\n of this option. Again this may change with future releases.\n .IP \"\\fB\\-plugin\\fR \\fIname\\fR\" 4\n .IX Item \"-plugin name\"\n Involve a plugin in the linking process. The \\fIname\\fR parameter is\n the absolute filename of the plugin. Usually this parameter is\n automatically added by the compiler, when using link time\n optimization, but users can also add their own plugins if they so\n@@ -844,15 +847,15 @@\n The option which are covered are: \\fB\\-Bdynamic\\fR, \\fB\\-Bstatic\\fR,\n \\&\\fB\\-dn\\fR, \\fB\\-dy\\fR, \\fB\\-call_shared\\fR, \\fB\\-non_shared\\fR,\n \\&\\fB\\-static\\fR, \\fB\\-N\\fR, \\fB\\-n\\fR, \\fB\\-\\-whole\\-archive\\fR,\n \\&\\fB\\-\\-no\\-whole\\-archive\\fR, \\fB\\-r\\fR, \\fB\\-Ur\\fR,\n \\&\\fB\\-\\-copy\\-dt\\-needed\\-entries\\fR, \\fB\\-\\-no\\-copy\\-dt\\-needed\\-entries\\fR,\n \\&\\fB\\-\\-as\\-needed\\fR, \\fB\\-\\-no\\-as\\-needed\\fR, and \\fB\\-a\\fR.\n .Sp\n-One target for this option are specifications for \\fIpkg-config\\fR. When\n+One target for this option are specifications for \\fIpkg\\-config\\fR. When\n used with the \\fB\\-\\-libs\\fR option all possibly needed libraries are\n listed and then possibly linked with all the time. It is better to return\n something as follows:\n .Sp\n .Vb 1\n \\& \\-Wl,\\-\\-push\\-state,\\-\\-as\\-needed \\-libone \\-libtwo \\-Wl,\\-\\-pop\\-state\n .Ve\n@@ -881,31 +884,31 @@\n .PD 0\n .IP \\fB\\-\\-relocatable\\fR 4\n .IX Item \"--relocatable\"\n .PD\n Generate relocatable output\\-\\-\\-i.e., generate an output file that can in\n turn serve as input to \\fBld\\fR. This is often called \\fIpartial\n linking\\fR. As a side effect, in environments that support standard Unix\n-magic numbers, this option also sets the output file's magic number to\n+magic numbers, this option also sets the output file\\*(Aqs magic number to\n \\&\\f(CW\\*(C`OMAGIC\\*(C'\\fR.\n If this option is not specified, an absolute file is produced. When\n linking C++ programs, this option \\fIwill not\\fR resolve references to\n constructors; to do that, use \\fB\\-Ur\\fR.\n .Sp\n When an input file does not have the same format as the output file,\n partial linking is only supported if that input file does not contain any\n relocations. Different output formats can have further restrictions; for\n example some \\f(CW\\*(C`a.out\\*(C'\\fR\\-based formats do not support partial linking\n with input files in other formats at all.\n .Sp\n When the relocatable output contains both contents which require\n-link-time optimization (LTO) and contents which don't require LTO,\n+link\\-time optimization (LTO) and contents which don\\*(Aqt require LTO,\n a .gnu_object_only section will be created to contain a relocatable\n object file, as if \\fB\\-r\\fR is applied to all relocatable inputs\n-which don't require LTO. When processing a relocatable input with\n+which don\\*(Aqt require LTO. When processing a relocatable input with\n a .gnu_object_only section, the linker will extract the .gnu_object_only\n section as a separate input.\n .Sp\n Note that since \\fB\\-r\\fR groups some sections from different input files\n together, there may be negative impacts on code size and locality in\n final executable or shared library.\n .Sp\n@@ -926,20 +929,20 @@\n the \\fB\\-rpath\\fR option.\n .IP \\fB\\-\\-rosegment\\fR 4\n .IX Item \"--rosegment\"\n .PD 0\n .IP \\fB\\-\\-no\\-rosegment\\fR 4\n .IX Item \"--no-rosegment\"\n .PD\n-Attempt to ensure that only a single read-only, non-code segment is\n+Attempt to ensure that only a single read\\-only, non\\-code segment is\n created. Only useful when used in conjunction with the \\fB\\-z\n-separate-code\\fR option. The resulting binaries should be smaller than\n-if \\fB\\-z separate-code\\fR is used on its own. Without this option,\n-or if \\fB\\-\\-no\\-rosegment\\fR is specified, the \\fB\\-z separate-code\\fR\n-option will create two read-only segments, one before the code segment\n+separate\\-code\\fR option. The resulting binaries should be smaller than\n+if \\fB\\-z separate\\-code\\fR is used on its own. Without this option,\n+or if \\fB\\-\\-no\\-rosegment\\fR is specified, the \\fB\\-z separate\\-code\\fR\n+option will create two read\\-only segments, one before the code segment\n and one after it.\n .Sp\n The name of the options are misleading, but they have been chosen in\n order for the linker to be compatible with the LLD and GOLD linkers.\n .Sp\n Thse options are only supported by ELF targets.\n .IP \\fB\\-s\\fR 4\n@@ -981,28 +984,28 @@\n .IP \"\\fB\\-T\\fR \\fIscriptfile\\fR\" 4\n .IX Item \"-T scriptfile\"\n .PD 0\n .IP \\fB\\-\\-script=\\fR\\fIscriptfile\\fR 4\n .IX Item \"--script=scriptfile\"\n .PD\n Use \\fIscriptfile\\fR as the linker script. This script replaces\n-\\&\\fBld\\fR's default linker script (rather than adding to it),\n+\\&\\fBld\\fR\\*(Aqs default linker script (rather than adding to it),\n unless the script contains \\f(CW\\*(C`INSERT\\*(C'\\fR, so \\fIcommandfile\\fR must\n specify everything necessary to describe the output file.\n .Sp\n If \\fIscriptfile\\fR does not exist in the current directory, \\f(CW\\*(C`ld\\*(C'\\fR\n looks for it in the directories specified by any preceding \\fB\\-L\\fR\n options.\n .Sp\n Command line options that appear before the \\fB\\-T\\fR option can\n affect the script, but command line options that appear after it do\n not.\n .Sp\n Multiple \\fB\\-T\\fR options will accumulate if they are augmenting the\n-current script, otherwise the last, non-augmenting, \\fB\\-T\\fR option\n+current script, otherwise the last, non\\-augmenting, \\fB\\-T\\fR option\n will be used.\n .Sp\n There are other ways of specifying linker scripts. See\n .IP \"\\fB\\-dT\\fR \\fIscriptfile\\fR\" 4\n .IX Item \"-dT scriptfile\"\n .PD 0\n .IP \\fB\\-\\-default\\-script=\\fR\\fIscriptfile\\fR 4\n@@ -1119,28 +1122,28 @@\n .IP \\fB\\-X\\fR 4\n .IX Item \"-X\"\n .PD 0\n .IP \\fB\\-\\-discard\\-locals\\fR 4\n .IX Item \"--discard-locals\"\n .PD\n Delete all temporary local symbols. (These symbols start with\n-system-specific local label prefixes, typically \\fB.L\\fR for ELF systems\n+system\\-specific local label prefixes, typically \\fB.L\\fR for ELF systems\n or \\fBL\\fR for traditional a.out systems.)\n .IP \"\\fB\\-y\\fR \\fIsymbol\\fR\" 4\n .IX Item \"-y symbol\"\n .PD 0\n .IP \\fB\\-\\-trace\\-symbol=\\fR\\fIsymbol\\fR 4\n .IX Item \"--trace-symbol=symbol\"\n .PD\n Print the name of each linked file in which \\fIsymbol\\fR appears. This\n option may be given any number of times. On many systems it is necessary\n to prepend an underscore.\n .Sp\n This option is useful when you have an undefined symbol in your link but\n-don't know where the reference is coming from.\n+don\\*(Aqt know where the reference is coming from.\n .IP \"\\fB\\-Y\\fR \\fIpath\\fR\" 4\n .IX Item \"-Y path\"\n Add \\fIpath\\fR to the default library search path. This option exists\n for Solaris compatibility.\n .IP \"\\fB\\-z\\fR \\fIkeyword\\fR\" 4\n .IX Item \"-z keyword\"\n The recognized keywords are:\n@@ -1201,26 +1204,26 @@\n .IX Item \"common-page-size=value\"\n Set the page size most commonly used to \\fIvalue\\fR. Memory image\n layout will be optimized to minimize memory pages if the system is\n using pages of this size.\n .IP \\fBdefs\\fR 4\n .IX Item \"defs\"\n Report unresolved symbol references from regular object files. This\n-is done even if the linker is creating a non-symbolic shared library.\n+is done even if the linker is creating a non\\-symbolic shared library.\n This option is the inverse of \\fB\\-z undefs\\fR.\n-.IP \\fBdynamic-undefined-weak\\fR 4\n+.IP \\fBdynamic\\-undefined\\-weak\\fR 4\n .IX Item \"dynamic-undefined-weak\"\n .PD 0\n-.IP \\fBnodynamic-undefined-weak\\fR 4\n+.IP \\fBnodynamic\\-undefined\\-weak\\fR 4\n .IX Item \"nodynamic-undefined-weak\"\n .PD\n Make undefined weak symbols dynamic when building a dynamic object,\n if they are referenced from a regular object file and not forced local\n by symbol visibility or versioning. Do not make them dynamic if\n-\\&\\fBnodynamic-undefined-weak\\fR. If neither option is given, a target\n+\\&\\fBnodynamic\\-undefined\\-weak\\fR. If neither option is given, a target\n may default to either option being in force, or make some other\n selection of undefined weak symbols dynamic. Not all targets support\n these options.\n .IP \\fBexecstack\\fR 4\n .IX Item \"execstack\"\n Marks the object as requiring executable stack.\n .IP \\fBglobal\\fR 4\n@@ -1230,38 +1233,38 @@\n of subsequently loaded libraries.\n .IP \\fBglobalaudit\\fR 4\n .IX Item \"globalaudit\"\n This option is only meaningful when building a dynamic executable.\n This option marks the executable as requiring global auditing by\n setting the \\f(CW\\*(C`DF_1_GLOBAUDIT\\*(C'\\fR bit in the \\f(CW\\*(C`DT_FLAGS_1\\*(C'\\fR dynamic\n tag. Global auditing requires that any auditing library defined via\n-the \\fB\\-\\-depaudit\\fR or \\fB\\-P\\fR command-line options be run for\n+the \\fB\\-\\-depaudit\\fR or \\fB\\-P\\fR command\\-line options be run for\n all dynamic objects loaded by the application.\n .IP \\fBibtplt\\fR 4\n .IX Item \"ibtplt\"\n Generate Intel Indirect Branch Tracking (IBT) enabled PLT entries.\n Supported for Linux/i386 and Linux/x86_64.\n .IP \\fBibt\\fR 4\n .IX Item \"ibt\"\n Generate GNU_PROPERTY_X86_FEATURE_1_IBT in .note.gnu.property section\n to indicate compatibility with IBT. This also implies \\fBibtplt\\fR.\n Supported for Linux/i386 and Linux/x86_64.\n-.IP \\fBindirect-extern-access\\fR 4\n+.IP \\fBindirect\\-extern\\-access\\fR 4\n .IX Item \"indirect-extern-access\"\n .PD 0\n-.IP \\fBnoindirect-extern-access\\fR 4\n+.IP \\fBnoindirect\\-extern\\-access\\fR 4\n .IX Item \"noindirect-extern-access\"\n .PD\n Generate GNU_PROPERTY_1_NEEDED_INDIRECT_EXTERN_ACCESS in\n \\&.note.gnu.property section to indicate that object file requires\n canonical function pointers and cannot be used with copy relocation.\n-This option also implies \\fBnoextern-protected-data\\fR and\n+This option also implies \\fBnoextern\\-protected\\-data\\fR and\n \\&\\fBnocopyreloc\\fR. Supported for i386 and x86\\-64.\n .Sp\n-\\&\\fBnoindirect-extern-access\\fR removes\n+\\&\\fBnoindirect\\-extern\\-access\\fR removes\n GNU_PROPERTY_1_NEEDED_INDIRECT_EXTERN_ACCESS from .note.gnu.property\n section.\n .IP \\fBinitfirst\\fR 4\n .IX Item \"initfirst\"\n This option is only meaningful when building a shared object.\n It marks the object so that its runtime initialization will occur\n before the runtime initialization of any other objects brought into\n@@ -1346,72 +1349,72 @@\n .IX Item \"lazy\"\n When generating an executable or shared library, mark it to tell the\n dynamic linker to defer function call resolution to the point when\n the function is called (lazy binding), rather than at load time.\n Lazy binding is the default.\n .IP \\fBloadfltr\\fR 4\n .IX Item \"loadfltr\"\n-Specify that the object's filters be processed immediately at runtime.\n+Specify that the object\\*(Aqs filters be processed immediately at runtime.\n .IP \\fBmax\\-page\\-size=\\fR\\fIvalue\\fR 4\n .IX Item \"max-page-size=value\"\n Set the maximum memory page size supported to \\fIvalue\\fR.\n-.IP \\fBmark-plt\\fR 4\n+.IP \\fBmark\\-plt\\fR 4\n .IX Item \"mark-plt\"\n .PD 0\n-.IP \\fBnomark-plt\\fR 4\n+.IP \\fBnomark\\-plt\\fR 4\n .IX Item \"nomark-plt\"\n .PD\n Mark PLT entries with dynamic tags, DT_X86_64_PLT, DT_X86_64_PLTSZ and\n-DT_X86_64_PLTENT. Since this option stores a non-zero value in the\n+DT_X86_64_PLTENT. Since this option stores a non\\-zero value in the\n r_addend field of R_X86_64_JUMP_SLOT relocations, the resulting\n executables and shared libraries are incompatible with dynamic linkers,\n such as those in older versions of glibc without the change to ignore\n r_addend in R_X86_64_GLOB_DAT and R_X86_64_JUMP_SLOT relocations, which\n-don't ignore the r_addend field of R_X86_64_JUMP_SLOT relocations.\n+don\\*(Aqt ignore the r_addend field of R_X86_64_JUMP_SLOT relocations.\n Supported for x86_64.\n .IP \\fBmuldefs\\fR 4\n .IX Item \"muldefs\"\n Allow multiple definitions.\n .IP \\fBnocopyreloc\\fR 4\n .IX Item \"nocopyreloc\"\n Disable linker generated .dynbss variables used in place of variables\n defined in shared libraries. May result in dynamic text relocations.\n .IP \\fBnodefaultlib\\fR 4\n .IX Item \"nodefaultlib\"\n Specify that the dynamic loader search for dependencies of this object\n should ignore any default library search paths.\n .IP \\fBnodelete\\fR 4\n .IX Item \"nodelete\"\n-Specify that the object shouldn't be unloaded at runtime.\n+Specify that the object shouldn\\*(Aqt be unloaded at runtime.\n .IP \\fBnodlopen\\fR 4\n .IX Item \"nodlopen\"\n Specify that the object is not available to \\f(CW\\*(C`dlopen\\*(C'\\fR.\n .IP \\fBnodump\\fR 4\n .IX Item \"nodump\"\n Specify that the object can not be dumped by \\f(CW\\*(C`dldump\\*(C'\\fR.\n .IP \\fBnoexecstack\\fR 4\n .IX Item \"noexecstack\"\n Marks the object as not requiring executable stack.\n-.IP \\fBnoextern-protected-data\\fR 4\n+.IP \\fBnoextern\\-protected\\-data\\fR 4\n .IX Item \"noextern-protected-data\"\n-Don't treat protected data symbols as external when building a shared\n+Don\\*(Aqt treat protected data symbols as external when building a shared\n library. This option overrides the linker backend default. It can be\n used to work around incorrect relocations against protected data symbols\n generated by compiler. Updates on protected data symbols by another\n-module aren't visible to the resulting shared library. Supported for\n+module aren\\*(Aqt visible to the resulting shared library. Supported for\n i386 and x86\\-64.\n-.IP \\fBnoreloc-overflow\\fR 4\n+.IP \\fBnoreloc\\-overflow\\fR 4\n .IX Item \"noreloc-overflow\"\n Disable relocation overflow check. This can be used to disable\n relocation overflow check if there will be no dynamic relocation\n-overflow at run-time. Supported for x86_64.\n-.IP \\fBmemory-seal\\fR 4\n+overflow at run\\-time. Supported for x86_64.\n+.IP \\fBmemory\\-seal\\fR 4\n .IX Item \"memory-seal\"\n .PD 0\n-.IP \\fBnomemory-seal\\fR 4\n+.IP \\fBnomemory\\-seal\\fR 4\n .IX Item \"nomemory-seal\"\n .PD\n Instruct the executable or shared library that the all PT_LOAD segments\n should be sealed to avoid further manipulation (such as changing the\n protection flags, the segment size, or remove the mapping).\n This is a security hardening that requires system support. This\n generates GNU_PROPERTY_MEMORY_SEAL in .note.gnu.property section\n@@ -1421,125 +1424,125 @@\n dynamic linker to resolve all symbols when the program is started, or\n when the shared library is loaded by dlopen, instead of deferring\n function call resolution to the point when the function is first\n called.\n .IP \\fBorigin\\fR 4\n .IX Item \"origin\"\n Specify that the object requires \\fR\\f(CB$ORIGIN\\fR\\fB\\fR handling in paths.\n-.IP \\fBpack-relative-relocs\\fR 4\n+.IP \\fBpack\\-relative\\-relocs\\fR 4\n .IX Item \"pack-relative-relocs\"\n .PD 0\n-.IP \\fBnopack-relative-relocs\\fR 4\n+.IP \\fBnopack\\-relative\\-relocs\\fR 4\n .IX Item \"nopack-relative-relocs\"\n .PD\n-Generate compact relative relocation in position-independent executable\n+Generate compact relative relocation in position\\-independent executable\n and shared library. It adds \\f(CW\\*(C`DT_RELR\\*(C'\\fR, \\f(CW\\*(C`DT_RELRSZ\\*(C'\\fR and\n \\&\\f(CW\\*(C`DT_RELRENT\\*(C'\\fR entries to the dynamic section. It is ignored when\n-building position-dependent executable and relocatable output.\n-\\&\\fBnopack-relative-relocs\\fR is the default, which disables compact\n+building position\\-dependent executable and relocatable output.\n+\\&\\fBnopack\\-relative\\-relocs\\fR is the default, which disables compact\n relative relocation. When linked against the GNU C Library, a\n GLIBC_ABI_DT_RELR symbol version dependency on the shared C Library is\n added to the output. Supported for i386 and x86\\-64.\n .IP \\fBrelro\\fR 4\n .IX Item \"relro\"\n .PD 0\n .IP \\fBnorelro\\fR 4\n .IX Item \"norelro\"\n .PD\n Create an ELF \\f(CW\\*(C`PT_GNU_RELRO\\*(C'\\fR segment header in the object. This\n-specifies a memory segment that should be made read-only after\n-relocation, if supported. Specifying \\fBcommon-page-size\\fR smaller\n+specifies a memory segment that should be made read\\-only after\n+relocation, if supported. Specifying \\fBcommon\\-page\\-size\\fR smaller\n than the system page size will render this protection ineffective.\n-Don't create an ELF \\f(CW\\*(C`PT_GNU_RELRO\\*(C'\\fR segment if \\fBnorelro\\fR.\n-.IP \\fBreport-relative-reloc\\fR 4\n+Don\\*(Aqt create an ELF \\f(CW\\*(C`PT_GNU_RELRO\\*(C'\\fR segment if \\fBnorelro\\fR.\n+.IP \\fBreport\\-relative\\-reloc\\fR 4\n .IX Item \"report-relative-reloc\"\n Report dynamic relative relocations generated by linker. Supported for\n Linux/i386 and Linux/x86_64.\n .IP \\fBsectionheader\\fR 4\n .IX Item \"sectionheader\"\n .PD 0\n .IP \\fBnosectionheader\\fR 4\n .IX Item \"nosectionheader\"\n .PD\n-Generate section header. Don't generate section header if\n+Generate section header. Don\\*(Aqt generate section header if\n \\&\\fBnosectionheader\\fR is used. \\fBsectionheader\\fR is the default.\n-.IP \\fBseparate-code\\fR 4\n+.IP \\fBseparate\\-code\\fR 4\n .IX Item \"separate-code\"\n .PD 0\n-.IP \\fBnoseparate-code\\fR 4\n+.IP \\fBnoseparate\\-code\\fR 4\n .IX Item \"noseparate-code\"\n .PD\n Create separate code \\f(CW\\*(C`PT_LOAD\\*(C'\\fR segment header in the object. This\n specifies a memory segment that should contain only instructions and must\n-be in wholly disjoint pages from any other data. Don't create separate\n-code \\f(CW\\*(C`PT_LOAD\\*(C'\\fR segment if \\fBnoseparate-code\\fR is used.\n+be in wholly disjoint pages from any other data. Don\\*(Aqt create separate\n+code \\f(CW\\*(C`PT_LOAD\\*(C'\\fR segment if \\fBnoseparate\\-code\\fR is used.\n .IP \\fBshstk\\fR 4\n .IX Item \"shstk\"\n Generate GNU_PROPERTY_X86_FEATURE_1_SHSTK in .note.gnu.property section\n to indicate compatibility with Intel Shadow Stack. Supported for\n Linux/i386 and Linux/x86_64.\n .IP \\fBstack\\-size=\\fR\\fIvalue\\fR 4\n .IX Item \"stack-size=value\"\n Specify a stack size for an ELF \\f(CW\\*(C`PT_GNU_STACK\\*(C'\\fR segment.\n-Specifying zero will override any default non-zero sized\n+Specifying zero will override any default non\\-zero sized\n \\&\\f(CW\\*(C`PT_GNU_STACK\\*(C'\\fR segment creation.\n-.IP \\fBstart-stop-gc\\fR 4\n+.IP \\fBstart\\-stop\\-gc\\fR 4\n .IX Item \"start-stop-gc\"\n .PD 0\n-.IP \\fBnostart-stop-gc\\fR 4\n+.IP \\fBnostart\\-stop\\-gc\\fR 4\n .IX Item \"nostart-stop-gc\"\n .PD\n When \\fB\\-\\-gc\\-sections\\fR is in effect, a reference from a retained\n section to \\f(CW\\*(C`_\\|_start_SECNAME\\*(C'\\fR or \\f(CW\\*(C`_\\|_stop_SECNAME\\*(C'\\fR causes all\n input sections named \\f(CW\\*(C`SECNAME\\*(C'\\fR to also be retained, if\n \\&\\f(CW\\*(C`SECNAME\\*(C'\\fR is representable as a C identifier and either\n \\&\\f(CW\\*(C`_\\|_start_SECNAME\\*(C'\\fR or \\f(CW\\*(C`_\\|_stop_SECNAME\\*(C'\\fR is synthesized by the\n-linker. \\fB\\-z start-stop-gc\\fR disables this effect, allowing\n+linker. \\fB\\-z start\\-stop\\-gc\\fR disables this effect, allowing\n sections to be garbage collected as if the special synthesized symbols\n-were not defined. \\fB\\-z start-stop-gc\\fR has no effect on a\n+were not defined. \\fB\\-z start\\-stop\\-gc\\fR has no effect on a\n definition of \\f(CW\\*(C`_\\|_start_SECNAME\\*(C'\\fR or \\f(CW\\*(C`_\\|_stop_SECNAME\\*(C'\\fR in an\n object file or linker script. Such a definition will prevent the\n linker providing a synthesized \\f(CW\\*(C`_\\|_start_SECNAME\\*(C'\\fR or\n \\&\\f(CW\\*(C`_\\|_stop_SECNAME\\*(C'\\fR respectively, and therefore the special\n treatment by garbage collection for those references.\n .IP \\fBstart\\-stop\\-visibility=\\fR\\fIvalue\\fR 4\n .IX Item \"start-stop-visibility=value\"\n Specify the ELF symbol visibility for synthesized\n \\&\\f(CW\\*(C`_\\|_start_SECNAME\\*(C'\\fR and \\f(CW\\*(C`_\\|_stop_SECNAME\\*(C'\\fR symbols. \\fIvalue\\fR must be exactly \\fBdefault\\fR,\n \\&\\fBinternal\\fR, \\fBhidden\\fR, or \\fBprotected\\fR. If no \\fB\\-z\n-start-stop-visibility\\fR option is given, \\fBprotected\\fR is used for\n-compatibility with historical practice. However, it's highly\n+start\\-stop\\-visibility\\fR option is given, \\fBprotected\\fR is used for\n+compatibility with historical practice. However, it\\*(Aqs highly\n recommended to use \\fB\\-z start\\-stop\\-visibility=hidden\\fR in new\n programs and shared libraries so that these symbols are not exported\n-between shared objects, which is not usually what's intended.\n+between shared objects, which is not usually what\\*(Aqs intended.\n .IP \\fBtext\\fR 4\n .IX Item \"text\"\n .PD 0\n .IP \\fBnotext\\fR 4\n .IX Item \"notext\"\n .IP \\fBtextoff\\fR 4\n .IX Item \"textoff\"\n .PD\n-Report an error if DT_TEXTREL is set, i.e., if the position-independent\n-or shared object has dynamic relocations in read-only sections. Don't\n+Report an error if DT_TEXTREL is set, i.e., if the position\\-independent\n+or shared object has dynamic relocations in read\\-only sections. Don\\*(Aqt\n report an error if \\fBnotext\\fR or \\fBtextoff\\fR.\n .IP \\fBundefs\\fR 4\n .IX Item \"undefs\"\n Do not report unresolved symbol references from regular object files,\n either when creating an executable, or when creating a shared library.\n This option is the inverse of \\fB\\-z defs\\fR.\n-.IP \\fBunique-symbol\\fR 4\n+.IP \\fBunique\\-symbol\\fR 4\n .IX Item \"unique-symbol\"\n .PD 0\n-.IP \\fBnounique-symbol\\fR 4\n+.IP \\fBnounique\\-symbol\\fR 4\n .IX Item \"nounique-symbol\"\n .PD\n Avoid duplicated local symbol names in the symbol string table. Append\n-\".\\f(CW\\*(C`number\\*(C'\\fR\" to duplicated local symbol names if \\fBunique-symbol\\fR\n-is used. \\fBnounique-symbol\\fR is the default.\n+\".\\f(CW\\*(C`number\\*(C'\\fR\" to duplicated local symbol names if \\fBunique\\-symbol\\fR\n+is used. \\fBnounique\\-symbol\\fR is the default.\n .IP \\fBx86\\-64\\-baseline\\fR 4\n .IX Item \"x86-64-baseline\"\n .PD 0\n .IP \\fBx86\\-64\\-v2\\fR 4\n .IX Item \"x86-64-v2\"\n .IP \\fBx86\\-64\\-v3\\fR 4\n .IX Item \"x86-64-v3\"\n@@ -1619,17 +1622,17 @@\n .PD\n This option affects ELF DT_NEEDED tags for dynamic libraries mentioned\n on the command line after the \\fB\\-\\-as\\-needed\\fR option. Normally\n the linker will add a DT_NEEDED tag for each dynamic library mentioned\n on the command line, regardless of whether the library is actually\n needed or not. \\fB\\-\\-as\\-needed\\fR causes a DT_NEEDED tag to only be\n emitted for a library that \\fIat that point in the link\\fR satisfies a\n-non-weak undefined symbol reference from a regular object file or, if\n+non\\-weak undefined symbol reference from a regular object file or, if\n the library is not found in the DT_NEEDED lists of other needed libraries, a\n-non-weak undefined symbol reference from another needed dynamic library.\n+non\\-weak undefined symbol reference from another needed dynamic library.\n Object files or libraries appearing on the command line \\fIafter\\fR\n the library in question do not affect whether the library is seen as\n needed. This is similar to the rules for extraction of object files\n from archives. \\fB\\-\\-no\\-as\\-needed\\fR restores the default behaviour.\n .Sp\n Note: On Linux based systems the \\fB\\-\\-as\\-needed\\fR option also has\n an affect on the behaviour of the \\fB\\-\\-rpath\\fR and\n@@ -1682,15 +1685,15 @@\n Do not link against shared libraries. This is only meaningful on\n platforms for which shared libraries are supported. The different\n variants of this option are for compatibility with various systems. You\n may use this option multiple times on the command line: it affects\n library searching for \\fB\\-l\\fR options which follow it. This\n option also implies \\fB\\-\\-unresolved\\-symbols=report\\-all\\fR. This\n option can be used with \\fB\\-shared\\fR. Doing so means that a\n-shared library is being created but that all of the library's external\n+shared library is being created but that all of the library\\*(Aqs external\n references must be resolved by pulling in entries from static\n libraries.\n .IP \\fB\\-Bsymbolic\\fR 4\n .IX Item \"-Bsymbolic\"\n When creating a shared library, bind references to global symbols to the\n definition within the shared library, if any. Normally, it is possible\n for a program linked against a shared library to override the definition\n@@ -1702,19 +1705,19 @@\n symbols to the definition within the shared library, if any.\n This option is only meaningful on ELF platforms which support shared\n libraries.\n .IP \\fB\\-Bno\\-symbolic\\fR 4\n .IX Item \"-Bno-symbolic\"\n This option can cancel previously specified \\fB\\-Bsymbolic\\fR and\n \\&\\fB\\-Bsymbolic\\-functions\\fR.\n-.IP \\fB\\-\\-dynamic\\-list=\\fR\\fIdynamic-list-file\\fR 4\n+.IP \\fB\\-\\-dynamic\\-list=\\fR\\fIdynamic\\-list\\-file\\fR 4\n .IX Item \"--dynamic-list=dynamic-list-file\"\n Specify the name of a dynamic list file to the linker. This is\n typically used when creating shared libraries to specify a list of\n-global symbols whose references shouldn't be bound to the definition\n+global symbols whose references shouldn\\*(Aqt be bound to the definition\n within the shared library, or creating dynamically linked executables\n to specify a list of symbols which should be added to the symbol table\n in the executable. This option is only meaningful on ELF platforms\n which support shared libraries.\n .Sp\n The format of the dynamic list is the same as the version node without\n scope and node name. See \\fBVERSION\\fR for more information.\n@@ -1735,27 +1738,27 @@\n .IX Item \"--no-check-sections\"\n .PD\n Asks the linker \\fInot\\fR to check section addresses after they have\n been assigned to see if there are any overlaps. Normally the linker will\n perform this check, and if it finds any overlaps it will produce\n suitable error messages. The linker does know about, and does make\n allowances for sections in overlays. The default behaviour can be\n-restored by using the command-line switch \\fB\\-\\-check\\-sections\\fR.\n+restored by using the command\\-line switch \\fB\\-\\-check\\-sections\\fR.\n Section overlap is not usually checked for relocatable links. You can\n force checking in that case by using the \\fB\\-\\-check\\-sections\\fR\n option.\n .IP \\fB\\-\\-copy\\-dt\\-needed\\-entries\\fR 4\n .IX Item \"--copy-dt-needed-entries\"\n .PD 0\n .IP \\fB\\-\\-no\\-copy\\-dt\\-needed\\-entries\\fR 4\n .IX Item \"--no-copy-dt-needed-entries\"\n .PD\n This option affects the treatment of dynamic libraries referred to\n by DT_NEEDED tags \\fIinside\\fR ELF dynamic libraries mentioned on the\n-command line. Normally the linker won't add a DT_NEEDED tag to the\n+command line. Normally the linker won\\*(Aqt add a DT_NEEDED tag to the\n output binary for each library mentioned in a DT_NEEDED tag in an\n input dynamic library. With \\fB\\-\\-copy\\-dt\\-needed\\-entries\\fR\n specified on the command line however any dynamic libraries that\n follow it will have their DT_NEEDED entries added. The default\n behaviour can be restored with \\fB\\-\\-no\\-copy\\-dt\\-needed\\-entries\\fR.\n .Sp\n This option also has an effect on the resolution of symbols in dynamic\n@@ -1792,39 +1795,39 @@\n usually wasted: the types are usually small but the names are often not.\n \\&\\fB\\-\\-ctf\\-variables\\fR causes the generation of such a section.\n The default behaviour can be restored with \\fB\\-\\-no\\-ctf\\-variables\\fR.\n .IP \\fB\\-\\-ctf\\-share\\-types=\\fR\\fImethod\\fR 4\n .IX Item \"--ctf-share-types=method\"\n Adjust the method used to share types between translation units in CTF.\n .RS 4\n-.IP \\fBshare-unconflicted\\fR 4\n+.IP \\fBshare\\-unconflicted\\fR 4\n .IX Item \"share-unconflicted\"\n Put all types that do not have ambiguous definitions into the shared dictionary,\n where debuggers can easily access them, even if they only occur in one\n translation unit. This is the default.\n-.IP \\fBshare-duplicated\\fR 4\n+.IP \\fBshare\\-duplicated\\fR 4\n .IX Item \"share-duplicated\"\n Put only types that occur in multiple translation units into the shared\n-dictionary: types with only one definition go into per-translation-unit\n+dictionary: types with only one definition go into per\\-translation\\-unit\n dictionaries. Types with ambiguous definitions in multiple translation units\n-always go into per-translation-unit dictionaries. This tends to make the CTF\n+always go into per\\-translation\\-unit dictionaries. This tends to make the CTF\n larger, but may reduce the amount of CTF in the shared dictionary. For very\n large projects this may speed up opening the CTF and save memory in the CTF\n consumer at runtime.\n .RE\n .RS 4\n .RE\n .IP \\fB\\-\\-no\\-define\\-common\\fR 4\n .IX Item \"--no-define-common\"\n This option inhibits the assignment of addresses to common symbols.\n The script command \\f(CW\\*(C`INHIBIT_COMMON_ALLOCATION\\*(C'\\fR has the same effect.\n .Sp\n The \\fB\\-\\-no\\-define\\-common\\fR option allows decoupling\n the decision to assign addresses to Common symbols from the choice\n-of the output file type; otherwise a non-Relocatable output type\n+of the output file type; otherwise a non\\-Relocatable output type\n forces assigning addresses to Common symbols.\n Using \\fB\\-\\-no\\-define\\-common\\fR allows Common symbols that are referenced\n from a shared library to be assigned addresses only in the main program.\n This eliminates the unused duplicate space in the shared library,\n and also prevents any possible confusion over resolving to the wrong\n duplicate when there are many dynamic modules with specialized search\n paths for runtime symbol resolution.\n@@ -1876,26 +1879,26 @@\n .IX Item \"-Ifile\"\n .PD 0\n .IP \\fB\\-\\-dynamic\\-linker=\\fR\\fIfile\\fR 4\n .IX Item \"--dynamic-linker=file\"\n .PD\n Set the name of the dynamic linker. This is only meaningful when\n generating dynamically linked ELF executables. The default dynamic\n-linker is normally correct; don't use this unless you know what you are\n+linker is normally correct; don\\*(Aqt use this unless you know what you are\n doing.\n .IP \\fB\\-\\-no\\-dynamic\\-linker\\fR 4\n .IX Item \"--no-dynamic-linker\"\n When producing an executable file, omit the request for a dynamic\n-linker to be used at load-time. This is only meaningful for ELF\n+linker to be used at load\\-time. This is only meaningful for ELF\n executables that contain dynamic relocations, and usually requires\n entry point code that is capable of processing these relocations.\n .IP \\fB\\-\\-embedded\\-relocs\\fR 4\n .IX Item \"--embedded-relocs\"\n This option is similar to the \\fB\\-\\-emit\\-relocs\\fR option except\n-that the relocs are stored in a target-specific section. This option\n+that the relocs are stored in a target\\-specific section. This option\n is only supported by the \\fBBFIN\\fR, \\fBCR16\\fR and \\fIM68K\\fR\n targets.\n .IP \\fB\\-\\-disable\\-multiple\\-abs\\-defs\\fR 4\n .IX Item \"--disable-multiple-abs-defs\"\n Do not allow multiple definitions with symbols included\n in filename invoked by \\-R or \\-\\-just\\-symbols\n .IP \\fB\\-\\-fatal\\-warnings\\fR 4\n@@ -1920,15 +1923,15 @@\n .IX Item \"--force-exe-suffix\"\n Make sure that an output file has a .exe suffix.\n .Sp\n If a successfully built fully linked output file does not have a\n \\&\\f(CW\\*(C`.exe\\*(C'\\fR or \\f(CW\\*(C`.dll\\*(C'\\fR suffix, this option forces the linker to copy\n the output file to one of the same name with a \\f(CW\\*(C`.exe\\*(C'\\fR suffix. This\n option is useful when using unmodified Unix makefiles on a Microsoft\n-Windows host, since some versions of Windows won't run an image unless\n+Windows host, since some versions of Windows won\\*(Aqt run an image unless\n it ends in a \\f(CW\\*(C`.exe\\*(C'\\fR suffix.\n .IP \\fB\\-\\-gc\\-sections\\fR 4\n .IX Item \"--gc-sections\"\n .PD 0\n .IP \\fB\\-\\-no\\-gc\\-sections\\fR 4\n .IX Item \"--no-gc-sections\"\n .PD\n@@ -1938,15 +1941,15 @@\n \\&\\fB\\-\\-no\\-gc\\-sections\\fR on the command line. Note that garbage\n collection for COFF and PE format targets is supported, but the\n implementation is currently considered to be experimental.\n .Sp\n \\&\\fB\\-\\-gc\\-sections\\fR decides which input sections are used by\n examining symbols and relocations. The section containing the entry\n symbol and all sections containing symbols undefined on the\n-command-line will be kept, as will sections containing symbols\n+command\\-line will be kept, as will sections containing symbols\n referenced by dynamic objects. Note that when building shared\n libraries, the linker must assume that any visible symbol is\n referenced. Once this initial set of sections has been determined,\n the linker recursively marks as used any section referenced by their\n relocations. See \\fB\\-\\-entry\\fR, \\fB\\-\\-undefined\\fR, and\n \\&\\fB\\-\\-gc\\-keep\\-exported\\fR.\n .Sp\n@@ -1979,15 +1982,15 @@\n collected regardless of the external visibility of contained symbols.\n Note that this option has no effect when linking shared objects since\n it is already the default behaviour. This option is only supported for\n ELF format targets.\n .IP \\fB\\-\\-print\\-output\\-format\\fR 4\n .IX Item \"--print-output-format\"\n Print the name of the default output format (perhaps influenced by\n-other command-line options). This is the string that would appear\n+other command\\-line options). This is the string that would appear\n in an \\f(CW\\*(C`OUTPUT_FORMAT\\*(C'\\fR linker script command.\n .IP \\fB\\-\\-print\\-memory\\-usage\\fR 4\n .IX Item \"--print-memory-usage\"\n Print used size, total size and used size of memory regions created with\n the \\fBMEMORY\\fR command. This is useful on embedded targets to have a\n quick view of amount of free memory. The format of the output has one\n headline and one line per region. It is both human readable and easily\n@@ -1996,18 +1999,18 @@\n .Vb 3\n \\& Memory region Used Size Region Size %age Used\n \\& ROM: 256 KB 1 MB 25.00%\n \\& RAM: 32 B 2 GB 0.00%\n .Ve\n .IP \\fB\\-\\-help\\fR 4\n .IX Item \"--help\"\n-Print a summary of the command-line options on the standard output and exit.\n+Print a summary of the command\\-line options on the standard output and exit.\n .IP \\fB\\-\\-target\\-help\\fR 4\n .IX Item \"--target-help\"\n-Print a summary of all target-specific options on the standard output and exit.\n+Print a summary of all target\\-specific options on the standard output and exit.\n .IP \\fB\\-Map=\\fR\\fImapfile\\fR 4\n .IX Item \"-Map=mapfile\"\n Print a link map to the file \\fImapfile\\fR. See the description of the\n \\&\\fB\\-M\\fR option, above. If \\fImapfile\\fR is just the character\n \\&\\f(CW\\*(C`\\-\\*(C'\\fR then the map will be written to stdout.\n .Sp\n Specifying a directory as \\fImapfile\\fR causes the linker map to be\n@@ -2045,15 +2048,15 @@\n .IP \\fB\\-\\-no\\-undefined\\fR 4\n .IX Item \"--no-undefined\"\n .PD 0\n .IP \"\\fB\\-z defs\\fR\" 4\n .IX Item \"-z defs\"\n .PD\n Report unresolved symbol references from regular object files. This\n-is done even if the linker is creating a non-symbolic shared library.\n+is done even if the linker is creating a non\\-symbolic shared library.\n The switch \\fB\\-\\-[no\\-]allow\\-shlib\\-undefined\\fR controls the\n behaviour for reporting unresolved references found in shared\n libraries being linked in.\n .Sp\n The effects of this option can be reverted by using \\f(CW\\*(C`\\-z undefs\\*(C'\\fR.\n .IP \\fB\\-\\-allow\\-multiple\\-definition\\fR 4\n .IX Item \"--allow-multiple-definition\"\n@@ -2101,15 +2104,15 @@\n .RE\n .IP \\fB\\-\\-error\\-handling\\-script=\\fR\\fIscriptname\\fR 4\n .IX Item \"--error-handling-script=scriptname\"\n If this option is provided then the linker will invoke\n \\&\\fIscriptname\\fR whenever an error is encountered. Currently however\n only two kinds of error are supported: missing symbols and missing\n libraries. Two arguments will be passed to script: the keyword\n-\"undefined-symbol\" or `missing\\-lib\" and the \\fIname\\fR of the\n+\"undefined\\-symbol\" or \\`missing\\-lib\" and the \\fIname\\fR of the\n undefined symbol or missing library. The intention is that the script\n will provide suggestions to the user as to where the symbol or library\n might be found. After the script has finished then the normal linker\n error message will be displayed.\n .Sp\n The availability of this option is controlled by a configure time\n switch, so it may not be present in specific implementations.\n@@ -2150,23 +2153,23 @@\n errors during the link process; it exits without writing an output file\n when it issues any error whatsoever.\n .IP \\fB\\-nostdlib\\fR 4\n .IX Item \"-nostdlib\"\n Only search library directories explicitly specified on the\n command line. Library directories specified in linker scripts\n (including linker scripts specified on the command line) are ignored.\n-.IP \\fB\\-\\-oformat=\\fR\\fIoutput-format\\fR 4\n+.IP \\fB\\-\\-oformat=\\fR\\fIoutput\\-format\\fR 4\n .IX Item \"--oformat=output-format\"\n \\&\\fBld\\fR may be configured to support more than one kind of object\n file. If your \\fBld\\fR is configured this way, you can use the\n \\&\\fB\\-\\-oformat\\fR option to specify the binary format for the output\n object file. Even when \\fBld\\fR is configured to support alternative\n-object formats, you don't usually need to specify this, as \\fBld\\fR\n+object formats, you don\\*(Aqt usually need to specify this, as \\fBld\\fR\n should be configured to produce as a default output format the most\n-usual format on each machine. \\fIoutput-format\\fR is a text string, the\n+usual format on each machine. \\fIoutput\\-format\\fR is a text string, the\n name of a particular format supported by the BFD libraries. (You can\n list the available binary formats with \\fBobjdump \\-i\\fR.) The script\n command \\f(CW\\*(C`OUTPUT_FORMAT\\*(C'\\fR can also specify the output format, but\n this option overrides it.\n .IP \"\\fB\\-\\-out\\-implib\\fR \\fIfile\\fR\" 4\n .IX Item \"--out-implib file\"\n Create an import library in \\fIfile\\fR corresponding to the executable\n@@ -2226,15 +2229,15 @@\n .IP \\fB\\-\\-retain\\-symbols\\-file=\\fR\\fIfilename\\fR 4\n .IX Item \"--retain-symbols-file=filename\"\n Retain \\fIonly\\fR the symbols listed in the file \\fIfilename\\fR,\n discarding all others. \\fIfilename\\fR is simply a flat file, with one\n symbol name per line. This option is especially useful in environments\n (such as VxWorks)\n where a large global symbol table is accumulated gradually, to conserve\n-run-time memory.\n+run\\-time memory.\n .Sp\n \\&\\fB\\-\\-retain\\-symbols\\-file\\fR does \\fInot\\fR discard undefined symbols,\n or symbols needed for relocations.\n .Sp\n You may only specify \\fB\\-\\-retain\\-symbols\\-file\\fR once in the command\n line. It overrides \\fB\\-s\\fR and \\fB\\-S\\fR.\n .IP \\fB\\-rpath=\\fR\\fIdir\\fR 4\n@@ -2267,16 +2270,16 @@\n the \\fB\\-rpath\\fR option.\n .IP \\fB\\-rpath\\-link=\\fR\\fIdir\\fR 4\n .IX Item \"-rpath-link=dir\"\n When using ELF or SunOS, one shared library may require another. This\n happens when an \\f(CW\\*(C`ld \\-shared\\*(C'\\fR link includes a shared library as one\n of the input files.\n .Sp\n-When the linker encounters such a dependency when doing a non-shared,\n-non-relocatable link, it will automatically try to locate the required\n+When the linker encounters such a dependency when doing a non\\-shared,\n+non\\-relocatable link, it will automatically try to locate the required\n shared library and include it in the link, if it is not included\n explicitly. In such a case, several directories are searched as\n described below. The \\fB\\-rpath\\-link\\fR option specifies the first\n set of directories to search. This option may specify a sequence of\n directory names either by providing a list of names separated by\n colons, or by appearing multiple times.\n .Sp\n@@ -2424,16 +2427,16 @@\n .IP \\fB\\-\\-sort\\-common=ascending\\fR 4\n .IX Item \"--sort-common=ascending\"\n .IP \\fB\\-\\-sort\\-common=descending\\fR 4\n .IX Item \"--sort-common=descending\"\n .PD\n This option tells \\fBld\\fR to sort the common symbols by alignment in\n ascending or descending order when it places them in the appropriate output\n-sections. The symbol alignments considered are sixteen-byte or larger,\n-eight-byte, four-byte, two-byte, and one-byte. This is to prevent gaps\n+sections. The symbol alignments considered are sixteen\\-byte or larger,\n+eight\\-byte, four\\-byte, two\\-byte, and one\\-byte. This is to prevent gaps\n between symbols due to alignment constraints. If no sorting order is\n specified, then descending order is assumed.\n .IP \\fB\\-\\-sort\\-section=name\\fR 4\n .IX Item \"--sort-section=name\"\n This option will apply \\f(CW\\*(C`SORT_BY_NAME\\*(C'\\fR to all wildcard section\n patterns in the linker script.\n .IP \\fB\\-\\-sort\\-section=alignment\\fR 4\n@@ -2465,19 +2468,19 @@\n .IP \\fB\\-\\-stats\\fR 4\n .IX Item \"--stats\"\n Compute and display statistics about the operation of the linker, such\n as execution time and memory usage.\n .IP \\fB\\-\\-sysroot=\\fR\\fIdirectory\\fR 4\n .IX Item \"--sysroot=directory\"\n Use \\fIdirectory\\fR as the location of the sysroot, overriding the\n-configure-time default. This option is only supported by linkers\n+configure\\-time default. This option is only supported by linkers\n that were configured using \\fB\\-\\-with\\-sysroot\\fR.\n .IP \\fB\\-\\-task\\-link\\fR 4\n .IX Item \"--task-link\"\n-This is used by COFF/PE based targets to create a task-linked object\n+This is used by COFF/PE based targets to create a task\\-linked object\n file where all of the global symbols have been converted to statics.\n .IP \\fB\\-\\-traditional\\-format\\fR 4\n .IX Item \"--traditional-format\"\n For some targets, the output of \\fBld\\fR is different in some ways from\n the output of some existing linker. This switch requests \\fBld\\fR to\n use the traditional format instead.\n .Sp\n@@ -2525,41 +2528,41 @@\n byte of the first segment. Note that when \\fB\\-pie\\fR is used with\n \\&\\fB\\-Ttext\\-segment=\\fR\\fIorg\\fR, the output executable is marked\n ET_EXEC so that the address of the first byte of the text segment will\n be guaranteed to be \\fIorg\\fR at run time.\n .IP \\fB\\-Trodata\\-segment=\\fR\\fIorg\\fR 4\n .IX Item \"-Trodata-segment=org\"\n When creating an ELF executable or shared object for a target where\n-the read-only data is in its own segment separate from the executable\n-text, it will set the address of the first byte of the read-only data segment.\n+the read\\-only data is in its own segment separate from the executable\n+text, it will set the address of the first byte of the read\\-only data segment.\n .IP \\fB\\-Tldata\\-segment=\\fR\\fIorg\\fR 4\n .IX Item \"-Tldata-segment=org\"\n When creating an ELF executable or shared object for x86\\-64 medium memory\n model, it will set the address of the first byte of the ldata segment.\n .IP \\fB\\-\\-unresolved\\-symbols=\\fR\\fImethod\\fR 4\n .IX Item \"--unresolved-symbols=method\"\n Determine how to handle unresolved symbols. There are four possible\n values for \\fBmethod\\fR:\n .RS 4\n-.IP \\fBignore-all\\fR 4\n+.IP \\fBignore\\-all\\fR 4\n .IX Item \"ignore-all\"\n Do not report any unresolved symbols.\n-.IP \\fBreport-all\\fR 4\n+.IP \\fBreport\\-all\\fR 4\n .IX Item \"report-all\"\n Report all unresolved symbols. This is the default.\n-.IP \\fBignore-in-object-files\\fR 4\n+.IP \\fBignore\\-in\\-object\\-files\\fR 4\n .IX Item \"ignore-in-object-files\"\n Report unresolved symbols that are contained in shared libraries, but\n ignore them if they come from regular object files.\n-.IP \\fBignore-in-shared-libs\\fR 4\n+.IP \\fBignore\\-in\\-shared\\-libs\\fR 4\n .IX Item \"ignore-in-shared-libs\"\n Report unresolved symbols that come from regular object files, but\n ignore them if they come from shared libraries. This can be useful\n when creating a dynamic binary and it is known that all the shared\n-libraries that it should be referencing are included on the linker's\n+libraries that it should be referencing are included on the linker\\*(Aqs\n command line.\n .RE\n .RS 4\n .Sp\n The behaviour for shared libraries on their own can also be controlled\n by the \\fB\\-\\-[no\\-]allow\\-shlib\\-undefined\\fR option.\n .Sp\n@@ -2573,22 +2576,22 @@\n .IP \\fB\\-\\-verbose[=\\fR\\fINUMBER\\fR\\fB]\\fR 4\n .IX Item \"--verbose[=NUMBER]\"\n .PD\n Display the version number for \\fBld\\fR and list the linker emulations\n supported. Display which input files can and cannot be opened. Display\n the linker script being used by the linker. If the optional \\fINUMBER\\fR\n argument > 1, plugin symbol status will also be displayed.\n-.IP \\fB\\-\\-version\\-script=\\fR\\fIversion-scriptfile\\fR 4\n+.IP \\fB\\-\\-version\\-script=\\fR\\fIversion\\-scriptfile\\fR 4\n .IX Item \"--version-script=version-scriptfile\"\n Specify the name of a version script to the linker. This is typically\n used when creating shared libraries to specify additional information\n about the version hierarchy for the library being created. This option\n is only fully supported on ELF platforms which support shared libraries;\n see \\fBVERSION\\fR. It is partially supported on PE platforms, which can\n-use version scripts to filter symbol visibility in auto-export mode: any\n+use version scripts to filter symbol visibility in auto\\-export mode: any\n symbols marked \\fBlocal\\fR in the version script will not be exported.\n .IP \\fB\\-\\-warn\\-common\\fR 4\n .IX Item \"--warn-common\"\n Warn when a common symbol is combined with another common symbol or with\n a symbol definition. Unix linkers allow this somewhat sloppy practice,\n but linkers on some other operating systems do not. This option allows\n you to find potential problems from combining global symbols.\n@@ -2638,15 +2641,15 @@\n .Sp\n .Vb 3\n \\& (
): warning: definition of \\`\\*(Aq\n \\& overriding common\n \\& (
): warning: common is here\n .Ve\n .IP 3. 4\n-Merging a common symbol with a previous same-sized common symbol.\n+Merging a common symbol with a previous same\\-sized common symbol.\n .Sp\n .Vb 3\n \\& (
): warning: multiple common\n \\& of \\`\\*(Aq\n \\& (
): warning: previous common is here\n .Ve\n .IP 4. 4\n@@ -2696,18 +2699,18 @@\n Only generate a warning if an object file requests an executable\n stack, but not if the \\fB\\-z execstack\\fR option is used.\n .RE\n .RS 4\n .Sp\n The default state depends upon how the linker was configured when it\n was built. The \\fB\\-\\-no\\-warn\\-execstack\\fR option always puts the\n-linker into the no-warnings state. The \\fB\\-\\-warn\\-execstack\\fR\n-option puts the linker into the warn-always state. The\n+linker into the no\\-warnings state. The \\fB\\-\\-warn\\-execstack\\fR\n+option puts the linker into the warn\\-always state. The\n \\&\\fB\\-\\-warn\\-execstack\\-objects\\fR option puts the linker into the\n-warn-for-object-files-only state.\n+warn\\-for\\-object\\-files\\-only state.\n .Sp\n Note: ELF format input files can specify that they need an executable\n stack by having a \\fI.note.GNU\\-stack\\fR section with the executable\n bit set in its section flags. They can specify that they do not need\n an executable stack by having the same section, but without the\n executable flag bit set. If an input file does not have a\n \\&\\fI.note.GNU\\-stack\\fR section then the default behaviour is target\n@@ -2720,52 +2723,52 @@\n .PD 0\n .IP \\fB\\-\\-no\\-error\\-execstack\\fR 4\n .IX Item \"--no-error-execstack\"\n .PD\n If the linker is going to generate a warning message about an\n executable stack then the \\fB\\-\\-error\\-execstack\\fR option will\n instead change that warning into an error. Note \\- this option does\n-not change the linker's execstack warning generation state. Use\n+not change the linker\\*(Aqs execstack warning generation state. Use\n \\&\\fB\\-\\-warn\\-execstack\\fR or \\fB\\-\\-warn\\-execstack\\-objects\\fR to set\n a specific warning state.\n .Sp\n The \\fB\\-\\-no\\-error\\-execstack\\fR option will restore the default\n behaviour of generating warning messages.\n .IP \\fB\\-\\-warn\\-multiple\\-gp\\fR 4\n .IX Item \"--warn-multiple-gp\"\n Warn if multiple global pointer values are required in the output file.\n This is only meaningful for certain processors, such as the Alpha.\n-Specifically, some processors put large-valued constants in a special\n+Specifically, some processors put large\\-valued constants in a special\n section. A special register (the global pointer) points into the middle\n of this section, so that constants can be loaded efficiently via a\n-base-register relative addressing mode. Since the offset in\n-base-register relative mode is fixed and relatively small (e.g., 16\n+base\\-register relative addressing mode. Since the offset in\n+base\\-register relative mode is fixed and relatively small (e.g., 16\n bits), this limits the maximum size of the constant pool. Thus, in\n large programs, it is often necessary to use multiple global pointer\n values in order to be able to address all possible constants. This\n option causes a warning to be issued whenever this case occurs.\n .IP \\fB\\-\\-warn\\-once\\fR 4\n .IX Item \"--warn-once\"\n Only warn once for each undefined symbol, rather than once per module\n which refers to it.\n .IP \\fB\\-\\-warn\\-rwx\\-segments\\fR 4\n .IX Item \"--warn-rwx-segments\"\n .PD 0\n .IP \\fB\\-\\-no\\-warn\\-rwx\\-segments\\fR 4\n .IX Item \"--no-warn-rwx-segments\"\n .PD\n-Warn if the linker creates a loadable, non-zero sized segment that has\n+Warn if the linker creates a loadable, non\\-zero sized segment that has\n all three of the read, write and execute permission flags set. Such a\n segment represents a potential security vulnerability. In addition\n warnings will be generated if a thread local storage segment is\n created with the execute permission flag set, regardless of whether or\n not it has the read and/or write flags set.\n .Sp\n These warnings are enabled by default. They can be disabled via the\n-\\&\\fB\\-\\-no\\-warn\\-rwx\\-segments\\fR option and re-enabled via the\n+\\&\\fB\\-\\-no\\-warn\\-rwx\\-segments\\fR option and re\\-enabled via the\n \\&\\fB\\-\\-warn\\-rwx\\-segments\\fR option.\n .IP \\fB\\-\\-error\\-rwx\\-segments\\fR 4\n .IX Item \"--error-rwx-segments\"\n .PD 0\n .IP \\fB\\-\\-no\\-error\\-rwx\\-segments\\fR 4\n .IX Item \"--no-error-rwx-segments\"\n .PD\n@@ -2784,40 +2787,40 @@\n Warn if the address of an output section is changed because of\n alignment. Typically, the alignment will be set by an input section.\n The address will only be changed if it not explicitly specified; that\n is, if the \\f(CW\\*(C`SECTIONS\\*(C'\\fR command does not specify a start address for\n the section.\n .IP \\fB\\-\\-warn\\-textrel\\fR 4\n .IX Item \"--warn-textrel\"\n-Warn if the linker adds DT_TEXTREL to a position-independent executable\n+Warn if the linker adds DT_TEXTREL to a position\\-independent executable\n or shared object.\n .IP \\fB\\-\\-warn\\-alternate\\-em\\fR 4\n .IX Item \"--warn-alternate-em\"\n Warn if an object has alternate ELF machine code.\n .IP \\fB\\-\\-warn\\-unresolved\\-symbols\\fR 4\n .IX Item \"--warn-unresolved-symbols\"\n If the linker is going to report an unresolved symbol (see the option\n \\&\\fB\\-\\-unresolved\\-symbols\\fR) it will normally generate an error.\n This option makes it generate a warning instead.\n .IP \\fB\\-\\-error\\-unresolved\\-symbols\\fR 4\n .IX Item \"--error-unresolved-symbols\"\n-This restores the linker's default behaviour of generating errors when\n+This restores the linker\\*(Aqs default behaviour of generating errors when\n it is reporting unresolved symbols.\n .IP \\fB\\-\\-whole\\-archive\\fR 4\n .IX Item \"--whole-archive\"\n For each archive mentioned on the command line after the\n \\&\\fB\\-\\-whole\\-archive\\fR option, include every object file in the archive\n in the link, rather than searching the archive for the required object\n files. This is normally used to turn an archive file into a shared\n library, forcing every object to be included in the resulting shared\n library. This option may be used more than once.\n .Sp\n-Two notes when using this option from gcc: First, gcc doesn't know\n+Two notes when using this option from gcc: First, gcc doesn\\*(Aqt know\n about this option, so you have to use \\fB\\-Wl,\\-whole\\-archive\\fR.\n-Second, don't forget to use \\fB\\-Wl,\\-no\\-whole\\-archive\\fR after your\n+Second, don\\*(Aqt forget to use \\fB\\-Wl,\\-no\\-whole\\-archive\\fR after your\n list of archives, because gcc will add its own list of archives to\n your link and you may not want this flag to affect those as well.\n .IP \\fB\\-\\-wrap=\\fR\\fIsymbol\\fR 4\n .IX Item \"--wrap=symbol\"\n Use a wrapper function for \\fIsymbol\\fR. Any undefined reference to\n \\&\\fIsymbol\\fR will be resolved to \\f(CW\\*(C`_\\|_wrap_\\fR\\f(CIsymbol\\fR\\f(CW\\*(C'\\fR. Any\n undefined reference to \\f(CW\\*(C`_\\|_real_\\fR\\f(CIsymbol\\fR\\f(CW\\*(C'\\fR will be resolved to\n@@ -2895,22 +2898,22 @@\n \\&\\fB\\-\\-enable\\-new\\-dtags\\fR, the new dynamic tags will be created as needed\n and older dynamic tags will be omitted.\n If you specify \\fB\\-\\-disable\\-new\\-dtags\\fR, no new dynamic tags will be\n created. By default, the new dynamic tags are not created. Note that\n those options are only available for ELF systems.\n .IP \\fB\\-\\-hash\\-size=\\fR\\fInumber\\fR 4\n .IX Item \"--hash-size=number\"\n-Set the default size of the linker's hash tables to a prime number\n+Set the default size of the linker\\*(Aqs hash tables to a prime number\n close to \\fInumber\\fR. Increasing this value can reduce the length of\n time it takes the linker to perform its tasks, at the expense of\n-increasing the linker's memory requirements. Similarly reducing this\n+increasing the linker\\*(Aqs memory requirements. Similarly reducing this\n value can reduce the memory requirements at the expense of speed.\n .IP \\fB\\-\\-hash\\-style=\\fR\\fIstyle\\fR 4\n .IX Item \"--hash-style=style\"\n-Set the type of linker's hash table(s). \\fIstyle\\fR can be either\n+Set the type of linker\\*(Aqs hash table(s). \\fIstyle\\fR can be either\n \\&\\f(CW\\*(C`sysv\\*(C'\\fR for classic ELF \\f(CW\\*(C`.hash\\*(C'\\fR section, \\f(CW\\*(C`gnu\\*(C'\\fR for\n new style GNU \\f(CW\\*(C`.gnu.hash\\*(C'\\fR section or \\f(CW\\*(C`both\\*(C'\\fR for both\n the classic ELF \\f(CW\\*(C`.hash\\*(C'\\fR and new style GNU \\f(CW\\*(C`.gnu.hash\\*(C'\\fR\n hash tables. The default depends upon how the linker was configured,\n but for most Linux based systems it will be \\f(CW\\*(C`both\\*(C'\\fR.\n .IP \\fB\\-\\-compress\\-debug\\-sections=none\\fR 4\n .IX Item \"--compress-debug-sections=none\"\n@@ -2923,45 +2926,45 @@\n .IX Item \"--compress-debug-sections=zlib-gabi\"\n .IP \\fB\\-\\-compress\\-debug\\-sections=zstd\\fR 4\n .IX Item \"--compress-debug-sections=zstd\"\n .PD\n On ELF platforms, these options control how DWARF debug sections are\n compressed using zlib.\n .Sp\n-\\&\\fB\\-\\-compress\\-debug\\-sections=none\\fR doesn't compress DWARF debug\n+\\&\\fB\\-\\-compress\\-debug\\-sections=none\\fR doesn\\*(Aqt compress DWARF debug\n sections. \\fB\\-\\-compress\\-debug\\-sections=zlib\\-gnu\\fR compresses\n DWARF debug sections and renames them to begin with \\fB.zdebug\\fR\n instead of \\fB.debug\\fR. \\fB\\-\\-compress\\-debug\\-sections=zlib\\-gabi\\fR\n also compresses DWARF debug sections, but rather than renaming them it\n-sets the SHF_COMPRESSED flag in the sections' headers.\n+sets the SHF_COMPRESSED flag in the sections\\*(Aq headers.\n .Sp\n The \\fB\\-\\-compress\\-debug\\-sections=zlib\\fR option is an alias for\n \\&\\fB\\-\\-compress\\-debug\\-sections=zlib\\-gabi\\fR.\n .Sp\n \\&\\fB\\-\\-compress\\-debug\\-sections=zstd\\fR compresses DWARF debug sections using\n zstd.\n .Sp\n Note that this option overrides any compression in input debug\n sections, so if a binary is linked with \\fB\\-\\-compress\\-debug\\-sections=none\\fR\n for example, then any compressed debug sections in input files will be\n uncompressed before they are copied into the output binary.\n .Sp\n The default compression behaviour varies depending upon the target\n involved and the configure options used to build the toolchain. The\n-default can be determined by examining the output from the linker's\n+default can be determined by examining the output from the linker\\*(Aqs\n \\&\\fB\\-\\-help\\fR option.\n .IP \\fB\\-\\-reduce\\-memory\\-overheads\\fR 4\n .IX Item \"--reduce-memory-overheads\"\n This option reduces memory requirements at ld runtime, at the expense of\n linking speed. This was introduced to select the old O(n^2) algorithm\n for link map file generation, rather than the new O(n) algorithm which uses\n about 40% more memory for symbol storage.\n .Sp\n Another effect of the switch is to set the default hash table size to\n-1021, which again saves memory at the cost of lengthening the linker's\n+1021, which again saves memory at the cost of lengthening the linker\\*(Aqs\n run time. This is not done however if the \\fB\\-\\-hash\\-size\\fR switch\n has been used.\n .Sp\n The \\fB\\-\\-reduce\\-memory\\-overheads\\fR switch may be also be used to\n enable other tradeoffs in future versions of the linker.\n .IP \\fB\\-\\-max\\-cache\\-size=\\fR\\fIsize\\fR 4\n .IX Item \"--max-cache-size=size\"\n@@ -2984,28 +2987,28 @@\n specified as an even number of hexadecimal digits (\\f(CW\\*(C`\\-\\*(C'\\fR and\n \\&\\f(CW\\*(C`:\\*(C'\\fR characters between digit pairs are ignored). If \\fIstyle\\fR\n is omitted, \\f(CW\\*(C`sha1\\*(C'\\fR is used.\n .Sp\n The \\f(CW\\*(C`md5\\*(C'\\fR, \\f(CW\\*(C`sha1\\*(C'\\fR, and \\f(CW\\*(C`xx\\*(C'\\fR styles produces an\n identifier that is always the same in an identical output file, but\n are almost certainly unique among all nonidentical output files. It\n-is not intended to be compared as a checksum for the file's contents.\n+is not intended to be compared as a checksum for the file\\*(Aqs contents.\n A linked file may be changed later by other tools, but the build ID\n bit string identifying the original linked file does not change.\n .Sp\n Passing \\f(CW\\*(C`none\\*(C'\\fR for \\fIstyle\\fR disables the setting from any\n \\&\\f(CW\\*(C`\\-\\-build\\-id\\*(C'\\fR options earlier on the command line.\n .IP \\fB\\-\\-package\\-metadata=\\fR\\fIJSON\\fR 4\n .IX Item \"--package-metadata=JSON\"\n Request the creation of a \\f(CW\\*(C`.note.package\\*(C'\\fR ELF note section. The\n contents of the note are in JSON format, as per the package metadata\n specification. For more information see:\n https://systemd.io/ELF_PACKAGE_METADATA/\n-The JSON argument support percent-encoding and following %[string]\n-(where string refers to the name in HTML's Named Character References)\n+The JSON argument support percent\\-encoding and following %[string]\n+(where string refers to the name in HTML\\*(Aqs Named Character References)\n encoding: \\fB%[comma]\\fR for \\fB,\\fR, \\fB%[lbrace]\\fR for \\fB{\\fR,\n \\&\\fB%[quot]\\fR for \\fB\"\\fR, \\fB%[rbrace]\\fR for \\fB}\\fR, and\n \\&\\fB%[space]\\fR for space character.\n If the JSON argument is missing/empty then this will disable the\n creation of the metadata note, if one had been enabled by an earlier\n occurrence of the \\-\\-package\\-metadata option.\n If the linker has been built with libjansson, then the JSON string\n@@ -3017,21 +3020,21 @@\n use this option. In addition, the linker fully supports the standard\n \\&\\f(CW\\*(C`*.def\\*(C'\\fR files, which may be specified on the linker command line\n like an object file (in fact, it should precede archives it exports\n symbols from, to ensure that they get linked in, just like a normal\n object file).\n .PP\n In addition to the options common to all targets, the i386 PE linker\n-support additional command-line options that are specific to the i386\n+support additional command\\-line options that are specific to the i386\n PE target. Options that take values may be separated from their\n values by either a space or an equals sign.\n .IP \\fB\\-\\-add\\-stdcall\\-alias\\fR 4\n .IX Item \"--add-stdcall-alias\"\n If given, symbols with a stdcall suffix (@\\fInn\\fR) will be exported\n-as-is and also with the suffix stripped.\n+as\\-is and also with the suffix stripped.\n [This option is specific to the i386 PE targeted port of the linker]\n .IP \"\\fB\\-\\-base\\-file\\fR \\fIfile\\fR\" 4\n .IX Item \"--base-file file\"\n Use \\fIfile\\fR as the name of a file in which to save the base\n addresses of all the relocations needed for generating DLLs with\n \\&\\fIdlltool\\fR.\n [This is an i386 PE specific option]\n@@ -3046,24 +3049,24 @@\n .PD 0\n .IP \\fB\\-\\-disable\\-long\\-section\\-names\\fR 4\n .IX Item \"--disable-long-section-names\"\n .PD\n The PE variants of the COFF object format add an extension that permits\n the use of section names longer than eight characters, the normal limit\n for COFF. By default, these names are only allowed in object files, as\n-fully-linked executable images do not carry the COFF string table required\n+fully\\-linked executable images do not carry the COFF string table required\n to support the longer names. As a GNU extension, it is possible to\n allow their use in executable images as well, or to (probably pointlessly!)\n disallow it in object files, by using these two options. Executable images\n-generated with these long section names are slightly non-standard, carrying\n+generated with these long section names are slightly non\\-standard, carrying\n as they do a string table, and may generate confusing output when examined\n-with non-GNU PE-aware tools, such as file viewers and dumpers. However,\n+with non\\-GNU PE\\-aware tools, such as file viewers and dumpers. However,\n GDB relies on the use of PE long section names to find Dwarf\\-2 debug\n information sections in an executable image at runtime, and so if neither\n-option is specified on the command-line, \\fBld\\fR will enable long\n+option is specified on the command\\-line, \\fBld\\fR will enable long\n section names, overriding the default and technically correct behaviour,\n when it finds the presence of debug information while linking an executable\n image and not stripping symbols.\n [This option is valid for all PE targeted ports of the linker]\n .IP \\fB\\-\\-enable\\-stdcall\\-fixup\\fR 4\n .IX Item \"--enable-stdcall-fixup\"\n .PD 0\n@@ -3074,49 +3077,49 @@\n do \"fuzzy linking\" by looking for another defined symbol that differs\n only in the format of the symbol name (cdecl vs stdcall) and will\n resolve that symbol by linking to the match. For example, the\n undefined symbol \\f(CW\\*(C`_foo\\*(C'\\fR might be linked to the function\n \\&\\f(CW\\*(C`_foo@12\\*(C'\\fR, or the undefined symbol \\f(CW\\*(C`_bar@16\\*(C'\\fR might be linked\n to the function \\f(CW\\*(C`_bar\\*(C'\\fR. When the linker does this, it prints a\n warning, since it normally should have failed to link, but sometimes\n-import libraries generated from third-party dlls may need this feature\n+import libraries generated from third\\-party dlls may need this feature\n to be usable. If you specify \\fB\\-\\-enable\\-stdcall\\-fixup\\fR, this\n feature is fully enabled and warnings are not printed. If you specify\n \\&\\fB\\-\\-disable\\-stdcall\\-fixup\\fR, this feature is disabled and such\n mismatches are considered to be errors.\n [This option is specific to the i386 PE targeted port of the linker]\n .IP \\fB\\-\\-leading\\-underscore\\fR 4\n .IX Item \"--leading-underscore\"\n .PD 0\n .IP \\fB\\-\\-no\\-leading\\-underscore\\fR 4\n .IX Item \"--no-leading-underscore\"\n .PD\n-For most targets default symbol-prefix is an underscore and is defined\n-in target's description. By this option it is possible to\n-disable/enable the default underscore symbol-prefix.\n+For most targets default symbol\\-prefix is an underscore and is defined\n+in target\\*(Aqs description. By this option it is possible to\n+disable/enable the default underscore symbol\\-prefix.\n .IP \\fB\\-\\-export\\-all\\-symbols\\fR 4\n .IX Item \"--export-all-symbols\"\n If given, all global symbols in the objects used to build a DLL will\n be exported by the DLL. Note that this is the default if there\n-otherwise wouldn't be any exported symbols. When symbols are\n+otherwise wouldn\\*(Aqt be any exported symbols. When symbols are\n explicitly exported via DEF files or implicitly exported via function\n attributes, the default is to not export anything else unless this\n option is given. Note that the symbols \\f(CW\\*(C`DllMain@12\\*(C'\\fR,\n \\&\\f(CW\\*(C`DllEntryPoint@0\\*(C'\\fR, \\f(CW\\*(C`DllMainCRTStartup@12\\*(C'\\fR, and\n \\&\\f(CW\\*(C`impure_ptr\\*(C'\\fR will not be automatically\n exported. Also, symbols imported from other DLLs will not be\n-re-exported, nor will symbols specifying the DLL's internal layout\n+re\\-exported, nor will symbols specifying the DLL\\*(Aqs internal layout\n such as those beginning with \\f(CW\\*(C`_head_\\*(C'\\fR or ending with\n \\&\\f(CW\\*(C`_iname\\*(C'\\fR. In addition, no symbols from \\f(CW\\*(C`libgcc\\*(C'\\fR,\n \\&\\f(CW\\*(C`libstd++\\*(C'\\fR, \\f(CW\\*(C`libmingw32\\*(C'\\fR, or \\f(CW\\*(C`crtX.o\\*(C'\\fR will be exported.\n Symbols whose names begin with \\f(CW\\*(C`_\\|_rtti_\\*(C'\\fR or \\f(CW\\*(C`_\\|_builtin_\\*(C'\\fR will\n not be exported, to help with C++ DLLs. Finally, there is an\n-extensive list of cygwin-private symbols that are not exported\n+extensive list of cygwin\\-private symbols that are not exported\n (obviously, this applies on when building DLLs for cygwin targets).\n-These cygwin-excludes are: \\f(CW\\*(C`_cygwin_dll_entry@12\\*(C'\\fR,\n+These cygwin\\-excludes are: \\f(CW\\*(C`_cygwin_dll_entry@12\\*(C'\\fR,\n \\&\\f(CW\\*(C`_cygwin_crt0_common@8\\*(C'\\fR, \\f(CW\\*(C`_cygwin_noncygwin_dll_entry@12\\*(C'\\fR,\n \\&\\f(CW\\*(C`_fmode\\*(C'\\fR, \\f(CW\\*(C`_impure_ptr\\*(C'\\fR, \\f(CW\\*(C`cygwin_attach_dll\\*(C'\\fR,\n \\&\\f(CW\\*(C`cygwin_premain0\\*(C'\\fR, \\f(CW\\*(C`cygwin_premain1\\*(C'\\fR, \\f(CW\\*(C`cygwin_premain2\\*(C'\\fR,\n \\&\\f(CW\\*(C`cygwin_premain3\\*(C'\\fR, and \\f(CW\\*(C`environ\\*(C'\\fR.\n [This option is specific to the i386 PE targeted port of the linker]\n .IP \"\\fB\\-\\-exclude\\-symbols\\fR \\fIsymbol\\fR\\fB,\\fR\\fIsymbol\\fR\\fB,...\\fR\" 4\n .IX Item \"--exclude-symbols symbol,symbol,...\"\n@@ -3200,21 +3203,21 @@\n .PD 0\n .IP \\fB\\-\\-enable\\-auto\\-image\\-base=\\fR\\fIvalue\\fR 4\n .IX Item \"--enable-auto-image-base=value\"\n .PD\n Automatically choose the image base for DLLs, optionally starting with base\n \\&\\fIvalue\\fR, unless one is specified using the \\f(CW\\*(C`\\-\\-image\\-base\\*(C'\\fR argument.\n By using a hash generated from the dllname to create unique image bases\n-for each DLL, in-memory collisions and relocations which can delay program\n+for each DLL, in\\-memory collisions and relocations which can delay program\n execution are avoided.\n [This option is specific to the i386 PE targeted port of the linker]\n .IP \\fB\\-\\-disable\\-auto\\-image\\-base\\fR 4\n .IX Item \"--disable-auto-image-base\"\n Do not automatically generate a unique image base. If there is no\n-user-specified image base (\\f(CW\\*(C`\\-\\-image\\-base\\*(C'\\fR) then use the platform\n+user\\-specified image base (\\f(CW\\*(C`\\-\\-image\\-base\\*(C'\\fR) then use the platform\n default.\n [This option is specific to the i386 PE targeted port of the linker]\n .IP \"\\fB\\-\\-dll\\-search\\-prefix\\fR \\fIstring\\fR\" 4\n .IX Item \"--dll-search-prefix string\"\n When linking dynamically to a dll without an import library,\n search for \\f(CW\\*(C`.dll\\*(C'\\fR in preference to\n \\&\\f(CW\\*(C`lib.dll\\*(C'\\fR. This behaviour allows easy distinction\n@@ -3228,29 +3231,29 @@\n DATA imports from DLLs, thus making it possible to bypass the dllimport\n mechanism on the user side and to reference unmangled symbol names.\n [This option is specific to the i386 PE targeted port of the linker]\n .Sp\n The following remarks pertain to the original implementation of the\n feature and are obsolete nowadays for Cygwin and MinGW targets.\n .Sp\n-Note: Use of the 'auto\\-import' extension will cause the text section\n+Note: Use of the \\*(Aqauto\\-import\\*(Aq extension will cause the text section\n of the image file to be made writable. This does not conform to the\n-PE-COFF format specification published by Microsoft.\n+PE\\-COFF format specification published by Microsoft.\n .Sp\n-Note \\- use of the 'auto\\-import' extension will also cause read only\n+Note \\- use of the \\*(Aqauto\\-import\\*(Aq extension will also cause read only\n data which would normally be placed into the .rdata section to be\n placed into the .data section instead. This is in order to work\n around a problem with consts that is described here:\n http://www.cygwin.com/ml/cygwin/2004\\-09/msg01101.html\n .Sp\n-Using 'auto\\-import' generally will 'just work' \\-\\- but sometimes you may\n+Using \\*(Aqauto\\-import\\*(Aq generally will \\*(Aqjust work\\*(Aq \\-\\- but sometimes you may\n see this message:\n .Sp\n-\"variable '' can't be auto-imported. Please read the\n-documentation for ld's \\f(CW\\*(C`\\-\\-enable\\-auto\\-import\\*(C'\\fR for details.\"\n+\"variable \\*(Aq\\*(Aq can\\*(Aqt be auto\\-imported. Please read the\n+documentation for ld\\*(Aqs \\f(CW\\*(C`\\-\\-enable\\-auto\\-import\\*(C'\\fR for details.\"\n .Sp\n This message occurs when some (sub)expression accesses an address\n ultimately given by the sum of two constants (Win32 import tables only\n allow one). Instances where this may occur include accesses to member\n fields of struct variables imported from a DLL, as well as using a\n constant index into an array variable imported from a DLL. Any\n multiword variable (arrays, structs, long long, etc) may trigger\n@@ -3261,18 +3264,18 @@\n There are several ways to address this difficulty, regardless of the\n data type of the exported variable:\n .Sp\n One way is to use \\-\\-enable\\-runtime\\-pseudo\\-reloc switch. This leaves the task\n of adjusting references in your client code for runtime environment, so\n this method works only when runtime environment supports this feature.\n .Sp\n-A second solution is to force one of the 'constants' to be a variable \\-\\-\n-that is, unknown and un-optimizable at compile time. For arrays,\n-there are two possibilities: a) make the indexee (the array's address)\n-a variable, or b) make the 'constant' index a variable. Thus:\n+A second solution is to force one of the \\*(Aqconstants\\*(Aq to be a variable \\-\\-\n+that is, unknown and un\\-optimizable at compile time. For arrays,\n+there are two possibilities: a) make the indexee (the array\\*(Aqs address)\n+a variable, or b) make the \\*(Aqconstant\\*(Aq index a variable. Thus:\n .Sp\n .Vb 3\n \\& extern type extern_array[];\n \\& extern_array[1] \\-\\->\n \\& { volatile type *t=extern_array; t[1] }\n .Ve\n .Sp\n@@ -3298,21 +3301,21 @@\n .Vb 3\n \\& extern long long extern_ll;\n \\& extern_ll \\-\\->\n \\& { volatile long long * local_ll=&extern_ll; *local_ll }\n .Ve\n .Sp\n A third method of dealing with this difficulty is to abandon\n-\\&'auto\\-import' for the offending symbol and mark it with\n+\\&\\*(Aqauto\\-import\\*(Aq for the offending symbol and mark it with\n \\&\\f(CW\\*(C`_\\|_declspec(dllimport)\\*(C'\\fR. However, in practice that\n-requires using compile-time #defines to indicate whether you are\n+requires using compile\\-time #defines to indicate whether you are\n building a DLL, building client code that will link to the DLL, or\n merely building/linking to a static library. In making the choice\n-between the various methods of resolving the 'direct address with\n-constant offset' problem, you should consider typical real-world usage:\n+between the various methods of resolving the \\*(Aqdirect address with\n+constant offset\\*(Aq problem, you should consider typical real\\-world usage:\n .Sp\n Original:\n .Sp\n .Vb 7\n \\& \\-\\-foo.h\n \\& extern int arr[];\n \\& \\-\\-foo.c\n@@ -3351,37 +3354,37 @@\n \\& \\-\\-foo.c\n \\& #include \"foo.h\"\n \\& void main(int argc, char **argv){\n \\& printf(\"%d\\en\",arr[1]);\n \\& }\n .Ve\n .Sp\n-A fourth way to avoid this problem is to re-code your\n+A fourth way to avoid this problem is to re\\-code your\n library to use a functional interface rather than a data interface\n for the offending variables (e.g. \\fBset_foo()\\fR and \\fBget_foo()\\fR accessor\n functions).\n .IP \\fB\\-\\-disable\\-auto\\-import\\fR 4\n .IX Item \"--disable-auto-import\"\n Do not attempt to do sophisticated linking of \\f(CW\\*(C`_symbol\\*(C'\\fR to\n \\&\\f(CW\\*(C`_\\|_imp_\\|_symbol\\*(C'\\fR for DATA imports from DLLs.\n [This option is specific to the i386 PE targeted port of the linker]\n .IP \\fB\\-\\-enable\\-runtime\\-pseudo\\-reloc\\fR 4\n .IX Item \"--enable-runtime-pseudo-reloc\"\n If your code contains expressions described in \\-\\-enable\\-auto\\-import section,\n-that is, DATA imports from DLL with non-zero offset, this switch will create\n-a vector of 'runtime pseudo relocations' which can be used by runtime\n+that is, DATA imports from DLL with non\\-zero offset, this switch will create\n+a vector of \\*(Aqruntime pseudo relocations\\*(Aq which can be used by runtime\n environment to adjust references to such data in your client code.\n [This option is specific to the i386 PE targeted port of the linker]\n .IP \\fB\\-\\-disable\\-runtime\\-pseudo\\-reloc\\fR 4\n .IX Item \"--disable-runtime-pseudo-reloc\"\n-Do not create pseudo relocations for non-zero offset DATA imports from DLLs.\n+Do not create pseudo relocations for non\\-zero offset DATA imports from DLLs.\n [This option is specific to the i386 PE targeted port of the linker]\n .IP \\fB\\-\\-enable\\-extra\\-pe\\-debug\\fR 4\n .IX Item \"--enable-extra-pe-debug\"\n-Show additional debug info related to auto-import symbol thunking.\n+Show additional debug info related to auto\\-import symbol thunking.\n [This option is specific to the i386 PE targeted port of the linker]\n .IP \\fB\\-\\-section\\-alignment\\fR 4\n .IX Item \"--section-alignment\"\n Sets the section alignment. Sections in memory will always begin at\n addresses which are a multiple of this number. Defaults to 0x1000.\n [This option is specific to the i386 PE targeted port of the linker]\n .IP \"\\fB\\-\\-stack\\fR \\fIreserve\\fR\" 4\n@@ -3624,71 +3627,71 @@\n .IP \\fB\\-N\\fR 4\n .IX Item \"-N\"\n .PD 0\n .IP \\fB\\-\\-omagic\\fR 4\n .IX Item \"--omagic\"\n .PD\n Mark the output as \\f(CW\\*(C`OMAGIC\\*(C'\\fR (0407) in the \\fIa.out\\fR header to\n-indicate that the text segment is not to be write-protected and\n+indicate that the text segment is not to be write\\-protected and\n shared. Since the text and data sections are both readable and\n writable, the data section is allocated immediately contiguous after\n the text segment. This is the oldest format for PDP11 executable\n programs and is the default for \\fBld\\fR on PDP11 Unix systems\n from the beginning through 2.11BSD.\n .IP \\fB\\-n\\fR 4\n .IX Item \"-n\"\n .PD 0\n .IP \\fB\\-\\-nmagic\\fR 4\n .IX Item \"--nmagic\"\n .PD\n Mark the output as \\f(CW\\*(C`NMAGIC\\*(C'\\fR (0410) in the \\fIa.out\\fR header to\n indicate that when the output file is executed, the text portion will\n-be read-only and shareable among all processes executing the same\n+be read\\-only and shareable among all processes executing the same\n file. This involves moving the data areas up to the first possible 8K\n byte page boundary following the end of the text. This option creates\n a \\fIpure executable\\fR format.\n .IP \\fB\\-z\\fR 4\n .IX Item \"-z\"\n .PD 0\n .IP \\fB\\-\\-imagic\\fR 4\n .IX Item \"--imagic\"\n .PD\n Mark the output as \\f(CW\\*(C`IMAGIC\\*(C'\\fR (0411) in the \\fIa.out\\fR header to\n indicate that when the output file is executed, the program text and\n data areas will be loaded into separate address spaces using the split\n instruction and data space feature of the memory management unit in\n larger models of the PDP11. This doubles the address space available\n-to the program. The text segment is again pure, write-protected, and\n+to the program. The text segment is again pure, write\\-protected, and\n shareable. The only difference in the output format between this\n option and the others, besides the magic number, is that both the text\n and data sections start at location 0. The \\fB\\-z\\fR option selected\n this format in 2.11BSD. This option creates a \\fIseparate\n executable\\fR format.\n .IP \\fB\\-\\-no\\-omagic\\fR 4\n .IX Item \"--no-omagic\"\n Equivalent to \\fB\\-\\-nmagic\\fR for pdp11\\-aout.\n .SH ENVIRONMENT\n .IX Header \"ENVIRONMENT\"\n You can change the behaviour of \\fBld\\fR with the environment variables\n \\&\\f(CW\\*(C`GNUTARGET\\*(C'\\fR,\n \\&\\f(CW\\*(C`LDEMULATION\\*(C'\\fR and \\f(CW\\*(C`COLLECT_NO_DEMANGLE\\*(C'\\fR.\n .PP\n-\\&\\f(CW\\*(C`GNUTARGET\\*(C'\\fR determines the input-file object format if you don't\n+\\&\\f(CW\\*(C`GNUTARGET\\*(C'\\fR determines the input\\-file object format if you don\\*(Aqt\n use \\fB\\-b\\fR (or its synonym \\fB\\-\\-format\\fR). Its value should be one\n of the BFD names for an input format. If there is no\n \\&\\f(CW\\*(C`GNUTARGET\\*(C'\\fR in the environment, \\fBld\\fR uses the natural format\n of the target. If \\f(CW\\*(C`GNUTARGET\\*(C'\\fR is set to \\f(CW\\*(C`default\\*(C'\\fR then BFD\n attempts to discover the input format by examining binary input files;\n this method often succeeds, but there are potential ambiguities, since\n there is no method of ensuring that the magic number used to specify\n-object-file formats is unique. However, the configuration procedure for\n+object\\-file formats is unique. However, the configuration procedure for\n BFD on each system places the conventional format for that system first\n-in the search-list, so ambiguities are resolved in favor of convention.\n+in the search\\-list, so ambiguities are resolved in favor of convention.\n .PP\n-\\&\\f(CW\\*(C`LDEMULATION\\*(C'\\fR determines the default emulation if you don't use the\n+\\&\\f(CW\\*(C`LDEMULATION\\*(C'\\fR determines the default emulation if you don\\*(Aqt use the\n \\&\\fB\\-m\\fR option. The emulation can affect various aspects of linker\n behaviour, particularly the default linker script. You can list the\n available emulations with the \\fB\\-\\-verbose\\fR or \\fB\\-V\\fR options. If\n the \\fB\\-m\\fR option is not used, and the \\f(CW\\*(C`LDEMULATION\\*(C'\\fR environment\n variable is not defined, the default emulation depends upon how the\n linker was configured.\n .PP\n@@ -3711,10 +3714,10 @@\n .SH COPYRIGHT\n .IX Header \"COPYRIGHT\"\n Copyright (c) 1991\\-2025 Free Software Foundation, Inc.\n .PP\n Permission is granted to copy, distribute and/or modify this document\n under the terms of the GNU Free Documentation License, Version 1.3\n or any later version published by the Free Software Foundation;\n-with no Invariant Sections, with no Front-Cover Texts, and with no\n-Back-Cover Texts. A copy of the license is included in the\n+with no Invariant Sections, with no Front\\-Cover Texts, and with no\n+Back\\-Cover Texts. A copy of the license is included in the\n section entitled \"GNU Free Documentation License\".\n"}]}]}]}]}, {"source1": "binutils-mingw-w64-x86-64_2.44-2+12_i386.deb", "source2": "binutils-mingw-w64-x86-64_2.44-2+12_i386.deb", "unified_diff": null, "details": [{"source1": "file list", "source2": "file list", "unified_diff": "@@ -1,3 +1,3 @@\n -rw-r--r-- 0 0 0 4 2024-10-26 18:03:55.000000 debian-binary\n -rw-r--r-- 0 0 0 2004 2024-10-26 18:03:55.000000 control.tar.xz\n--rw-r--r-- 0 0 0 3619892 2024-10-26 18:03:55.000000 data.tar.xz\n+-rw-r--r-- 0 0 0 3620036 2024-10-26 18:03:55.000000 data.tar.xz\n"}, {"source1": "control.tar.xz", "source2": "control.tar.xz", "unified_diff": null, "details": [{"source1": "control.tar", "source2": "control.tar", "unified_diff": null, "details": [{"source1": "./md5sums", "source2": "./md5sums", "unified_diff": null, "details": [{"source1": "./md5sums", "source2": "./md5sums", "comments": ["Files differ"], "unified_diff": null}]}]}]}, {"source1": "data.tar.xz", "source2": "data.tar.xz", "unified_diff": null, "details": [{"source1": "data.tar", "source2": "data.tar", "unified_diff": null, "details": [{"source1": "file list", "source2": "file list", "unified_diff": "@@ -35,15 +35,15 @@\n -rw-r--r-- 0 root (0) root (0) 7222 2024-10-26 18:03:55.000000 ./usr/share/man/man1/x86_64-w64-mingw32-ar.1.gz\n -rw-r--r-- 0 root (0) root (0) 29664 2024-10-26 18:03:55.000000 ./usr/share/man/man1/x86_64-w64-mingw32-as.1.gz\n -rw-r--r-- 0 root (0) root (0) 3765 2024-10-26 18:03:55.000000 ./usr/share/man/man1/x86_64-w64-mingw32-c++filt.1.gz\n -rw-r--r-- 0 root (0) root (0) 5720 2024-10-26 18:03:55.000000 ./usr/share/man/man1/x86_64-w64-mingw32-dlltool.1.gz\n -rw-r--r-- 0 root (0) root (0) 311 2024-10-26 18:03:55.000000 ./usr/share/man/man1/x86_64-w64-mingw32-dllwrap.1.gz\n -rw-r--r-- 0 root (0) root (0) 2557 2024-10-26 18:03:55.000000 ./usr/share/man/man1/x86_64-w64-mingw32-elfedit.1.gz\n -rw-r--r-- 0 root (0) root (0) 7943 2024-10-26 18:03:55.000000 ./usr/share/man/man1/x86_64-w64-mingw32-gprof.1.gz\n--rw-r--r-- 0 root (0) root (0) 47205 2024-10-26 18:03:55.000000 ./usr/share/man/man1/x86_64-w64-mingw32-ld.1.gz\n+-rw-r--r-- 0 root (0) root (0) 47328 2024-10-26 18:03:55.000000 ./usr/share/man/man1/x86_64-w64-mingw32-ld.1.gz\n -rw-r--r-- 0 root (0) root (0) 7084 2024-10-26 18:03:55.000000 ./usr/share/man/man1/x86_64-w64-mingw32-nm.1.gz\n -rw-r--r-- 0 root (0) root (0) 14200 2024-10-26 18:03:55.000000 ./usr/share/man/man1/x86_64-w64-mingw32-objcopy.1.gz\n -rw-r--r-- 0 root (0) root (0) 14901 2024-10-26 18:03:55.000000 ./usr/share/man/man1/x86_64-w64-mingw32-objdump.1.gz\n -rw-r--r-- 0 root (0) root (0) 2144 2024-10-26 18:03:55.000000 ./usr/share/man/man1/x86_64-w64-mingw32-ranlib.1.gz\n -rw-r--r-- 0 root (0) root (0) 8507 2024-10-26 18:03:55.000000 ./usr/share/man/man1/x86_64-w64-mingw32-readelf.1.gz\n -rw-r--r-- 0 root (0) root (0) 3015 2024-10-26 18:03:55.000000 ./usr/share/man/man1/x86_64-w64-mingw32-size.1.gz\n -rw-r--r-- 0 root (0) root (0) 3806 2024-10-26 18:03:55.000000 ./usr/share/man/man1/x86_64-w64-mingw32-strings.1.gz\n"}, {"source1": "./usr/share/man/man1/x86_64-w64-mingw32-ld.1.gz", "source2": "./usr/share/man/man1/x86_64-w64-mingw32-ld.1.gz", "unified_diff": null, "details": [{"source1": "x86_64-w64-mingw32-ld.1", "source2": "x86_64-w64-mingw32-ld.1", "unified_diff": "@@ -1,9 +1,9 @@\n .\\\" -*- mode: troff; coding: utf-8 -*-\n-.\\\" Automatically generated by Pod::Man 5.0102 (Pod::Simple 3.45)\n+.\\\" Automatically generated by Pod::Man v6.0.2 (Pod::Simple 3.45)\n .\\\"\n .\\\" Standard preamble:\n .\\\" ========================================================================\n .de Sp \\\" Vertical space (when we can't use .PP)\n .if t .sp .5v\n .if n .sp\n ..\n@@ -48,18 +48,21 @@\n . if !\\nF==2 \\{\\\n . nr % 0\n . nr F 2\n . \\}\n . \\}\n .\\}\n .rr rF\n+.\\\"\n+.\\\" Required to disable full justification in groff 1.23.0.\n+.if n .ds AD l\n .\\\" ========================================================================\n .\\\"\n .IX Title \"LD 1\"\n-.TH LD 1 2024-10-26 binutils-2.44 \"GNU Development Tools\"\n+.TH LD 1 2025-02-02 binutils-2.44 \"GNU Development Tools\"\n .\\\" For nroff, turn off justification. Always turn off hyphenation; it makes\n .\\\" way too many mistakes in technical documents.\n .if n .ad l\n .nh\n .SH NAME\n ld \\- The GNU linker\n .SH SYNOPSIS\n@@ -68,15 +71,15 @@\n .SH DESCRIPTION\n .IX Header \"DESCRIPTION\"\n \\&\\fBld\\fR combines a number of object and archive files, relocates\n their data and ties up symbol references. Usually the last step in\n compiling a program is to run \\fBld\\fR.\n .PP\n \\&\\fBld\\fR accepts Linker Command Language files written in\n-a superset of AT&T's Link Editor Command Language syntax,\n+a superset of AT&T\\*(Aqs Link Editor Command Language syntax,\n to provide explicit and total control over the linking process.\n .PP\n This man page does not describe the command language; see the\n \\&\\fBld\\fR entry in \\f(CW\\*(C`info\\*(C'\\fR for full details on the command\n language and on other aspects of the GNU linker.\n .PP\n This version of \\fBld\\fR uses the general purpose BFD libraries\n@@ -92,41 +95,41 @@\n (or, in some cases, to get an output file in spite of the error).\n .PP\n The GNU linker \\fBld\\fR is meant to cover a broad range of situations,\n and to be as compatible as possible with other linkers. As a result,\n you have many choices to control its behavior.\n .SH OPTIONS\n .IX Header \"OPTIONS\"\n-The linker supports a plethora of command-line options, but in actual\n+The linker supports a plethora of command\\-line options, but in actual\n practice few of them are used in any particular context.\n For instance, a frequent use of \\fBld\\fR is to link standard Unix\n object files on a standard, supported Unix system. On such a system, to\n link a file \\f(CW\\*(C`hello.o\\*(C'\\fR:\n .PP\n .Vb 1\n \\& ld \\-o /lib/crt0.o hello.o \\-lc\n .Ve\n .PP\n This tells \\fBld\\fR to produce a file called \\fIoutput\\fR as the\n result of linking the file \\f(CW\\*(C`/lib/crt0.o\\*(C'\\fR with \\f(CW\\*(C`hello.o\\*(C'\\fR and\n the library \\f(CW\\*(C`libc.a\\*(C'\\fR, which will come from the standard search\n directories. (See the discussion of the \\fB\\-l\\fR option below.)\n .PP\n-Some of the command-line options to \\fBld\\fR may be specified at any\n+Some of the command\\-line options to \\fBld\\fR may be specified at any\n point in the command line. However, options which refer to files, such\n as \\fB\\-l\\fR or \\fB\\-T\\fR, cause the file to be read at the point at\n which the option appears in the command line, relative to the object\n-files and other file options. Repeating non-file options with a\n+files and other file options. Repeating non\\-file options with a\n different argument will either have no further effect, or override prior\n occurrences (those further to the left on the command line) of that\n option. Options which may be meaningfully specified more than once are\n noted in the descriptions below.\n .PP\n-Non-option arguments are object files or archives which are to be linked\n-together. They may follow, precede, or be mixed in with command-line\n+Non\\-option arguments are object files or archives which are to be linked\n+together. They may follow, precede, or be mixed in with command\\-line\n options, except that an object file argument may not be placed between\n an option and its argument.\n .PP\n Usually the linker is invoked with at least one object file, but you can\n specify other forms of binary input files using \\fB\\-l\\fR, \\fB\\-R\\fR,\n and the script command language. If \\fIno\\fR binary input files at all\n are specified, the linker does not produce any output, and issues the\n@@ -148,53 +151,53 @@\n option arguments must either follow the option letter without intervening\n whitespace, or be given as separate arguments immediately following the\n option that requires them.\n .PP\n For options whose names are multiple letters, either one dash or two can\n precede the option name; for example, \\fB\\-trace\\-symbol\\fR and\n \\&\\fB\\-\\-trace\\-symbol\\fR are equivalent. Note\\-\\-\\-there is one exception to\n-this rule. Multiple letter options that start with a lower case 'o' can\n+this rule. Multiple letter options that start with a lower case \\*(Aqo\\*(Aq can\n only be preceded by two dashes. This is to reduce confusion with the\n \\&\\fB\\-o\\fR option. So for example \\fB\\-omagic\\fR sets the output file\n name to \\fBmagic\\fR whereas \\fB\\-\\-omagic\\fR sets the NMAGIC flag on the\n output.\n .PP\n-Arguments to multiple-letter options must either be separated from the\n+Arguments to multiple\\-letter options must either be separated from the\n option name by an equals sign, or be given as separate arguments\n immediately following the option that requires them. For example,\n \\&\\fB\\-\\-trace\\-symbol foo\\fR and \\fB\\-\\-trace\\-symbol=foo\\fR are equivalent.\n-Unique abbreviations of the names of multiple-letter options are\n+Unique abbreviations of the names of multiple\\-letter options are\n accepted.\n .PP\n Note\\-\\-\\-if the linker is being invoked indirectly, via a compiler driver\n-(e.g. \\fBgcc\\fR) then all the linker command-line options should be\n+(e.g. \\fBgcc\\fR) then all the linker command\\-line options should be\n prefixed by \\fB\\-Wl,\\fR (or whatever is appropriate for the particular\n compiler driver) like this:\n .PP\n .Vb 1\n \\& gcc \\-Wl,\\-\\-start\\-group foo.o bar.o \\-Wl,\\-\\-end\\-group\n .Ve\n .PP\n This is important, because otherwise the compiler driver program may\n silently drop the linker options, resulting in a bad link. Confusion\n may also arise when passing options that require values through a\n driver, as the use of a space between option and argument acts as\n a separator, and causes the driver to pass only the option to the linker\n and the argument to the compiler. In this case, it is simplest to use\n-the joined forms of both single\\- and multiple-letter options, such as:\n+the joined forms of both single\\- and multiple\\-letter options, such as:\n .PP\n .Vb 1\n \\& gcc foo.o bar.o \\-Wl,\\-eENTRY \\-Wl,\\-Map=a.map\n .Ve\n .PP\n-Here is a table of the generic command-line switches accepted by the GNU\n+Here is a table of the generic command\\-line switches accepted by the GNU\n linker:\n .IP \\fB@\\fR\\fIfile\\fR 4\n .IX Item \"@file\"\n-Read command-line options from \\fIfile\\fR. The options read are\n+Read command\\-line options from \\fIfile\\fR. The options read are\n inserted in place of the original @\\fIfile\\fR option. If \\fIfile\\fR\n does not exist, or cannot be read, then the option will be treated\n literally, and not removed.\n .Sp\n Options in \\fIfile\\fR are separated by whitespace. A whitespace\n character may be included in an option by surrounding the entire\n option in either single or double quotes. Any character (including a\n@@ -212,57 +215,57 @@\n .IX Item \"--audit AUDITLIB\"\n Adds \\fIAUDITLIB\\fR to the \\f(CW\\*(C`DT_AUDIT\\*(C'\\fR entry of the dynamic section.\n \\&\\fIAUDITLIB\\fR is not checked for existence, nor will it use the DT_SONAME\n specified in the library. If specified multiple times \\f(CW\\*(C`DT_AUDIT\\*(C'\\fR\n will contain a colon separated list of audit interfaces to use. If the linker\n finds an object with an audit entry while searching for shared libraries,\n it will add a corresponding \\f(CW\\*(C`DT_DEPAUDIT\\*(C'\\fR entry in the output file.\n-This option is only meaningful on ELF platforms supporting the rtld-audit\n+This option is only meaningful on ELF platforms supporting the rtld\\-audit\n interface.\n-.IP \"\\fB\\-b\\fR \\fIinput-format\\fR\" 4\n+.IP \"\\fB\\-b\\fR \\fIinput\\-format\\fR\" 4\n .IX Item \"-b input-format\"\n .PD 0\n-.IP \\fB\\-\\-format=\\fR\\fIinput-format\\fR 4\n+.IP \\fB\\-\\-format=\\fR\\fIinput\\-format\\fR 4\n .IX Item \"--format=input-format\"\n .PD\n \\&\\fBld\\fR may be configured to support more than one kind of object\n file. If your \\fBld\\fR is configured this way, you can use the\n \\&\\fB\\-b\\fR option to specify the binary format for input object files\n that follow this option on the command line. Even when \\fBld\\fR is\n-configured to support alternative object formats, you don't usually need\n+configured to support alternative object formats, you don\\*(Aqt usually need\n to specify this, as \\fBld\\fR should be configured to expect as a\n default input format the most usual format on each machine.\n-\\&\\fIinput-format\\fR is a text string, the name of a particular format\n+\\&\\fIinput\\-format\\fR is a text string, the name of a particular format\n supported by the BFD libraries. (You can list the available binary\n formats with \\fBobjdump \\-i\\fR.)\n .Sp\n You may want to use this option if you are linking files with an unusual\n binary format. You can also use \\fB\\-b\\fR to switch formats explicitly (when\n linking object files of different formats), by including\n-\\&\\fB\\-b\\fR \\fIinput-format\\fR before each group of object files in a\n+\\&\\fB\\-b\\fR \\fIinput\\-format\\fR before each group of object files in a\n particular format.\n .Sp\n The default format is taken from the environment variable\n \\&\\f(CW\\*(C`GNUTARGET\\*(C'\\fR.\n .Sp\n You can also define the input format from a script, using the command\n \\&\\f(CW\\*(C`TARGET\\*(C'\\fR;\n-.IP \"\\fB\\-c\\fR \\fIMRI-commandfile\\fR\" 4\n+.IP \"\\fB\\-c\\fR \\fIMRI\\-commandfile\\fR\" 4\n .IX Item \"-c MRI-commandfile\"\n .PD 0\n-.IP \\fB\\-\\-mri\\-script=\\fR\\fIMRI-commandfile\\fR 4\n+.IP \\fB\\-\\-mri\\-script=\\fR\\fIMRI\\-commandfile\\fR 4\n .IX Item \"--mri-script=MRI-commandfile\"\n .PD\n For compatibility with linkers produced by MRI, \\fBld\\fR accepts script\n files written in an alternate, restricted command language, described in\n the MRI Compatible Script Files section of GNU ld documentation.\n Introduce MRI script files with\n the option \\fB\\-c\\fR; use the \\fB\\-T\\fR option to run linker\n-scripts written in the general-purpose \\fBld\\fR scripting language.\n-If \\fIMRI-cmdfile\\fR does not exist, \\fBld\\fR looks for it in the directories\n+scripts written in the general\\-purpose \\fBld\\fR scripting language.\n+If \\fIMRI\\-cmdfile\\fR does not exist, \\fBld\\fR looks for it in the directories\n specified by any \\fB\\-L\\fR options.\n .IP \\fB\\-d\\fR 4\n .IX Item \"-d\"\n .PD 0\n .IP \\fB\\-dc\\fR 4\n .IX Item \"-dc\"\n .IP \\fB\\-dp\\fR 4\n@@ -278,15 +281,15 @@\n .IP \"\\fB\\-P\\fR \\fIAUDITLIB\\fR\" 4\n .IX Item \"-P AUDITLIB\"\n .PD\n Adds \\fIAUDITLIB\\fR to the \\f(CW\\*(C`DT_DEPAUDIT\\*(C'\\fR entry of the dynamic section.\n \\&\\fIAUDITLIB\\fR is not checked for existence, nor will it use the DT_SONAME\n specified in the library. If specified multiple times \\f(CW\\*(C`DT_DEPAUDIT\\*(C'\\fR\n will contain a colon separated list of audit interfaces to use. This\n-option is only meaningful on ELF platforms supporting the rtld-audit interface.\n+option is only meaningful on ELF platforms supporting the rtld\\-audit interface.\n The \\-P option is provided for Solaris compatibility.\n .IP \\fB\\-\\-enable\\-linker\\-version\\fR 4\n .IX Item \"--enable-linker-version\"\n Enables the \\f(CW\\*(C`LINKER_VERSION\\*(C'\\fR linker script directive, described\n in \\fBOutput Section Data\\fR. If this directive is used in a linker\n script and this option has been enabled then a string containing the\n linker version will be inserted at the current point.\n@@ -300,15 +303,15 @@\n does not insert a version string. This is the default.\n .IP \\fB\\-\\-enable\\-non\\-contiguous\\-regions\\fR 4\n .IX Item \"--enable-non-contiguous-regions\"\n This option avoids generating an error if an input section does not\n fit a matching output section. The linker tries to allocate the input\n section to subseque nt matching output sections, and generates an\n error only if no output section is large enough. This is useful when\n-several non-contiguous memory regions are available and the input\n+several non\\-contiguous memory regions are available and the input\n section does not require a particular one. The order in which input\n sections are evaluated does not change, for instance:\n .Sp\n .Vb 10\n \\& MEMORY {\n \\& MEM1 (rwx) : ORIGIN = 0x1000, LENGTH = 0x14\n \\& MEM2 (rwx) : ORIGIN = 0x1000, LENGTH = 0x40\n@@ -364,16 +367,16 @@\n .IX Item \"--exclude-modules-for-implib module,module,...\"\n Specifies a list of object files or archive members, from which symbols\n should not be automatically exported, but which should be copied wholesale\n into the import library being generated during the link. The module names\n may be delimited by commas or colons, and must match exactly the filenames\n used by \\fBld\\fR to open the files; for archive members, this is simply\n the member name, but for object files the name listed must include and\n-match precisely any path used to specify the input file on the linker's\n-command-line. This option is available only for the i386 PE targeted port\n+match precisely any path used to specify the input file on the linker\\*(Aqs\n+command\\-line. This option is available only for the i386 PE targeted port\n of the linker. Symbols explicitly listed in a .def file are still exported,\n regardless of this option.\n .IP \\fB\\-E\\fR 4\n .IX Item \"-E\"\n .PD 0\n .IP \\fB\\-\\-export\\-dynamic\\fR 4\n .IX Item \"--export-dynamic\"\n@@ -404,28 +407,28 @@\n the description of \\fB\\-\\-export\\-all\\-symbols\\fR below.\n .IP \\fB\\-\\-export\\-dynamic\\-symbol=\\fR\\fIglob\\fR 4\n .IX Item \"--export-dynamic-symbol=glob\"\n When creating a dynamically linked executable, symbols matching\n \\&\\fIglob\\fR will be added to the dynamic symbol table. When creating a\n shared library, references to symbols matching \\fIglob\\fR will not be\n bound to the definitions within the shared library. This option is a\n-no-op when creating a shared library and \\fB\\-Bsymbolic\\fR or\n+no\\-op when creating a shared library and \\fB\\-Bsymbolic\\fR or\n \\&\\fB\\-\\-dynamic\\-list\\fR are not specified. This option is only meaningful\n on ELF platforms which support shared libraries.\n .IP \\fB\\-\\-export\\-dynamic\\-symbol\\-list=\\fR\\fIfile\\fR 4\n .IX Item \"--export-dynamic-symbol-list=file\"\n Specify a \\fB\\-\\-export\\-dynamic\\-symbol\\fR for each pattern in the file.\n The format of the file is the same as the version node without\n scope and node name. See \\fBVERSION\\fR for more information.\n .IP \\fB\\-EB\\fR 4\n .IX Item \"-EB\"\n-Link big-endian objects. This affects the default output format.\n+Link big\\-endian objects. This affects the default output format.\n .IP \\fB\\-EL\\fR 4\n .IX Item \"-EL\"\n-Link little-endian objects. This affects the default output format.\n+Link little\\-endian objects. This affects the default output format.\n .IP \"\\fB\\-f\\fR \\fIname\\fR\" 4\n .IX Item \"-f name\"\n .PD 0\n .IP \\fB\\-\\-auxiliary=\\fR\\fIname\\fR 4\n .IX Item \"--auxiliary=name\"\n .PD\n When creating an ELF shared object, set the internal DT_AUXILIARY field\n@@ -437,15 +440,15 @@\n run the program, the dynamic linker will see the DT_AUXILIARY field. If\n the dynamic linker resolves any symbols from the filter object, it will\n first check whether there is a definition in the shared object\n \\&\\fIname\\fR. If there is one, it will be used instead of the definition\n in the filter object. The shared object \\fIname\\fR need not exist.\n Thus the shared object \\fIname\\fR may be used to provide an alternative\n implementation of certain functions, perhaps for debugging or for\n-machine-specific performance.\n+machine\\-specific performance.\n .Sp\n This option may be specified more than once. The DT_AUXILIARY entries\n will be created in the order in which they appear on the command line.\n .IP \"\\fB\\-F\\fR \\fIname\\fR\" 4\n .IX Item \"-F name\"\n .PD 0\n .IP \\fB\\-\\-filter=\\fR\\fIname\\fR 4\n@@ -461,15 +464,15 @@\n dynamic linker will resolve symbols according to the symbol table of the\n filter object as usual, but it will actually link to the definitions\n found in the shared object \\fIname\\fR. Thus the filter object can be\n used to select a subset of the symbols provided by the object\n \\&\\fIname\\fR.\n .Sp\n Some older linkers used the \\fB\\-F\\fR option throughout a compilation\n-toolchain for specifying object-file format for both input and output\n+toolchain for specifying object\\-file format for both input and output\n object files.\n The GNU linker uses other mechanisms for this purpose: the\n \\&\\fB\\-b\\fR, \\fB\\-\\-format\\fR, \\fB\\-\\-oformat\\fR options, the\n \\&\\f(CW\\*(C`TARGET\\*(C'\\fR command in linker scripts, and the \\f(CW\\*(C`GNUTARGET\\*(C'\\fR\n environment variable.\n The GNU linker will ignore the \\fB\\-F\\fR option when not\n creating an ELF shared object.\n@@ -698,15 +701,15 @@\n .Sp\n .Vb 1\n \\& Removed property 0xc0000002 to merge foo.o (0x1) and bar.o (not found)\n .Ve\n .Sp\n This indicates that property 0xc0000002 is removed from output when\n merging properties in \\fIfoo.o\\fR, whose property 0xc0000002 value\n-is 0x1, and \\fIbar.o\\fR, which doesn't have property 0xc0000002.\n+is 0x1, and \\fIbar.o\\fR, which doesn\\*(Aqt have property 0xc0000002.\n .Sp\n .Vb 1\n \\& Updated property 0xc0010001 (0x1) to merge foo.o (0x1) and bar.o (0x1)\n .Ve\n .Sp\n This indicates that property 0xc0010001 value is updated to 0x1 in output\n when merging properties in \\fIfoo.o\\fR, whose 0xc0010001 property value\n@@ -754,24 +757,24 @@\n .IP \\fB\\-N\\fR 4\n .IX Item \"-N\"\n .PD 0\n .IP \\fB\\-\\-omagic\\fR 4\n .IX Item \"--omagic\"\n .PD\n Set the text and data sections to be readable and writable. Also, do\n-not page-align the data segment, and disable linking against shared\n+not page\\-align the data segment, and disable linking against shared\n libraries. If the output format supports Unix style magic numbers,\n mark the output as \\f(CW\\*(C`OMAGIC\\*(C'\\fR. Note: Although a writable text section\n-is allowed for PE-COFF targets, it does not conform to the format\n+is allowed for PE\\-COFF targets, it does not conform to the format\n specification published by Microsoft.\n .IP \\fB\\-\\-no\\-omagic\\fR 4\n .IX Item \"--no-omagic\"\n This option negates most of the effects of the \\fB\\-N\\fR option. It\n-sets the text section to be read-only, and forces the data segment to\n-be page-aligned. Note \\- this option does not enable linking against\n+sets the text section to be read\\-only, and forces the data segment to\n+be page\\-aligned. Note \\- this option does not enable linking against\n shared libraries. Use \\fB\\-Bdynamic\\fR for this.\n .IP \"\\fB\\-o\\fR \\fIoutput\\fR\" 4\n .IX Item \"-o output\"\n .PD 0\n .IP \\fB\\-\\-output=\\fR\\fIoutput\\fR 4\n .IX Item \"--output=output\"\n .PD\n@@ -800,29 +803,29 @@\n files (the first one in the \\fBOBJS\\fR list) whereas the second\n version of the linker command will generate an error message and not\n delete anything.\n .IP \\fB\\-\\-dependency\\-file=\\fR\\fIdepfile\\fR 4\n .IX Item \"--dependency-file=depfile\"\n Write a \\fIdependency file\\fR to \\fIdepfile\\fR. This file contains a rule\n suitable for \\f(CW\\*(C`make\\*(C'\\fR describing the output file and all the input files\n-that were read to produce it. The output is similar to the compiler's\n-output with \\fB\\-M \\-MP\\fR. Note that there is no option like the compiler's \\fB\\-MM\\fR,\n-to exclude \"system files\" (which is not a well-specified concept in the\n+that were read to produce it. The output is similar to the compiler\\*(Aqs\n+output with \\fB\\-M \\-MP\\fR. Note that there is no option like the compiler\\*(Aqs \\fB\\-MM\\fR,\n+to exclude \"system files\" (which is not a well\\-specified concept in the\n linker, unlike \"system headers\" in the compiler). So the output from\n \\&\\fB\\-\\-dependency\\-file\\fR is always specific to the exact state of the\n installation where it was produced, and should not be copied into\n distributed makefiles without careful editing.\n .IP \"\\fB\\-O\\fR \\fIlevel\\fR\" 4\n .IX Item \"-O level\"\n If \\fIlevel\\fR is a numeric values greater than zero \\fBld\\fR optimizes\n the output. This might take significantly longer and therefore probably\n should only be enabled for the final binary. At the moment this\n option only affects ELF shared library generation. Future releases of\n the linker may make more use of this option. Also currently there is\n-no difference in the linker's behaviour for different non-zero values\n+no difference in the linker\\*(Aqs behaviour for different non\\-zero values\n of this option. Again this may change with future releases.\n .IP \"\\fB\\-plugin\\fR \\fIname\\fR\" 4\n .IX Item \"-plugin name\"\n Involve a plugin in the linking process. The \\fIname\\fR parameter is\n the absolute filename of the plugin. Usually this parameter is\n automatically added by the compiler, when using link time\n optimization, but users can also add their own plugins if they so\n@@ -844,15 +847,15 @@\n The option which are covered are: \\fB\\-Bdynamic\\fR, \\fB\\-Bstatic\\fR,\n \\&\\fB\\-dn\\fR, \\fB\\-dy\\fR, \\fB\\-call_shared\\fR, \\fB\\-non_shared\\fR,\n \\&\\fB\\-static\\fR, \\fB\\-N\\fR, \\fB\\-n\\fR, \\fB\\-\\-whole\\-archive\\fR,\n \\&\\fB\\-\\-no\\-whole\\-archive\\fR, \\fB\\-r\\fR, \\fB\\-Ur\\fR,\n \\&\\fB\\-\\-copy\\-dt\\-needed\\-entries\\fR, \\fB\\-\\-no\\-copy\\-dt\\-needed\\-entries\\fR,\n \\&\\fB\\-\\-as\\-needed\\fR, \\fB\\-\\-no\\-as\\-needed\\fR, and \\fB\\-a\\fR.\n .Sp\n-One target for this option are specifications for \\fIpkg-config\\fR. When\n+One target for this option are specifications for \\fIpkg\\-config\\fR. When\n used with the \\fB\\-\\-libs\\fR option all possibly needed libraries are\n listed and then possibly linked with all the time. It is better to return\n something as follows:\n .Sp\n .Vb 1\n \\& \\-Wl,\\-\\-push\\-state,\\-\\-as\\-needed \\-libone \\-libtwo \\-Wl,\\-\\-pop\\-state\n .Ve\n@@ -881,31 +884,31 @@\n .PD 0\n .IP \\fB\\-\\-relocatable\\fR 4\n .IX Item \"--relocatable\"\n .PD\n Generate relocatable output\\-\\-\\-i.e., generate an output file that can in\n turn serve as input to \\fBld\\fR. This is often called \\fIpartial\n linking\\fR. As a side effect, in environments that support standard Unix\n-magic numbers, this option also sets the output file's magic number to\n+magic numbers, this option also sets the output file\\*(Aqs magic number to\n \\&\\f(CW\\*(C`OMAGIC\\*(C'\\fR.\n If this option is not specified, an absolute file is produced. When\n linking C++ programs, this option \\fIwill not\\fR resolve references to\n constructors; to do that, use \\fB\\-Ur\\fR.\n .Sp\n When an input file does not have the same format as the output file,\n partial linking is only supported if that input file does not contain any\n relocations. Different output formats can have further restrictions; for\n example some \\f(CW\\*(C`a.out\\*(C'\\fR\\-based formats do not support partial linking\n with input files in other formats at all.\n .Sp\n When the relocatable output contains both contents which require\n-link-time optimization (LTO) and contents which don't require LTO,\n+link\\-time optimization (LTO) and contents which don\\*(Aqt require LTO,\n a .gnu_object_only section will be created to contain a relocatable\n object file, as if \\fB\\-r\\fR is applied to all relocatable inputs\n-which don't require LTO. When processing a relocatable input with\n+which don\\*(Aqt require LTO. When processing a relocatable input with\n a .gnu_object_only section, the linker will extract the .gnu_object_only\n section as a separate input.\n .Sp\n Note that since \\fB\\-r\\fR groups some sections from different input files\n together, there may be negative impacts on code size and locality in\n final executable or shared library.\n .Sp\n@@ -926,20 +929,20 @@\n the \\fB\\-rpath\\fR option.\n .IP \\fB\\-\\-rosegment\\fR 4\n .IX Item \"--rosegment\"\n .PD 0\n .IP \\fB\\-\\-no\\-rosegment\\fR 4\n .IX Item \"--no-rosegment\"\n .PD\n-Attempt to ensure that only a single read-only, non-code segment is\n+Attempt to ensure that only a single read\\-only, non\\-code segment is\n created. Only useful when used in conjunction with the \\fB\\-z\n-separate-code\\fR option. The resulting binaries should be smaller than\n-if \\fB\\-z separate-code\\fR is used on its own. Without this option,\n-or if \\fB\\-\\-no\\-rosegment\\fR is specified, the \\fB\\-z separate-code\\fR\n-option will create two read-only segments, one before the code segment\n+separate\\-code\\fR option. The resulting binaries should be smaller than\n+if \\fB\\-z separate\\-code\\fR is used on its own. Without this option,\n+or if \\fB\\-\\-no\\-rosegment\\fR is specified, the \\fB\\-z separate\\-code\\fR\n+option will create two read\\-only segments, one before the code segment\n and one after it.\n .Sp\n The name of the options are misleading, but they have been chosen in\n order for the linker to be compatible with the LLD and GOLD linkers.\n .Sp\n Thse options are only supported by ELF targets.\n .IP \\fB\\-s\\fR 4\n@@ -981,28 +984,28 @@\n .IP \"\\fB\\-T\\fR \\fIscriptfile\\fR\" 4\n .IX Item \"-T scriptfile\"\n .PD 0\n .IP \\fB\\-\\-script=\\fR\\fIscriptfile\\fR 4\n .IX Item \"--script=scriptfile\"\n .PD\n Use \\fIscriptfile\\fR as the linker script. This script replaces\n-\\&\\fBld\\fR's default linker script (rather than adding to it),\n+\\&\\fBld\\fR\\*(Aqs default linker script (rather than adding to it),\n unless the script contains \\f(CW\\*(C`INSERT\\*(C'\\fR, so \\fIcommandfile\\fR must\n specify everything necessary to describe the output file.\n .Sp\n If \\fIscriptfile\\fR does not exist in the current directory, \\f(CW\\*(C`ld\\*(C'\\fR\n looks for it in the directories specified by any preceding \\fB\\-L\\fR\n options.\n .Sp\n Command line options that appear before the \\fB\\-T\\fR option can\n affect the script, but command line options that appear after it do\n not.\n .Sp\n Multiple \\fB\\-T\\fR options will accumulate if they are augmenting the\n-current script, otherwise the last, non-augmenting, \\fB\\-T\\fR option\n+current script, otherwise the last, non\\-augmenting, \\fB\\-T\\fR option\n will be used.\n .Sp\n There are other ways of specifying linker scripts. See\n .IP \"\\fB\\-dT\\fR \\fIscriptfile\\fR\" 4\n .IX Item \"-dT scriptfile\"\n .PD 0\n .IP \\fB\\-\\-default\\-script=\\fR\\fIscriptfile\\fR 4\n@@ -1119,28 +1122,28 @@\n .IP \\fB\\-X\\fR 4\n .IX Item \"-X\"\n .PD 0\n .IP \\fB\\-\\-discard\\-locals\\fR 4\n .IX Item \"--discard-locals\"\n .PD\n Delete all temporary local symbols. (These symbols start with\n-system-specific local label prefixes, typically \\fB.L\\fR for ELF systems\n+system\\-specific local label prefixes, typically \\fB.L\\fR for ELF systems\n or \\fBL\\fR for traditional a.out systems.)\n .IP \"\\fB\\-y\\fR \\fIsymbol\\fR\" 4\n .IX Item \"-y symbol\"\n .PD 0\n .IP \\fB\\-\\-trace\\-symbol=\\fR\\fIsymbol\\fR 4\n .IX Item \"--trace-symbol=symbol\"\n .PD\n Print the name of each linked file in which \\fIsymbol\\fR appears. This\n option may be given any number of times. On many systems it is necessary\n to prepend an underscore.\n .Sp\n This option is useful when you have an undefined symbol in your link but\n-don't know where the reference is coming from.\n+don\\*(Aqt know where the reference is coming from.\n .IP \"\\fB\\-Y\\fR \\fIpath\\fR\" 4\n .IX Item \"-Y path\"\n Add \\fIpath\\fR to the default library search path. This option exists\n for Solaris compatibility.\n .IP \"\\fB\\-z\\fR \\fIkeyword\\fR\" 4\n .IX Item \"-z keyword\"\n The recognized keywords are:\n@@ -1201,26 +1204,26 @@\n .IX Item \"common-page-size=value\"\n Set the page size most commonly used to \\fIvalue\\fR. Memory image\n layout will be optimized to minimize memory pages if the system is\n using pages of this size.\n .IP \\fBdefs\\fR 4\n .IX Item \"defs\"\n Report unresolved symbol references from regular object files. This\n-is done even if the linker is creating a non-symbolic shared library.\n+is done even if the linker is creating a non\\-symbolic shared library.\n This option is the inverse of \\fB\\-z undefs\\fR.\n-.IP \\fBdynamic-undefined-weak\\fR 4\n+.IP \\fBdynamic\\-undefined\\-weak\\fR 4\n .IX Item \"dynamic-undefined-weak\"\n .PD 0\n-.IP \\fBnodynamic-undefined-weak\\fR 4\n+.IP \\fBnodynamic\\-undefined\\-weak\\fR 4\n .IX Item \"nodynamic-undefined-weak\"\n .PD\n Make undefined weak symbols dynamic when building a dynamic object,\n if they are referenced from a regular object file and not forced local\n by symbol visibility or versioning. Do not make them dynamic if\n-\\&\\fBnodynamic-undefined-weak\\fR. If neither option is given, a target\n+\\&\\fBnodynamic\\-undefined\\-weak\\fR. If neither option is given, a target\n may default to either option being in force, or make some other\n selection of undefined weak symbols dynamic. Not all targets support\n these options.\n .IP \\fBexecstack\\fR 4\n .IX Item \"execstack\"\n Marks the object as requiring executable stack.\n .IP \\fBglobal\\fR 4\n@@ -1230,38 +1233,38 @@\n of subsequently loaded libraries.\n .IP \\fBglobalaudit\\fR 4\n .IX Item \"globalaudit\"\n This option is only meaningful when building a dynamic executable.\n This option marks the executable as requiring global auditing by\n setting the \\f(CW\\*(C`DF_1_GLOBAUDIT\\*(C'\\fR bit in the \\f(CW\\*(C`DT_FLAGS_1\\*(C'\\fR dynamic\n tag. Global auditing requires that any auditing library defined via\n-the \\fB\\-\\-depaudit\\fR or \\fB\\-P\\fR command-line options be run for\n+the \\fB\\-\\-depaudit\\fR or \\fB\\-P\\fR command\\-line options be run for\n all dynamic objects loaded by the application.\n .IP \\fBibtplt\\fR 4\n .IX Item \"ibtplt\"\n Generate Intel Indirect Branch Tracking (IBT) enabled PLT entries.\n Supported for Linux/i386 and Linux/x86_64.\n .IP \\fBibt\\fR 4\n .IX Item \"ibt\"\n Generate GNU_PROPERTY_X86_FEATURE_1_IBT in .note.gnu.property section\n to indicate compatibility with IBT. This also implies \\fBibtplt\\fR.\n Supported for Linux/i386 and Linux/x86_64.\n-.IP \\fBindirect-extern-access\\fR 4\n+.IP \\fBindirect\\-extern\\-access\\fR 4\n .IX Item \"indirect-extern-access\"\n .PD 0\n-.IP \\fBnoindirect-extern-access\\fR 4\n+.IP \\fBnoindirect\\-extern\\-access\\fR 4\n .IX Item \"noindirect-extern-access\"\n .PD\n Generate GNU_PROPERTY_1_NEEDED_INDIRECT_EXTERN_ACCESS in\n \\&.note.gnu.property section to indicate that object file requires\n canonical function pointers and cannot be used with copy relocation.\n-This option also implies \\fBnoextern-protected-data\\fR and\n+This option also implies \\fBnoextern\\-protected\\-data\\fR and\n \\&\\fBnocopyreloc\\fR. Supported for i386 and x86\\-64.\n .Sp\n-\\&\\fBnoindirect-extern-access\\fR removes\n+\\&\\fBnoindirect\\-extern\\-access\\fR removes\n GNU_PROPERTY_1_NEEDED_INDIRECT_EXTERN_ACCESS from .note.gnu.property\n section.\n .IP \\fBinitfirst\\fR 4\n .IX Item \"initfirst\"\n This option is only meaningful when building a shared object.\n It marks the object so that its runtime initialization will occur\n before the runtime initialization of any other objects brought into\n@@ -1346,72 +1349,72 @@\n .IX Item \"lazy\"\n When generating an executable or shared library, mark it to tell the\n dynamic linker to defer function call resolution to the point when\n the function is called (lazy binding), rather than at load time.\n Lazy binding is the default.\n .IP \\fBloadfltr\\fR 4\n .IX Item \"loadfltr\"\n-Specify that the object's filters be processed immediately at runtime.\n+Specify that the object\\*(Aqs filters be processed immediately at runtime.\n .IP \\fBmax\\-page\\-size=\\fR\\fIvalue\\fR 4\n .IX Item \"max-page-size=value\"\n Set the maximum memory page size supported to \\fIvalue\\fR.\n-.IP \\fBmark-plt\\fR 4\n+.IP \\fBmark\\-plt\\fR 4\n .IX Item \"mark-plt\"\n .PD 0\n-.IP \\fBnomark-plt\\fR 4\n+.IP \\fBnomark\\-plt\\fR 4\n .IX Item \"nomark-plt\"\n .PD\n Mark PLT entries with dynamic tags, DT_X86_64_PLT, DT_X86_64_PLTSZ and\n-DT_X86_64_PLTENT. Since this option stores a non-zero value in the\n+DT_X86_64_PLTENT. Since this option stores a non\\-zero value in the\n r_addend field of R_X86_64_JUMP_SLOT relocations, the resulting\n executables and shared libraries are incompatible with dynamic linkers,\n such as those in older versions of glibc without the change to ignore\n r_addend in R_X86_64_GLOB_DAT and R_X86_64_JUMP_SLOT relocations, which\n-don't ignore the r_addend field of R_X86_64_JUMP_SLOT relocations.\n+don\\*(Aqt ignore the r_addend field of R_X86_64_JUMP_SLOT relocations.\n Supported for x86_64.\n .IP \\fBmuldefs\\fR 4\n .IX Item \"muldefs\"\n Allow multiple definitions.\n .IP \\fBnocopyreloc\\fR 4\n .IX Item \"nocopyreloc\"\n Disable linker generated .dynbss variables used in place of variables\n defined in shared libraries. May result in dynamic text relocations.\n .IP \\fBnodefaultlib\\fR 4\n .IX Item \"nodefaultlib\"\n Specify that the dynamic loader search for dependencies of this object\n should ignore any default library search paths.\n .IP \\fBnodelete\\fR 4\n .IX Item \"nodelete\"\n-Specify that the object shouldn't be unloaded at runtime.\n+Specify that the object shouldn\\*(Aqt be unloaded at runtime.\n .IP \\fBnodlopen\\fR 4\n .IX Item \"nodlopen\"\n Specify that the object is not available to \\f(CW\\*(C`dlopen\\*(C'\\fR.\n .IP \\fBnodump\\fR 4\n .IX Item \"nodump\"\n Specify that the object can not be dumped by \\f(CW\\*(C`dldump\\*(C'\\fR.\n .IP \\fBnoexecstack\\fR 4\n .IX Item \"noexecstack\"\n Marks the object as not requiring executable stack.\n-.IP \\fBnoextern-protected-data\\fR 4\n+.IP \\fBnoextern\\-protected\\-data\\fR 4\n .IX Item \"noextern-protected-data\"\n-Don't treat protected data symbols as external when building a shared\n+Don\\*(Aqt treat protected data symbols as external when building a shared\n library. This option overrides the linker backend default. It can be\n used to work around incorrect relocations against protected data symbols\n generated by compiler. Updates on protected data symbols by another\n-module aren't visible to the resulting shared library. Supported for\n+module aren\\*(Aqt visible to the resulting shared library. Supported for\n i386 and x86\\-64.\n-.IP \\fBnoreloc-overflow\\fR 4\n+.IP \\fBnoreloc\\-overflow\\fR 4\n .IX Item \"noreloc-overflow\"\n Disable relocation overflow check. This can be used to disable\n relocation overflow check if there will be no dynamic relocation\n-overflow at run-time. Supported for x86_64.\n-.IP \\fBmemory-seal\\fR 4\n+overflow at run\\-time. Supported for x86_64.\n+.IP \\fBmemory\\-seal\\fR 4\n .IX Item \"memory-seal\"\n .PD 0\n-.IP \\fBnomemory-seal\\fR 4\n+.IP \\fBnomemory\\-seal\\fR 4\n .IX Item \"nomemory-seal\"\n .PD\n Instruct the executable or shared library that the all PT_LOAD segments\n should be sealed to avoid further manipulation (such as changing the\n protection flags, the segment size, or remove the mapping).\n This is a security hardening that requires system support. This\n generates GNU_PROPERTY_MEMORY_SEAL in .note.gnu.property section\n@@ -1421,125 +1424,125 @@\n dynamic linker to resolve all symbols when the program is started, or\n when the shared library is loaded by dlopen, instead of deferring\n function call resolution to the point when the function is first\n called.\n .IP \\fBorigin\\fR 4\n .IX Item \"origin\"\n Specify that the object requires \\fR\\f(CB$ORIGIN\\fR\\fB\\fR handling in paths.\n-.IP \\fBpack-relative-relocs\\fR 4\n+.IP \\fBpack\\-relative\\-relocs\\fR 4\n .IX Item \"pack-relative-relocs\"\n .PD 0\n-.IP \\fBnopack-relative-relocs\\fR 4\n+.IP \\fBnopack\\-relative\\-relocs\\fR 4\n .IX Item \"nopack-relative-relocs\"\n .PD\n-Generate compact relative relocation in position-independent executable\n+Generate compact relative relocation in position\\-independent executable\n and shared library. It adds \\f(CW\\*(C`DT_RELR\\*(C'\\fR, \\f(CW\\*(C`DT_RELRSZ\\*(C'\\fR and\n \\&\\f(CW\\*(C`DT_RELRENT\\*(C'\\fR entries to the dynamic section. It is ignored when\n-building position-dependent executable and relocatable output.\n-\\&\\fBnopack-relative-relocs\\fR is the default, which disables compact\n+building position\\-dependent executable and relocatable output.\n+\\&\\fBnopack\\-relative\\-relocs\\fR is the default, which disables compact\n relative relocation. When linked against the GNU C Library, a\n GLIBC_ABI_DT_RELR symbol version dependency on the shared C Library is\n added to the output. Supported for i386 and x86\\-64.\n .IP \\fBrelro\\fR 4\n .IX Item \"relro\"\n .PD 0\n .IP \\fBnorelro\\fR 4\n .IX Item \"norelro\"\n .PD\n Create an ELF \\f(CW\\*(C`PT_GNU_RELRO\\*(C'\\fR segment header in the object. This\n-specifies a memory segment that should be made read-only after\n-relocation, if supported. Specifying \\fBcommon-page-size\\fR smaller\n+specifies a memory segment that should be made read\\-only after\n+relocation, if supported. Specifying \\fBcommon\\-page\\-size\\fR smaller\n than the system page size will render this protection ineffective.\n-Don't create an ELF \\f(CW\\*(C`PT_GNU_RELRO\\*(C'\\fR segment if \\fBnorelro\\fR.\n-.IP \\fBreport-relative-reloc\\fR 4\n+Don\\*(Aqt create an ELF \\f(CW\\*(C`PT_GNU_RELRO\\*(C'\\fR segment if \\fBnorelro\\fR.\n+.IP \\fBreport\\-relative\\-reloc\\fR 4\n .IX Item \"report-relative-reloc\"\n Report dynamic relative relocations generated by linker. Supported for\n Linux/i386 and Linux/x86_64.\n .IP \\fBsectionheader\\fR 4\n .IX Item \"sectionheader\"\n .PD 0\n .IP \\fBnosectionheader\\fR 4\n .IX Item \"nosectionheader\"\n .PD\n-Generate section header. Don't generate section header if\n+Generate section header. Don\\*(Aqt generate section header if\n \\&\\fBnosectionheader\\fR is used. \\fBsectionheader\\fR is the default.\n-.IP \\fBseparate-code\\fR 4\n+.IP \\fBseparate\\-code\\fR 4\n .IX Item \"separate-code\"\n .PD 0\n-.IP \\fBnoseparate-code\\fR 4\n+.IP \\fBnoseparate\\-code\\fR 4\n .IX Item \"noseparate-code\"\n .PD\n Create separate code \\f(CW\\*(C`PT_LOAD\\*(C'\\fR segment header in the object. This\n specifies a memory segment that should contain only instructions and must\n-be in wholly disjoint pages from any other data. Don't create separate\n-code \\f(CW\\*(C`PT_LOAD\\*(C'\\fR segment if \\fBnoseparate-code\\fR is used.\n+be in wholly disjoint pages from any other data. Don\\*(Aqt create separate\n+code \\f(CW\\*(C`PT_LOAD\\*(C'\\fR segment if \\fBnoseparate\\-code\\fR is used.\n .IP \\fBshstk\\fR 4\n .IX Item \"shstk\"\n Generate GNU_PROPERTY_X86_FEATURE_1_SHSTK in .note.gnu.property section\n to indicate compatibility with Intel Shadow Stack. Supported for\n Linux/i386 and Linux/x86_64.\n .IP \\fBstack\\-size=\\fR\\fIvalue\\fR 4\n .IX Item \"stack-size=value\"\n Specify a stack size for an ELF \\f(CW\\*(C`PT_GNU_STACK\\*(C'\\fR segment.\n-Specifying zero will override any default non-zero sized\n+Specifying zero will override any default non\\-zero sized\n \\&\\f(CW\\*(C`PT_GNU_STACK\\*(C'\\fR segment creation.\n-.IP \\fBstart-stop-gc\\fR 4\n+.IP \\fBstart\\-stop\\-gc\\fR 4\n .IX Item \"start-stop-gc\"\n .PD 0\n-.IP \\fBnostart-stop-gc\\fR 4\n+.IP \\fBnostart\\-stop\\-gc\\fR 4\n .IX Item \"nostart-stop-gc\"\n .PD\n When \\fB\\-\\-gc\\-sections\\fR is in effect, a reference from a retained\n section to \\f(CW\\*(C`_\\|_start_SECNAME\\*(C'\\fR or \\f(CW\\*(C`_\\|_stop_SECNAME\\*(C'\\fR causes all\n input sections named \\f(CW\\*(C`SECNAME\\*(C'\\fR to also be retained, if\n \\&\\f(CW\\*(C`SECNAME\\*(C'\\fR is representable as a C identifier and either\n \\&\\f(CW\\*(C`_\\|_start_SECNAME\\*(C'\\fR or \\f(CW\\*(C`_\\|_stop_SECNAME\\*(C'\\fR is synthesized by the\n-linker. \\fB\\-z start-stop-gc\\fR disables this effect, allowing\n+linker. \\fB\\-z start\\-stop\\-gc\\fR disables this effect, allowing\n sections to be garbage collected as if the special synthesized symbols\n-were not defined. \\fB\\-z start-stop-gc\\fR has no effect on a\n+were not defined. \\fB\\-z start\\-stop\\-gc\\fR has no effect on a\n definition of \\f(CW\\*(C`_\\|_start_SECNAME\\*(C'\\fR or \\f(CW\\*(C`_\\|_stop_SECNAME\\*(C'\\fR in an\n object file or linker script. Such a definition will prevent the\n linker providing a synthesized \\f(CW\\*(C`_\\|_start_SECNAME\\*(C'\\fR or\n \\&\\f(CW\\*(C`_\\|_stop_SECNAME\\*(C'\\fR respectively, and therefore the special\n treatment by garbage collection for those references.\n .IP \\fBstart\\-stop\\-visibility=\\fR\\fIvalue\\fR 4\n .IX Item \"start-stop-visibility=value\"\n Specify the ELF symbol visibility for synthesized\n \\&\\f(CW\\*(C`_\\|_start_SECNAME\\*(C'\\fR and \\f(CW\\*(C`_\\|_stop_SECNAME\\*(C'\\fR symbols. \\fIvalue\\fR must be exactly \\fBdefault\\fR,\n \\&\\fBinternal\\fR, \\fBhidden\\fR, or \\fBprotected\\fR. If no \\fB\\-z\n-start-stop-visibility\\fR option is given, \\fBprotected\\fR is used for\n-compatibility with historical practice. However, it's highly\n+start\\-stop\\-visibility\\fR option is given, \\fBprotected\\fR is used for\n+compatibility with historical practice. However, it\\*(Aqs highly\n recommended to use \\fB\\-z start\\-stop\\-visibility=hidden\\fR in new\n programs and shared libraries so that these symbols are not exported\n-between shared objects, which is not usually what's intended.\n+between shared objects, which is not usually what\\*(Aqs intended.\n .IP \\fBtext\\fR 4\n .IX Item \"text\"\n .PD 0\n .IP \\fBnotext\\fR 4\n .IX Item \"notext\"\n .IP \\fBtextoff\\fR 4\n .IX Item \"textoff\"\n .PD\n-Report an error if DT_TEXTREL is set, i.e., if the position-independent\n-or shared object has dynamic relocations in read-only sections. Don't\n+Report an error if DT_TEXTREL is set, i.e., if the position\\-independent\n+or shared object has dynamic relocations in read\\-only sections. Don\\*(Aqt\n report an error if \\fBnotext\\fR or \\fBtextoff\\fR.\n .IP \\fBundefs\\fR 4\n .IX Item \"undefs\"\n Do not report unresolved symbol references from regular object files,\n either when creating an executable, or when creating a shared library.\n This option is the inverse of \\fB\\-z defs\\fR.\n-.IP \\fBunique-symbol\\fR 4\n+.IP \\fBunique\\-symbol\\fR 4\n .IX Item \"unique-symbol\"\n .PD 0\n-.IP \\fBnounique-symbol\\fR 4\n+.IP \\fBnounique\\-symbol\\fR 4\n .IX Item \"nounique-symbol\"\n .PD\n Avoid duplicated local symbol names in the symbol string table. Append\n-\".\\f(CW\\*(C`number\\*(C'\\fR\" to duplicated local symbol names if \\fBunique-symbol\\fR\n-is used. \\fBnounique-symbol\\fR is the default.\n+\".\\f(CW\\*(C`number\\*(C'\\fR\" to duplicated local symbol names if \\fBunique\\-symbol\\fR\n+is used. \\fBnounique\\-symbol\\fR is the default.\n .IP \\fBx86\\-64\\-baseline\\fR 4\n .IX Item \"x86-64-baseline\"\n .PD 0\n .IP \\fBx86\\-64\\-v2\\fR 4\n .IX Item \"x86-64-v2\"\n .IP \\fBx86\\-64\\-v3\\fR 4\n .IX Item \"x86-64-v3\"\n@@ -1619,17 +1622,17 @@\n .PD\n This option affects ELF DT_NEEDED tags for dynamic libraries mentioned\n on the command line after the \\fB\\-\\-as\\-needed\\fR option. Normally\n the linker will add a DT_NEEDED tag for each dynamic library mentioned\n on the command line, regardless of whether the library is actually\n needed or not. \\fB\\-\\-as\\-needed\\fR causes a DT_NEEDED tag to only be\n emitted for a library that \\fIat that point in the link\\fR satisfies a\n-non-weak undefined symbol reference from a regular object file or, if\n+non\\-weak undefined symbol reference from a regular object file or, if\n the library is not found in the DT_NEEDED lists of other needed libraries, a\n-non-weak undefined symbol reference from another needed dynamic library.\n+non\\-weak undefined symbol reference from another needed dynamic library.\n Object files or libraries appearing on the command line \\fIafter\\fR\n the library in question do not affect whether the library is seen as\n needed. This is similar to the rules for extraction of object files\n from archives. \\fB\\-\\-no\\-as\\-needed\\fR restores the default behaviour.\n .Sp\n Note: On Linux based systems the \\fB\\-\\-as\\-needed\\fR option also has\n an affect on the behaviour of the \\fB\\-\\-rpath\\fR and\n@@ -1682,15 +1685,15 @@\n Do not link against shared libraries. This is only meaningful on\n platforms for which shared libraries are supported. The different\n variants of this option are for compatibility with various systems. You\n may use this option multiple times on the command line: it affects\n library searching for \\fB\\-l\\fR options which follow it. This\n option also implies \\fB\\-\\-unresolved\\-symbols=report\\-all\\fR. This\n option can be used with \\fB\\-shared\\fR. Doing so means that a\n-shared library is being created but that all of the library's external\n+shared library is being created but that all of the library\\*(Aqs external\n references must be resolved by pulling in entries from static\n libraries.\n .IP \\fB\\-Bsymbolic\\fR 4\n .IX Item \"-Bsymbolic\"\n When creating a shared library, bind references to global symbols to the\n definition within the shared library, if any. Normally, it is possible\n for a program linked against a shared library to override the definition\n@@ -1702,19 +1705,19 @@\n symbols to the definition within the shared library, if any.\n This option is only meaningful on ELF platforms which support shared\n libraries.\n .IP \\fB\\-Bno\\-symbolic\\fR 4\n .IX Item \"-Bno-symbolic\"\n This option can cancel previously specified \\fB\\-Bsymbolic\\fR and\n \\&\\fB\\-Bsymbolic\\-functions\\fR.\n-.IP \\fB\\-\\-dynamic\\-list=\\fR\\fIdynamic-list-file\\fR 4\n+.IP \\fB\\-\\-dynamic\\-list=\\fR\\fIdynamic\\-list\\-file\\fR 4\n .IX Item \"--dynamic-list=dynamic-list-file\"\n Specify the name of a dynamic list file to the linker. This is\n typically used when creating shared libraries to specify a list of\n-global symbols whose references shouldn't be bound to the definition\n+global symbols whose references shouldn\\*(Aqt be bound to the definition\n within the shared library, or creating dynamically linked executables\n to specify a list of symbols which should be added to the symbol table\n in the executable. This option is only meaningful on ELF platforms\n which support shared libraries.\n .Sp\n The format of the dynamic list is the same as the version node without\n scope and node name. See \\fBVERSION\\fR for more information.\n@@ -1735,27 +1738,27 @@\n .IX Item \"--no-check-sections\"\n .PD\n Asks the linker \\fInot\\fR to check section addresses after they have\n been assigned to see if there are any overlaps. Normally the linker will\n perform this check, and if it finds any overlaps it will produce\n suitable error messages. The linker does know about, and does make\n allowances for sections in overlays. The default behaviour can be\n-restored by using the command-line switch \\fB\\-\\-check\\-sections\\fR.\n+restored by using the command\\-line switch \\fB\\-\\-check\\-sections\\fR.\n Section overlap is not usually checked for relocatable links. You can\n force checking in that case by using the \\fB\\-\\-check\\-sections\\fR\n option.\n .IP \\fB\\-\\-copy\\-dt\\-needed\\-entries\\fR 4\n .IX Item \"--copy-dt-needed-entries\"\n .PD 0\n .IP \\fB\\-\\-no\\-copy\\-dt\\-needed\\-entries\\fR 4\n .IX Item \"--no-copy-dt-needed-entries\"\n .PD\n This option affects the treatment of dynamic libraries referred to\n by DT_NEEDED tags \\fIinside\\fR ELF dynamic libraries mentioned on the\n-command line. Normally the linker won't add a DT_NEEDED tag to the\n+command line. Normally the linker won\\*(Aqt add a DT_NEEDED tag to the\n output binary for each library mentioned in a DT_NEEDED tag in an\n input dynamic library. With \\fB\\-\\-copy\\-dt\\-needed\\-entries\\fR\n specified on the command line however any dynamic libraries that\n follow it will have their DT_NEEDED entries added. The default\n behaviour can be restored with \\fB\\-\\-no\\-copy\\-dt\\-needed\\-entries\\fR.\n .Sp\n This option also has an effect on the resolution of symbols in dynamic\n@@ -1792,39 +1795,39 @@\n usually wasted: the types are usually small but the names are often not.\n \\&\\fB\\-\\-ctf\\-variables\\fR causes the generation of such a section.\n The default behaviour can be restored with \\fB\\-\\-no\\-ctf\\-variables\\fR.\n .IP \\fB\\-\\-ctf\\-share\\-types=\\fR\\fImethod\\fR 4\n .IX Item \"--ctf-share-types=method\"\n Adjust the method used to share types between translation units in CTF.\n .RS 4\n-.IP \\fBshare-unconflicted\\fR 4\n+.IP \\fBshare\\-unconflicted\\fR 4\n .IX Item \"share-unconflicted\"\n Put all types that do not have ambiguous definitions into the shared dictionary,\n where debuggers can easily access them, even if they only occur in one\n translation unit. This is the default.\n-.IP \\fBshare-duplicated\\fR 4\n+.IP \\fBshare\\-duplicated\\fR 4\n .IX Item \"share-duplicated\"\n Put only types that occur in multiple translation units into the shared\n-dictionary: types with only one definition go into per-translation-unit\n+dictionary: types with only one definition go into per\\-translation\\-unit\n dictionaries. Types with ambiguous definitions in multiple translation units\n-always go into per-translation-unit dictionaries. This tends to make the CTF\n+always go into per\\-translation\\-unit dictionaries. This tends to make the CTF\n larger, but may reduce the amount of CTF in the shared dictionary. For very\n large projects this may speed up opening the CTF and save memory in the CTF\n consumer at runtime.\n .RE\n .RS 4\n .RE\n .IP \\fB\\-\\-no\\-define\\-common\\fR 4\n .IX Item \"--no-define-common\"\n This option inhibits the assignment of addresses to common symbols.\n The script command \\f(CW\\*(C`INHIBIT_COMMON_ALLOCATION\\*(C'\\fR has the same effect.\n .Sp\n The \\fB\\-\\-no\\-define\\-common\\fR option allows decoupling\n the decision to assign addresses to Common symbols from the choice\n-of the output file type; otherwise a non-Relocatable output type\n+of the output file type; otherwise a non\\-Relocatable output type\n forces assigning addresses to Common symbols.\n Using \\fB\\-\\-no\\-define\\-common\\fR allows Common symbols that are referenced\n from a shared library to be assigned addresses only in the main program.\n This eliminates the unused duplicate space in the shared library,\n and also prevents any possible confusion over resolving to the wrong\n duplicate when there are many dynamic modules with specialized search\n paths for runtime symbol resolution.\n@@ -1876,26 +1879,26 @@\n .IX Item \"-Ifile\"\n .PD 0\n .IP \\fB\\-\\-dynamic\\-linker=\\fR\\fIfile\\fR 4\n .IX Item \"--dynamic-linker=file\"\n .PD\n Set the name of the dynamic linker. This is only meaningful when\n generating dynamically linked ELF executables. The default dynamic\n-linker is normally correct; don't use this unless you know what you are\n+linker is normally correct; don\\*(Aqt use this unless you know what you are\n doing.\n .IP \\fB\\-\\-no\\-dynamic\\-linker\\fR 4\n .IX Item \"--no-dynamic-linker\"\n When producing an executable file, omit the request for a dynamic\n-linker to be used at load-time. This is only meaningful for ELF\n+linker to be used at load\\-time. This is only meaningful for ELF\n executables that contain dynamic relocations, and usually requires\n entry point code that is capable of processing these relocations.\n .IP \\fB\\-\\-embedded\\-relocs\\fR 4\n .IX Item \"--embedded-relocs\"\n This option is similar to the \\fB\\-\\-emit\\-relocs\\fR option except\n-that the relocs are stored in a target-specific section. This option\n+that the relocs are stored in a target\\-specific section. This option\n is only supported by the \\fBBFIN\\fR, \\fBCR16\\fR and \\fIM68K\\fR\n targets.\n .IP \\fB\\-\\-disable\\-multiple\\-abs\\-defs\\fR 4\n .IX Item \"--disable-multiple-abs-defs\"\n Do not allow multiple definitions with symbols included\n in filename invoked by \\-R or \\-\\-just\\-symbols\n .IP \\fB\\-\\-fatal\\-warnings\\fR 4\n@@ -1920,15 +1923,15 @@\n .IX Item \"--force-exe-suffix\"\n Make sure that an output file has a .exe suffix.\n .Sp\n If a successfully built fully linked output file does not have a\n \\&\\f(CW\\*(C`.exe\\*(C'\\fR or \\f(CW\\*(C`.dll\\*(C'\\fR suffix, this option forces the linker to copy\n the output file to one of the same name with a \\f(CW\\*(C`.exe\\*(C'\\fR suffix. This\n option is useful when using unmodified Unix makefiles on a Microsoft\n-Windows host, since some versions of Windows won't run an image unless\n+Windows host, since some versions of Windows won\\*(Aqt run an image unless\n it ends in a \\f(CW\\*(C`.exe\\*(C'\\fR suffix.\n .IP \\fB\\-\\-gc\\-sections\\fR 4\n .IX Item \"--gc-sections\"\n .PD 0\n .IP \\fB\\-\\-no\\-gc\\-sections\\fR 4\n .IX Item \"--no-gc-sections\"\n .PD\n@@ -1938,15 +1941,15 @@\n \\&\\fB\\-\\-no\\-gc\\-sections\\fR on the command line. Note that garbage\n collection for COFF and PE format targets is supported, but the\n implementation is currently considered to be experimental.\n .Sp\n \\&\\fB\\-\\-gc\\-sections\\fR decides which input sections are used by\n examining symbols and relocations. The section containing the entry\n symbol and all sections containing symbols undefined on the\n-command-line will be kept, as will sections containing symbols\n+command\\-line will be kept, as will sections containing symbols\n referenced by dynamic objects. Note that when building shared\n libraries, the linker must assume that any visible symbol is\n referenced. Once this initial set of sections has been determined,\n the linker recursively marks as used any section referenced by their\n relocations. See \\fB\\-\\-entry\\fR, \\fB\\-\\-undefined\\fR, and\n \\&\\fB\\-\\-gc\\-keep\\-exported\\fR.\n .Sp\n@@ -1979,15 +1982,15 @@\n collected regardless of the external visibility of contained symbols.\n Note that this option has no effect when linking shared objects since\n it is already the default behaviour. This option is only supported for\n ELF format targets.\n .IP \\fB\\-\\-print\\-output\\-format\\fR 4\n .IX Item \"--print-output-format\"\n Print the name of the default output format (perhaps influenced by\n-other command-line options). This is the string that would appear\n+other command\\-line options). This is the string that would appear\n in an \\f(CW\\*(C`OUTPUT_FORMAT\\*(C'\\fR linker script command.\n .IP \\fB\\-\\-print\\-memory\\-usage\\fR 4\n .IX Item \"--print-memory-usage\"\n Print used size, total size and used size of memory regions created with\n the \\fBMEMORY\\fR command. This is useful on embedded targets to have a\n quick view of amount of free memory. The format of the output has one\n headline and one line per region. It is both human readable and easily\n@@ -1996,18 +1999,18 @@\n .Vb 3\n \\& Memory region Used Size Region Size %age Used\n \\& ROM: 256 KB 1 MB 25.00%\n \\& RAM: 32 B 2 GB 0.00%\n .Ve\n .IP \\fB\\-\\-help\\fR 4\n .IX Item \"--help\"\n-Print a summary of the command-line options on the standard output and exit.\n+Print a summary of the command\\-line options on the standard output and exit.\n .IP \\fB\\-\\-target\\-help\\fR 4\n .IX Item \"--target-help\"\n-Print a summary of all target-specific options on the standard output and exit.\n+Print a summary of all target\\-specific options on the standard output and exit.\n .IP \\fB\\-Map=\\fR\\fImapfile\\fR 4\n .IX Item \"-Map=mapfile\"\n Print a link map to the file \\fImapfile\\fR. See the description of the\n \\&\\fB\\-M\\fR option, above. If \\fImapfile\\fR is just the character\n \\&\\f(CW\\*(C`\\-\\*(C'\\fR then the map will be written to stdout.\n .Sp\n Specifying a directory as \\fImapfile\\fR causes the linker map to be\n@@ -2045,15 +2048,15 @@\n .IP \\fB\\-\\-no\\-undefined\\fR 4\n .IX Item \"--no-undefined\"\n .PD 0\n .IP \"\\fB\\-z defs\\fR\" 4\n .IX Item \"-z defs\"\n .PD\n Report unresolved symbol references from regular object files. This\n-is done even if the linker is creating a non-symbolic shared library.\n+is done even if the linker is creating a non\\-symbolic shared library.\n The switch \\fB\\-\\-[no\\-]allow\\-shlib\\-undefined\\fR controls the\n behaviour for reporting unresolved references found in shared\n libraries being linked in.\n .Sp\n The effects of this option can be reverted by using \\f(CW\\*(C`\\-z undefs\\*(C'\\fR.\n .IP \\fB\\-\\-allow\\-multiple\\-definition\\fR 4\n .IX Item \"--allow-multiple-definition\"\n@@ -2101,15 +2104,15 @@\n .RE\n .IP \\fB\\-\\-error\\-handling\\-script=\\fR\\fIscriptname\\fR 4\n .IX Item \"--error-handling-script=scriptname\"\n If this option is provided then the linker will invoke\n \\&\\fIscriptname\\fR whenever an error is encountered. Currently however\n only two kinds of error are supported: missing symbols and missing\n libraries. Two arguments will be passed to script: the keyword\n-\"undefined-symbol\" or `missing\\-lib\" and the \\fIname\\fR of the\n+\"undefined\\-symbol\" or \\`missing\\-lib\" and the \\fIname\\fR of the\n undefined symbol or missing library. The intention is that the script\n will provide suggestions to the user as to where the symbol or library\n might be found. After the script has finished then the normal linker\n error message will be displayed.\n .Sp\n The availability of this option is controlled by a configure time\n switch, so it may not be present in specific implementations.\n@@ -2150,23 +2153,23 @@\n errors during the link process; it exits without writing an output file\n when it issues any error whatsoever.\n .IP \\fB\\-nostdlib\\fR 4\n .IX Item \"-nostdlib\"\n Only search library directories explicitly specified on the\n command line. Library directories specified in linker scripts\n (including linker scripts specified on the command line) are ignored.\n-.IP \\fB\\-\\-oformat=\\fR\\fIoutput-format\\fR 4\n+.IP \\fB\\-\\-oformat=\\fR\\fIoutput\\-format\\fR 4\n .IX Item \"--oformat=output-format\"\n \\&\\fBld\\fR may be configured to support more than one kind of object\n file. If your \\fBld\\fR is configured this way, you can use the\n \\&\\fB\\-\\-oformat\\fR option to specify the binary format for the output\n object file. Even when \\fBld\\fR is configured to support alternative\n-object formats, you don't usually need to specify this, as \\fBld\\fR\n+object formats, you don\\*(Aqt usually need to specify this, as \\fBld\\fR\n should be configured to produce as a default output format the most\n-usual format on each machine. \\fIoutput-format\\fR is a text string, the\n+usual format on each machine. \\fIoutput\\-format\\fR is a text string, the\n name of a particular format supported by the BFD libraries. (You can\n list the available binary formats with \\fBobjdump \\-i\\fR.) The script\n command \\f(CW\\*(C`OUTPUT_FORMAT\\*(C'\\fR can also specify the output format, but\n this option overrides it.\n .IP \"\\fB\\-\\-out\\-implib\\fR \\fIfile\\fR\" 4\n .IX Item \"--out-implib file\"\n Create an import library in \\fIfile\\fR corresponding to the executable\n@@ -2226,15 +2229,15 @@\n .IP \\fB\\-\\-retain\\-symbols\\-file=\\fR\\fIfilename\\fR 4\n .IX Item \"--retain-symbols-file=filename\"\n Retain \\fIonly\\fR the symbols listed in the file \\fIfilename\\fR,\n discarding all others. \\fIfilename\\fR is simply a flat file, with one\n symbol name per line. This option is especially useful in environments\n (such as VxWorks)\n where a large global symbol table is accumulated gradually, to conserve\n-run-time memory.\n+run\\-time memory.\n .Sp\n \\&\\fB\\-\\-retain\\-symbols\\-file\\fR does \\fInot\\fR discard undefined symbols,\n or symbols needed for relocations.\n .Sp\n You may only specify \\fB\\-\\-retain\\-symbols\\-file\\fR once in the command\n line. It overrides \\fB\\-s\\fR and \\fB\\-S\\fR.\n .IP \\fB\\-rpath=\\fR\\fIdir\\fR 4\n@@ -2267,16 +2270,16 @@\n the \\fB\\-rpath\\fR option.\n .IP \\fB\\-rpath\\-link=\\fR\\fIdir\\fR 4\n .IX Item \"-rpath-link=dir\"\n When using ELF or SunOS, one shared library may require another. This\n happens when an \\f(CW\\*(C`ld \\-shared\\*(C'\\fR link includes a shared library as one\n of the input files.\n .Sp\n-When the linker encounters such a dependency when doing a non-shared,\n-non-relocatable link, it will automatically try to locate the required\n+When the linker encounters such a dependency when doing a non\\-shared,\n+non\\-relocatable link, it will automatically try to locate the required\n shared library and include it in the link, if it is not included\n explicitly. In such a case, several directories are searched as\n described below. The \\fB\\-rpath\\-link\\fR option specifies the first\n set of directories to search. This option may specify a sequence of\n directory names either by providing a list of names separated by\n colons, or by appearing multiple times.\n .Sp\n@@ -2424,16 +2427,16 @@\n .IP \\fB\\-\\-sort\\-common=ascending\\fR 4\n .IX Item \"--sort-common=ascending\"\n .IP \\fB\\-\\-sort\\-common=descending\\fR 4\n .IX Item \"--sort-common=descending\"\n .PD\n This option tells \\fBld\\fR to sort the common symbols by alignment in\n ascending or descending order when it places them in the appropriate output\n-sections. The symbol alignments considered are sixteen-byte or larger,\n-eight-byte, four-byte, two-byte, and one-byte. This is to prevent gaps\n+sections. The symbol alignments considered are sixteen\\-byte or larger,\n+eight\\-byte, four\\-byte, two\\-byte, and one\\-byte. This is to prevent gaps\n between symbols due to alignment constraints. If no sorting order is\n specified, then descending order is assumed.\n .IP \\fB\\-\\-sort\\-section=name\\fR 4\n .IX Item \"--sort-section=name\"\n This option will apply \\f(CW\\*(C`SORT_BY_NAME\\*(C'\\fR to all wildcard section\n patterns in the linker script.\n .IP \\fB\\-\\-sort\\-section=alignment\\fR 4\n@@ -2465,19 +2468,19 @@\n .IP \\fB\\-\\-stats\\fR 4\n .IX Item \"--stats\"\n Compute and display statistics about the operation of the linker, such\n as execution time and memory usage.\n .IP \\fB\\-\\-sysroot=\\fR\\fIdirectory\\fR 4\n .IX Item \"--sysroot=directory\"\n Use \\fIdirectory\\fR as the location of the sysroot, overriding the\n-configure-time default. This option is only supported by linkers\n+configure\\-time default. This option is only supported by linkers\n that were configured using \\fB\\-\\-with\\-sysroot\\fR.\n .IP \\fB\\-\\-task\\-link\\fR 4\n .IX Item \"--task-link\"\n-This is used by COFF/PE based targets to create a task-linked object\n+This is used by COFF/PE based targets to create a task\\-linked object\n file where all of the global symbols have been converted to statics.\n .IP \\fB\\-\\-traditional\\-format\\fR 4\n .IX Item \"--traditional-format\"\n For some targets, the output of \\fBld\\fR is different in some ways from\n the output of some existing linker. This switch requests \\fBld\\fR to\n use the traditional format instead.\n .Sp\n@@ -2525,41 +2528,41 @@\n byte of the first segment. Note that when \\fB\\-pie\\fR is used with\n \\&\\fB\\-Ttext\\-segment=\\fR\\fIorg\\fR, the output executable is marked\n ET_EXEC so that the address of the first byte of the text segment will\n be guaranteed to be \\fIorg\\fR at run time.\n .IP \\fB\\-Trodata\\-segment=\\fR\\fIorg\\fR 4\n .IX Item \"-Trodata-segment=org\"\n When creating an ELF executable or shared object for a target where\n-the read-only data is in its own segment separate from the executable\n-text, it will set the address of the first byte of the read-only data segment.\n+the read\\-only data is in its own segment separate from the executable\n+text, it will set the address of the first byte of the read\\-only data segment.\n .IP \\fB\\-Tldata\\-segment=\\fR\\fIorg\\fR 4\n .IX Item \"-Tldata-segment=org\"\n When creating an ELF executable or shared object for x86\\-64 medium memory\n model, it will set the address of the first byte of the ldata segment.\n .IP \\fB\\-\\-unresolved\\-symbols=\\fR\\fImethod\\fR 4\n .IX Item \"--unresolved-symbols=method\"\n Determine how to handle unresolved symbols. There are four possible\n values for \\fBmethod\\fR:\n .RS 4\n-.IP \\fBignore-all\\fR 4\n+.IP \\fBignore\\-all\\fR 4\n .IX Item \"ignore-all\"\n Do not report any unresolved symbols.\n-.IP \\fBreport-all\\fR 4\n+.IP \\fBreport\\-all\\fR 4\n .IX Item \"report-all\"\n Report all unresolved symbols. This is the default.\n-.IP \\fBignore-in-object-files\\fR 4\n+.IP \\fBignore\\-in\\-object\\-files\\fR 4\n .IX Item \"ignore-in-object-files\"\n Report unresolved symbols that are contained in shared libraries, but\n ignore them if they come from regular object files.\n-.IP \\fBignore-in-shared-libs\\fR 4\n+.IP \\fBignore\\-in\\-shared\\-libs\\fR 4\n .IX Item \"ignore-in-shared-libs\"\n Report unresolved symbols that come from regular object files, but\n ignore them if they come from shared libraries. This can be useful\n when creating a dynamic binary and it is known that all the shared\n-libraries that it should be referencing are included on the linker's\n+libraries that it should be referencing are included on the linker\\*(Aqs\n command line.\n .RE\n .RS 4\n .Sp\n The behaviour for shared libraries on their own can also be controlled\n by the \\fB\\-\\-[no\\-]allow\\-shlib\\-undefined\\fR option.\n .Sp\n@@ -2573,22 +2576,22 @@\n .IP \\fB\\-\\-verbose[=\\fR\\fINUMBER\\fR\\fB]\\fR 4\n .IX Item \"--verbose[=NUMBER]\"\n .PD\n Display the version number for \\fBld\\fR and list the linker emulations\n supported. Display which input files can and cannot be opened. Display\n the linker script being used by the linker. If the optional \\fINUMBER\\fR\n argument > 1, plugin symbol status will also be displayed.\n-.IP \\fB\\-\\-version\\-script=\\fR\\fIversion-scriptfile\\fR 4\n+.IP \\fB\\-\\-version\\-script=\\fR\\fIversion\\-scriptfile\\fR 4\n .IX Item \"--version-script=version-scriptfile\"\n Specify the name of a version script to the linker. This is typically\n used when creating shared libraries to specify additional information\n about the version hierarchy for the library being created. This option\n is only fully supported on ELF platforms which support shared libraries;\n see \\fBVERSION\\fR. It is partially supported on PE platforms, which can\n-use version scripts to filter symbol visibility in auto-export mode: any\n+use version scripts to filter symbol visibility in auto\\-export mode: any\n symbols marked \\fBlocal\\fR in the version script will not be exported.\n .IP \\fB\\-\\-warn\\-common\\fR 4\n .IX Item \"--warn-common\"\n Warn when a common symbol is combined with another common symbol or with\n a symbol definition. Unix linkers allow this somewhat sloppy practice,\n but linkers on some other operating systems do not. This option allows\n you to find potential problems from combining global symbols.\n@@ -2638,15 +2641,15 @@\n .Sp\n .Vb 3\n \\& (
): warning: definition of \\`\\*(Aq\n \\& overriding common\n \\& (
): warning: common is here\n .Ve\n .IP 3. 4\n-Merging a common symbol with a previous same-sized common symbol.\n+Merging a common symbol with a previous same\\-sized common symbol.\n .Sp\n .Vb 3\n \\& (
): warning: multiple common\n \\& of \\`\\*(Aq\n \\& (
): warning: previous common is here\n .Ve\n .IP 4. 4\n@@ -2696,18 +2699,18 @@\n Only generate a warning if an object file requests an executable\n stack, but not if the \\fB\\-z execstack\\fR option is used.\n .RE\n .RS 4\n .Sp\n The default state depends upon how the linker was configured when it\n was built. The \\fB\\-\\-no\\-warn\\-execstack\\fR option always puts the\n-linker into the no-warnings state. The \\fB\\-\\-warn\\-execstack\\fR\n-option puts the linker into the warn-always state. The\n+linker into the no\\-warnings state. The \\fB\\-\\-warn\\-execstack\\fR\n+option puts the linker into the warn\\-always state. The\n \\&\\fB\\-\\-warn\\-execstack\\-objects\\fR option puts the linker into the\n-warn-for-object-files-only state.\n+warn\\-for\\-object\\-files\\-only state.\n .Sp\n Note: ELF format input files can specify that they need an executable\n stack by having a \\fI.note.GNU\\-stack\\fR section with the executable\n bit set in its section flags. They can specify that they do not need\n an executable stack by having the same section, but without the\n executable flag bit set. If an input file does not have a\n \\&\\fI.note.GNU\\-stack\\fR section then the default behaviour is target\n@@ -2720,52 +2723,52 @@\n .PD 0\n .IP \\fB\\-\\-no\\-error\\-execstack\\fR 4\n .IX Item \"--no-error-execstack\"\n .PD\n If the linker is going to generate a warning message about an\n executable stack then the \\fB\\-\\-error\\-execstack\\fR option will\n instead change that warning into an error. Note \\- this option does\n-not change the linker's execstack warning generation state. Use\n+not change the linker\\*(Aqs execstack warning generation state. Use\n \\&\\fB\\-\\-warn\\-execstack\\fR or \\fB\\-\\-warn\\-execstack\\-objects\\fR to set\n a specific warning state.\n .Sp\n The \\fB\\-\\-no\\-error\\-execstack\\fR option will restore the default\n behaviour of generating warning messages.\n .IP \\fB\\-\\-warn\\-multiple\\-gp\\fR 4\n .IX Item \"--warn-multiple-gp\"\n Warn if multiple global pointer values are required in the output file.\n This is only meaningful for certain processors, such as the Alpha.\n-Specifically, some processors put large-valued constants in a special\n+Specifically, some processors put large\\-valued constants in a special\n section. A special register (the global pointer) points into the middle\n of this section, so that constants can be loaded efficiently via a\n-base-register relative addressing mode. Since the offset in\n-base-register relative mode is fixed and relatively small (e.g., 16\n+base\\-register relative addressing mode. Since the offset in\n+base\\-register relative mode is fixed and relatively small (e.g., 16\n bits), this limits the maximum size of the constant pool. Thus, in\n large programs, it is often necessary to use multiple global pointer\n values in order to be able to address all possible constants. This\n option causes a warning to be issued whenever this case occurs.\n .IP \\fB\\-\\-warn\\-once\\fR 4\n .IX Item \"--warn-once\"\n Only warn once for each undefined symbol, rather than once per module\n which refers to it.\n .IP \\fB\\-\\-warn\\-rwx\\-segments\\fR 4\n .IX Item \"--warn-rwx-segments\"\n .PD 0\n .IP \\fB\\-\\-no\\-warn\\-rwx\\-segments\\fR 4\n .IX Item \"--no-warn-rwx-segments\"\n .PD\n-Warn if the linker creates a loadable, non-zero sized segment that has\n+Warn if the linker creates a loadable, non\\-zero sized segment that has\n all three of the read, write and execute permission flags set. Such a\n segment represents a potential security vulnerability. In addition\n warnings will be generated if a thread local storage segment is\n created with the execute permission flag set, regardless of whether or\n not it has the read and/or write flags set.\n .Sp\n These warnings are enabled by default. They can be disabled via the\n-\\&\\fB\\-\\-no\\-warn\\-rwx\\-segments\\fR option and re-enabled via the\n+\\&\\fB\\-\\-no\\-warn\\-rwx\\-segments\\fR option and re\\-enabled via the\n \\&\\fB\\-\\-warn\\-rwx\\-segments\\fR option.\n .IP \\fB\\-\\-error\\-rwx\\-segments\\fR 4\n .IX Item \"--error-rwx-segments\"\n .PD 0\n .IP \\fB\\-\\-no\\-error\\-rwx\\-segments\\fR 4\n .IX Item \"--no-error-rwx-segments\"\n .PD\n@@ -2784,40 +2787,40 @@\n Warn if the address of an output section is changed because of\n alignment. Typically, the alignment will be set by an input section.\n The address will only be changed if it not explicitly specified; that\n is, if the \\f(CW\\*(C`SECTIONS\\*(C'\\fR command does not specify a start address for\n the section.\n .IP \\fB\\-\\-warn\\-textrel\\fR 4\n .IX Item \"--warn-textrel\"\n-Warn if the linker adds DT_TEXTREL to a position-independent executable\n+Warn if the linker adds DT_TEXTREL to a position\\-independent executable\n or shared object.\n .IP \\fB\\-\\-warn\\-alternate\\-em\\fR 4\n .IX Item \"--warn-alternate-em\"\n Warn if an object has alternate ELF machine code.\n .IP \\fB\\-\\-warn\\-unresolved\\-symbols\\fR 4\n .IX Item \"--warn-unresolved-symbols\"\n If the linker is going to report an unresolved symbol (see the option\n \\&\\fB\\-\\-unresolved\\-symbols\\fR) it will normally generate an error.\n This option makes it generate a warning instead.\n .IP \\fB\\-\\-error\\-unresolved\\-symbols\\fR 4\n .IX Item \"--error-unresolved-symbols\"\n-This restores the linker's default behaviour of generating errors when\n+This restores the linker\\*(Aqs default behaviour of generating errors when\n it is reporting unresolved symbols.\n .IP \\fB\\-\\-whole\\-archive\\fR 4\n .IX Item \"--whole-archive\"\n For each archive mentioned on the command line after the\n \\&\\fB\\-\\-whole\\-archive\\fR option, include every object file in the archive\n in the link, rather than searching the archive for the required object\n files. This is normally used to turn an archive file into a shared\n library, forcing every object to be included in the resulting shared\n library. This option may be used more than once.\n .Sp\n-Two notes when using this option from gcc: First, gcc doesn't know\n+Two notes when using this option from gcc: First, gcc doesn\\*(Aqt know\n about this option, so you have to use \\fB\\-Wl,\\-whole\\-archive\\fR.\n-Second, don't forget to use \\fB\\-Wl,\\-no\\-whole\\-archive\\fR after your\n+Second, don\\*(Aqt forget to use \\fB\\-Wl,\\-no\\-whole\\-archive\\fR after your\n list of archives, because gcc will add its own list of archives to\n your link and you may not want this flag to affect those as well.\n .IP \\fB\\-\\-wrap=\\fR\\fIsymbol\\fR 4\n .IX Item \"--wrap=symbol\"\n Use a wrapper function for \\fIsymbol\\fR. Any undefined reference to\n \\&\\fIsymbol\\fR will be resolved to \\f(CW\\*(C`_\\|_wrap_\\fR\\f(CIsymbol\\fR\\f(CW\\*(C'\\fR. Any\n undefined reference to \\f(CW\\*(C`_\\|_real_\\fR\\f(CIsymbol\\fR\\f(CW\\*(C'\\fR will be resolved to\n@@ -2895,22 +2898,22 @@\n \\&\\fB\\-\\-enable\\-new\\-dtags\\fR, the new dynamic tags will be created as needed\n and older dynamic tags will be omitted.\n If you specify \\fB\\-\\-disable\\-new\\-dtags\\fR, no new dynamic tags will be\n created. By default, the new dynamic tags are not created. Note that\n those options are only available for ELF systems.\n .IP \\fB\\-\\-hash\\-size=\\fR\\fInumber\\fR 4\n .IX Item \"--hash-size=number\"\n-Set the default size of the linker's hash tables to a prime number\n+Set the default size of the linker\\*(Aqs hash tables to a prime number\n close to \\fInumber\\fR. Increasing this value can reduce the length of\n time it takes the linker to perform its tasks, at the expense of\n-increasing the linker's memory requirements. Similarly reducing this\n+increasing the linker\\*(Aqs memory requirements. Similarly reducing this\n value can reduce the memory requirements at the expense of speed.\n .IP \\fB\\-\\-hash\\-style=\\fR\\fIstyle\\fR 4\n .IX Item \"--hash-style=style\"\n-Set the type of linker's hash table(s). \\fIstyle\\fR can be either\n+Set the type of linker\\*(Aqs hash table(s). \\fIstyle\\fR can be either\n \\&\\f(CW\\*(C`sysv\\*(C'\\fR for classic ELF \\f(CW\\*(C`.hash\\*(C'\\fR section, \\f(CW\\*(C`gnu\\*(C'\\fR for\n new style GNU \\f(CW\\*(C`.gnu.hash\\*(C'\\fR section or \\f(CW\\*(C`both\\*(C'\\fR for both\n the classic ELF \\f(CW\\*(C`.hash\\*(C'\\fR and new style GNU \\f(CW\\*(C`.gnu.hash\\*(C'\\fR\n hash tables. The default depends upon how the linker was configured,\n but for most Linux based systems it will be \\f(CW\\*(C`both\\*(C'\\fR.\n .IP \\fB\\-\\-compress\\-debug\\-sections=none\\fR 4\n .IX Item \"--compress-debug-sections=none\"\n@@ -2923,45 +2926,45 @@\n .IX Item \"--compress-debug-sections=zlib-gabi\"\n .IP \\fB\\-\\-compress\\-debug\\-sections=zstd\\fR 4\n .IX Item \"--compress-debug-sections=zstd\"\n .PD\n On ELF platforms, these options control how DWARF debug sections are\n compressed using zlib.\n .Sp\n-\\&\\fB\\-\\-compress\\-debug\\-sections=none\\fR doesn't compress DWARF debug\n+\\&\\fB\\-\\-compress\\-debug\\-sections=none\\fR doesn\\*(Aqt compress DWARF debug\n sections. \\fB\\-\\-compress\\-debug\\-sections=zlib\\-gnu\\fR compresses\n DWARF debug sections and renames them to begin with \\fB.zdebug\\fR\n instead of \\fB.debug\\fR. \\fB\\-\\-compress\\-debug\\-sections=zlib\\-gabi\\fR\n also compresses DWARF debug sections, but rather than renaming them it\n-sets the SHF_COMPRESSED flag in the sections' headers.\n+sets the SHF_COMPRESSED flag in the sections\\*(Aq headers.\n .Sp\n The \\fB\\-\\-compress\\-debug\\-sections=zlib\\fR option is an alias for\n \\&\\fB\\-\\-compress\\-debug\\-sections=zlib\\-gabi\\fR.\n .Sp\n \\&\\fB\\-\\-compress\\-debug\\-sections=zstd\\fR compresses DWARF debug sections using\n zstd.\n .Sp\n Note that this option overrides any compression in input debug\n sections, so if a binary is linked with \\fB\\-\\-compress\\-debug\\-sections=none\\fR\n for example, then any compressed debug sections in input files will be\n uncompressed before they are copied into the output binary.\n .Sp\n The default compression behaviour varies depending upon the target\n involved and the configure options used to build the toolchain. The\n-default can be determined by examining the output from the linker's\n+default can be determined by examining the output from the linker\\*(Aqs\n \\&\\fB\\-\\-help\\fR option.\n .IP \\fB\\-\\-reduce\\-memory\\-overheads\\fR 4\n .IX Item \"--reduce-memory-overheads\"\n This option reduces memory requirements at ld runtime, at the expense of\n linking speed. This was introduced to select the old O(n^2) algorithm\n for link map file generation, rather than the new O(n) algorithm which uses\n about 40% more memory for symbol storage.\n .Sp\n Another effect of the switch is to set the default hash table size to\n-1021, which again saves memory at the cost of lengthening the linker's\n+1021, which again saves memory at the cost of lengthening the linker\\*(Aqs\n run time. This is not done however if the \\fB\\-\\-hash\\-size\\fR switch\n has been used.\n .Sp\n The \\fB\\-\\-reduce\\-memory\\-overheads\\fR switch may be also be used to\n enable other tradeoffs in future versions of the linker.\n .IP \\fB\\-\\-max\\-cache\\-size=\\fR\\fIsize\\fR 4\n .IX Item \"--max-cache-size=size\"\n@@ -2984,28 +2987,28 @@\n specified as an even number of hexadecimal digits (\\f(CW\\*(C`\\-\\*(C'\\fR and\n \\&\\f(CW\\*(C`:\\*(C'\\fR characters between digit pairs are ignored). If \\fIstyle\\fR\n is omitted, \\f(CW\\*(C`sha1\\*(C'\\fR is used.\n .Sp\n The \\f(CW\\*(C`md5\\*(C'\\fR, \\f(CW\\*(C`sha1\\*(C'\\fR, and \\f(CW\\*(C`xx\\*(C'\\fR styles produces an\n identifier that is always the same in an identical output file, but\n are almost certainly unique among all nonidentical output files. It\n-is not intended to be compared as a checksum for the file's contents.\n+is not intended to be compared as a checksum for the file\\*(Aqs contents.\n A linked file may be changed later by other tools, but the build ID\n bit string identifying the original linked file does not change.\n .Sp\n Passing \\f(CW\\*(C`none\\*(C'\\fR for \\fIstyle\\fR disables the setting from any\n \\&\\f(CW\\*(C`\\-\\-build\\-id\\*(C'\\fR options earlier on the command line.\n .IP \\fB\\-\\-package\\-metadata=\\fR\\fIJSON\\fR 4\n .IX Item \"--package-metadata=JSON\"\n Request the creation of a \\f(CW\\*(C`.note.package\\*(C'\\fR ELF note section. The\n contents of the note are in JSON format, as per the package metadata\n specification. For more information see:\n https://systemd.io/ELF_PACKAGE_METADATA/\n-The JSON argument support percent-encoding and following %[string]\n-(where string refers to the name in HTML's Named Character References)\n+The JSON argument support percent\\-encoding and following %[string]\n+(where string refers to the name in HTML\\*(Aqs Named Character References)\n encoding: \\fB%[comma]\\fR for \\fB,\\fR, \\fB%[lbrace]\\fR for \\fB{\\fR,\n \\&\\fB%[quot]\\fR for \\fB\"\\fR, \\fB%[rbrace]\\fR for \\fB}\\fR, and\n \\&\\fB%[space]\\fR for space character.\n If the JSON argument is missing/empty then this will disable the\n creation of the metadata note, if one had been enabled by an earlier\n occurrence of the \\-\\-package\\-metadata option.\n If the linker has been built with libjansson, then the JSON string\n@@ -3017,21 +3020,21 @@\n use this option. In addition, the linker fully supports the standard\n \\&\\f(CW\\*(C`*.def\\*(C'\\fR files, which may be specified on the linker command line\n like an object file (in fact, it should precede archives it exports\n symbols from, to ensure that they get linked in, just like a normal\n object file).\n .PP\n In addition to the options common to all targets, the i386 PE linker\n-support additional command-line options that are specific to the i386\n+support additional command\\-line options that are specific to the i386\n PE target. Options that take values may be separated from their\n values by either a space or an equals sign.\n .IP \\fB\\-\\-add\\-stdcall\\-alias\\fR 4\n .IX Item \"--add-stdcall-alias\"\n If given, symbols with a stdcall suffix (@\\fInn\\fR) will be exported\n-as-is and also with the suffix stripped.\n+as\\-is and also with the suffix stripped.\n [This option is specific to the i386 PE targeted port of the linker]\n .IP \"\\fB\\-\\-base\\-file\\fR \\fIfile\\fR\" 4\n .IX Item \"--base-file file\"\n Use \\fIfile\\fR as the name of a file in which to save the base\n addresses of all the relocations needed for generating DLLs with\n \\&\\fIdlltool\\fR.\n [This is an i386 PE specific option]\n@@ -3046,24 +3049,24 @@\n .PD 0\n .IP \\fB\\-\\-disable\\-long\\-section\\-names\\fR 4\n .IX Item \"--disable-long-section-names\"\n .PD\n The PE variants of the COFF object format add an extension that permits\n the use of section names longer than eight characters, the normal limit\n for COFF. By default, these names are only allowed in object files, as\n-fully-linked executable images do not carry the COFF string table required\n+fully\\-linked executable images do not carry the COFF string table required\n to support the longer names. As a GNU extension, it is possible to\n allow their use in executable images as well, or to (probably pointlessly!)\n disallow it in object files, by using these two options. Executable images\n-generated with these long section names are slightly non-standard, carrying\n+generated with these long section names are slightly non\\-standard, carrying\n as they do a string table, and may generate confusing output when examined\n-with non-GNU PE-aware tools, such as file viewers and dumpers. However,\n+with non\\-GNU PE\\-aware tools, such as file viewers and dumpers. However,\n GDB relies on the use of PE long section names to find Dwarf\\-2 debug\n information sections in an executable image at runtime, and so if neither\n-option is specified on the command-line, \\fBld\\fR will enable long\n+option is specified on the command\\-line, \\fBld\\fR will enable long\n section names, overriding the default and technically correct behaviour,\n when it finds the presence of debug information while linking an executable\n image and not stripping symbols.\n [This option is valid for all PE targeted ports of the linker]\n .IP \\fB\\-\\-enable\\-stdcall\\-fixup\\fR 4\n .IX Item \"--enable-stdcall-fixup\"\n .PD 0\n@@ -3074,49 +3077,49 @@\n do \"fuzzy linking\" by looking for another defined symbol that differs\n only in the format of the symbol name (cdecl vs stdcall) and will\n resolve that symbol by linking to the match. For example, the\n undefined symbol \\f(CW\\*(C`_foo\\*(C'\\fR might be linked to the function\n \\&\\f(CW\\*(C`_foo@12\\*(C'\\fR, or the undefined symbol \\f(CW\\*(C`_bar@16\\*(C'\\fR might be linked\n to the function \\f(CW\\*(C`_bar\\*(C'\\fR. When the linker does this, it prints a\n warning, since it normally should have failed to link, but sometimes\n-import libraries generated from third-party dlls may need this feature\n+import libraries generated from third\\-party dlls may need this feature\n to be usable. If you specify \\fB\\-\\-enable\\-stdcall\\-fixup\\fR, this\n feature is fully enabled and warnings are not printed. If you specify\n \\&\\fB\\-\\-disable\\-stdcall\\-fixup\\fR, this feature is disabled and such\n mismatches are considered to be errors.\n [This option is specific to the i386 PE targeted port of the linker]\n .IP \\fB\\-\\-leading\\-underscore\\fR 4\n .IX Item \"--leading-underscore\"\n .PD 0\n .IP \\fB\\-\\-no\\-leading\\-underscore\\fR 4\n .IX Item \"--no-leading-underscore\"\n .PD\n-For most targets default symbol-prefix is an underscore and is defined\n-in target's description. By this option it is possible to\n-disable/enable the default underscore symbol-prefix.\n+For most targets default symbol\\-prefix is an underscore and is defined\n+in target\\*(Aqs description. By this option it is possible to\n+disable/enable the default underscore symbol\\-prefix.\n .IP \\fB\\-\\-export\\-all\\-symbols\\fR 4\n .IX Item \"--export-all-symbols\"\n If given, all global symbols in the objects used to build a DLL will\n be exported by the DLL. Note that this is the default if there\n-otherwise wouldn't be any exported symbols. When symbols are\n+otherwise wouldn\\*(Aqt be any exported symbols. When symbols are\n explicitly exported via DEF files or implicitly exported via function\n attributes, the default is to not export anything else unless this\n option is given. Note that the symbols \\f(CW\\*(C`DllMain@12\\*(C'\\fR,\n \\&\\f(CW\\*(C`DllEntryPoint@0\\*(C'\\fR, \\f(CW\\*(C`DllMainCRTStartup@12\\*(C'\\fR, and\n \\&\\f(CW\\*(C`impure_ptr\\*(C'\\fR will not be automatically\n exported. Also, symbols imported from other DLLs will not be\n-re-exported, nor will symbols specifying the DLL's internal layout\n+re\\-exported, nor will symbols specifying the DLL\\*(Aqs internal layout\n such as those beginning with \\f(CW\\*(C`_head_\\*(C'\\fR or ending with\n \\&\\f(CW\\*(C`_iname\\*(C'\\fR. In addition, no symbols from \\f(CW\\*(C`libgcc\\*(C'\\fR,\n \\&\\f(CW\\*(C`libstd++\\*(C'\\fR, \\f(CW\\*(C`libmingw32\\*(C'\\fR, or \\f(CW\\*(C`crtX.o\\*(C'\\fR will be exported.\n Symbols whose names begin with \\f(CW\\*(C`_\\|_rtti_\\*(C'\\fR or \\f(CW\\*(C`_\\|_builtin_\\*(C'\\fR will\n not be exported, to help with C++ DLLs. Finally, there is an\n-extensive list of cygwin-private symbols that are not exported\n+extensive list of cygwin\\-private symbols that are not exported\n (obviously, this applies on when building DLLs for cygwin targets).\n-These cygwin-excludes are: \\f(CW\\*(C`_cygwin_dll_entry@12\\*(C'\\fR,\n+These cygwin\\-excludes are: \\f(CW\\*(C`_cygwin_dll_entry@12\\*(C'\\fR,\n \\&\\f(CW\\*(C`_cygwin_crt0_common@8\\*(C'\\fR, \\f(CW\\*(C`_cygwin_noncygwin_dll_entry@12\\*(C'\\fR,\n \\&\\f(CW\\*(C`_fmode\\*(C'\\fR, \\f(CW\\*(C`_impure_ptr\\*(C'\\fR, \\f(CW\\*(C`cygwin_attach_dll\\*(C'\\fR,\n \\&\\f(CW\\*(C`cygwin_premain0\\*(C'\\fR, \\f(CW\\*(C`cygwin_premain1\\*(C'\\fR, \\f(CW\\*(C`cygwin_premain2\\*(C'\\fR,\n \\&\\f(CW\\*(C`cygwin_premain3\\*(C'\\fR, and \\f(CW\\*(C`environ\\*(C'\\fR.\n [This option is specific to the i386 PE targeted port of the linker]\n .IP \"\\fB\\-\\-exclude\\-symbols\\fR \\fIsymbol\\fR\\fB,\\fR\\fIsymbol\\fR\\fB,...\\fR\" 4\n .IX Item \"--exclude-symbols symbol,symbol,...\"\n@@ -3200,21 +3203,21 @@\n .PD 0\n .IP \\fB\\-\\-enable\\-auto\\-image\\-base=\\fR\\fIvalue\\fR 4\n .IX Item \"--enable-auto-image-base=value\"\n .PD\n Automatically choose the image base for DLLs, optionally starting with base\n \\&\\fIvalue\\fR, unless one is specified using the \\f(CW\\*(C`\\-\\-image\\-base\\*(C'\\fR argument.\n By using a hash generated from the dllname to create unique image bases\n-for each DLL, in-memory collisions and relocations which can delay program\n+for each DLL, in\\-memory collisions and relocations which can delay program\n execution are avoided.\n [This option is specific to the i386 PE targeted port of the linker]\n .IP \\fB\\-\\-disable\\-auto\\-image\\-base\\fR 4\n .IX Item \"--disable-auto-image-base\"\n Do not automatically generate a unique image base. If there is no\n-user-specified image base (\\f(CW\\*(C`\\-\\-image\\-base\\*(C'\\fR) then use the platform\n+user\\-specified image base (\\f(CW\\*(C`\\-\\-image\\-base\\*(C'\\fR) then use the platform\n default.\n [This option is specific to the i386 PE targeted port of the linker]\n .IP \"\\fB\\-\\-dll\\-search\\-prefix\\fR \\fIstring\\fR\" 4\n .IX Item \"--dll-search-prefix string\"\n When linking dynamically to a dll without an import library,\n search for \\f(CW\\*(C`.dll\\*(C'\\fR in preference to\n \\&\\f(CW\\*(C`lib.dll\\*(C'\\fR. This behaviour allows easy distinction\n@@ -3228,29 +3231,29 @@\n DATA imports from DLLs, thus making it possible to bypass the dllimport\n mechanism on the user side and to reference unmangled symbol names.\n [This option is specific to the i386 PE targeted port of the linker]\n .Sp\n The following remarks pertain to the original implementation of the\n feature and are obsolete nowadays for Cygwin and MinGW targets.\n .Sp\n-Note: Use of the 'auto\\-import' extension will cause the text section\n+Note: Use of the \\*(Aqauto\\-import\\*(Aq extension will cause the text section\n of the image file to be made writable. This does not conform to the\n-PE-COFF format specification published by Microsoft.\n+PE\\-COFF format specification published by Microsoft.\n .Sp\n-Note \\- use of the 'auto\\-import' extension will also cause read only\n+Note \\- use of the \\*(Aqauto\\-import\\*(Aq extension will also cause read only\n data which would normally be placed into the .rdata section to be\n placed into the .data section instead. This is in order to work\n around a problem with consts that is described here:\n http://www.cygwin.com/ml/cygwin/2004\\-09/msg01101.html\n .Sp\n-Using 'auto\\-import' generally will 'just work' \\-\\- but sometimes you may\n+Using \\*(Aqauto\\-import\\*(Aq generally will \\*(Aqjust work\\*(Aq \\-\\- but sometimes you may\n see this message:\n .Sp\n-\"variable '' can't be auto-imported. Please read the\n-documentation for ld's \\f(CW\\*(C`\\-\\-enable\\-auto\\-import\\*(C'\\fR for details.\"\n+\"variable \\*(Aq\\*(Aq can\\*(Aqt be auto\\-imported. Please read the\n+documentation for ld\\*(Aqs \\f(CW\\*(C`\\-\\-enable\\-auto\\-import\\*(C'\\fR for details.\"\n .Sp\n This message occurs when some (sub)expression accesses an address\n ultimately given by the sum of two constants (Win32 import tables only\n allow one). Instances where this may occur include accesses to member\n fields of struct variables imported from a DLL, as well as using a\n constant index into an array variable imported from a DLL. Any\n multiword variable (arrays, structs, long long, etc) may trigger\n@@ -3261,18 +3264,18 @@\n There are several ways to address this difficulty, regardless of the\n data type of the exported variable:\n .Sp\n One way is to use \\-\\-enable\\-runtime\\-pseudo\\-reloc switch. This leaves the task\n of adjusting references in your client code for runtime environment, so\n this method works only when runtime environment supports this feature.\n .Sp\n-A second solution is to force one of the 'constants' to be a variable \\-\\-\n-that is, unknown and un-optimizable at compile time. For arrays,\n-there are two possibilities: a) make the indexee (the array's address)\n-a variable, or b) make the 'constant' index a variable. Thus:\n+A second solution is to force one of the \\*(Aqconstants\\*(Aq to be a variable \\-\\-\n+that is, unknown and un\\-optimizable at compile time. For arrays,\n+there are two possibilities: a) make the indexee (the array\\*(Aqs address)\n+a variable, or b) make the \\*(Aqconstant\\*(Aq index a variable. Thus:\n .Sp\n .Vb 3\n \\& extern type extern_array[];\n \\& extern_array[1] \\-\\->\n \\& { volatile type *t=extern_array; t[1] }\n .Ve\n .Sp\n@@ -3298,21 +3301,21 @@\n .Vb 3\n \\& extern long long extern_ll;\n \\& extern_ll \\-\\->\n \\& { volatile long long * local_ll=&extern_ll; *local_ll }\n .Ve\n .Sp\n A third method of dealing with this difficulty is to abandon\n-\\&'auto\\-import' for the offending symbol and mark it with\n+\\&\\*(Aqauto\\-import\\*(Aq for the offending symbol and mark it with\n \\&\\f(CW\\*(C`_\\|_declspec(dllimport)\\*(C'\\fR. However, in practice that\n-requires using compile-time #defines to indicate whether you are\n+requires using compile\\-time #defines to indicate whether you are\n building a DLL, building client code that will link to the DLL, or\n merely building/linking to a static library. In making the choice\n-between the various methods of resolving the 'direct address with\n-constant offset' problem, you should consider typical real-world usage:\n+between the various methods of resolving the \\*(Aqdirect address with\n+constant offset\\*(Aq problem, you should consider typical real\\-world usage:\n .Sp\n Original:\n .Sp\n .Vb 7\n \\& \\-\\-foo.h\n \\& extern int arr[];\n \\& \\-\\-foo.c\n@@ -3351,37 +3354,37 @@\n \\& \\-\\-foo.c\n \\& #include \"foo.h\"\n \\& void main(int argc, char **argv){\n \\& printf(\"%d\\en\",arr[1]);\n \\& }\n .Ve\n .Sp\n-A fourth way to avoid this problem is to re-code your\n+A fourth way to avoid this problem is to re\\-code your\n library to use a functional interface rather than a data interface\n for the offending variables (e.g. \\fBset_foo()\\fR and \\fBget_foo()\\fR accessor\n functions).\n .IP \\fB\\-\\-disable\\-auto\\-import\\fR 4\n .IX Item \"--disable-auto-import\"\n Do not attempt to do sophisticated linking of \\f(CW\\*(C`_symbol\\*(C'\\fR to\n \\&\\f(CW\\*(C`_\\|_imp_\\|_symbol\\*(C'\\fR for DATA imports from DLLs.\n [This option is specific to the i386 PE targeted port of the linker]\n .IP \\fB\\-\\-enable\\-runtime\\-pseudo\\-reloc\\fR 4\n .IX Item \"--enable-runtime-pseudo-reloc\"\n If your code contains expressions described in \\-\\-enable\\-auto\\-import section,\n-that is, DATA imports from DLL with non-zero offset, this switch will create\n-a vector of 'runtime pseudo relocations' which can be used by runtime\n+that is, DATA imports from DLL with non\\-zero offset, this switch will create\n+a vector of \\*(Aqruntime pseudo relocations\\*(Aq which can be used by runtime\n environment to adjust references to such data in your client code.\n [This option is specific to the i386 PE targeted port of the linker]\n .IP \\fB\\-\\-disable\\-runtime\\-pseudo\\-reloc\\fR 4\n .IX Item \"--disable-runtime-pseudo-reloc\"\n-Do not create pseudo relocations for non-zero offset DATA imports from DLLs.\n+Do not create pseudo relocations for non\\-zero offset DATA imports from DLLs.\n [This option is specific to the i386 PE targeted port of the linker]\n .IP \\fB\\-\\-enable\\-extra\\-pe\\-debug\\fR 4\n .IX Item \"--enable-extra-pe-debug\"\n-Show additional debug info related to auto-import symbol thunking.\n+Show additional debug info related to auto\\-import symbol thunking.\n [This option is specific to the i386 PE targeted port of the linker]\n .IP \\fB\\-\\-section\\-alignment\\fR 4\n .IX Item \"--section-alignment\"\n Sets the section alignment. Sections in memory will always begin at\n addresses which are a multiple of this number. Defaults to 0x1000.\n [This option is specific to the i386 PE targeted port of the linker]\n .IP \"\\fB\\-\\-stack\\fR \\fIreserve\\fR\" 4\n@@ -3624,71 +3627,71 @@\n .IP \\fB\\-N\\fR 4\n .IX Item \"-N\"\n .PD 0\n .IP \\fB\\-\\-omagic\\fR 4\n .IX Item \"--omagic\"\n .PD\n Mark the output as \\f(CW\\*(C`OMAGIC\\*(C'\\fR (0407) in the \\fIa.out\\fR header to\n-indicate that the text segment is not to be write-protected and\n+indicate that the text segment is not to be write\\-protected and\n shared. Since the text and data sections are both readable and\n writable, the data section is allocated immediately contiguous after\n the text segment. This is the oldest format for PDP11 executable\n programs and is the default for \\fBld\\fR on PDP11 Unix systems\n from the beginning through 2.11BSD.\n .IP \\fB\\-n\\fR 4\n .IX Item \"-n\"\n .PD 0\n .IP \\fB\\-\\-nmagic\\fR 4\n .IX Item \"--nmagic\"\n .PD\n Mark the output as \\f(CW\\*(C`NMAGIC\\*(C'\\fR (0410) in the \\fIa.out\\fR header to\n indicate that when the output file is executed, the text portion will\n-be read-only and shareable among all processes executing the same\n+be read\\-only and shareable among all processes executing the same\n file. This involves moving the data areas up to the first possible 8K\n byte page boundary following the end of the text. This option creates\n a \\fIpure executable\\fR format.\n .IP \\fB\\-z\\fR 4\n .IX Item \"-z\"\n .PD 0\n .IP \\fB\\-\\-imagic\\fR 4\n .IX Item \"--imagic\"\n .PD\n Mark the output as \\f(CW\\*(C`IMAGIC\\*(C'\\fR (0411) in the \\fIa.out\\fR header to\n indicate that when the output file is executed, the program text and\n data areas will be loaded into separate address spaces using the split\n instruction and data space feature of the memory management unit in\n larger models of the PDP11. This doubles the address space available\n-to the program. The text segment is again pure, write-protected, and\n+to the program. The text segment is again pure, write\\-protected, and\n shareable. The only difference in the output format between this\n option and the others, besides the magic number, is that both the text\n and data sections start at location 0. The \\fB\\-z\\fR option selected\n this format in 2.11BSD. This option creates a \\fIseparate\n executable\\fR format.\n .IP \\fB\\-\\-no\\-omagic\\fR 4\n .IX Item \"--no-omagic\"\n Equivalent to \\fB\\-\\-nmagic\\fR for pdp11\\-aout.\n .SH ENVIRONMENT\n .IX Header \"ENVIRONMENT\"\n You can change the behaviour of \\fBld\\fR with the environment variables\n \\&\\f(CW\\*(C`GNUTARGET\\*(C'\\fR,\n \\&\\f(CW\\*(C`LDEMULATION\\*(C'\\fR and \\f(CW\\*(C`COLLECT_NO_DEMANGLE\\*(C'\\fR.\n .PP\n-\\&\\f(CW\\*(C`GNUTARGET\\*(C'\\fR determines the input-file object format if you don't\n+\\&\\f(CW\\*(C`GNUTARGET\\*(C'\\fR determines the input\\-file object format if you don\\*(Aqt\n use \\fB\\-b\\fR (or its synonym \\fB\\-\\-format\\fR). Its value should be one\n of the BFD names for an input format. If there is no\n \\&\\f(CW\\*(C`GNUTARGET\\*(C'\\fR in the environment, \\fBld\\fR uses the natural format\n of the target. If \\f(CW\\*(C`GNUTARGET\\*(C'\\fR is set to \\f(CW\\*(C`default\\*(C'\\fR then BFD\n attempts to discover the input format by examining binary input files;\n this method often succeeds, but there are potential ambiguities, since\n there is no method of ensuring that the magic number used to specify\n-object-file formats is unique. However, the configuration procedure for\n+object\\-file formats is unique. However, the configuration procedure for\n BFD on each system places the conventional format for that system first\n-in the search-list, so ambiguities are resolved in favor of convention.\n+in the search\\-list, so ambiguities are resolved in favor of convention.\n .PP\n-\\&\\f(CW\\*(C`LDEMULATION\\*(C'\\fR determines the default emulation if you don't use the\n+\\&\\f(CW\\*(C`LDEMULATION\\*(C'\\fR determines the default emulation if you don\\*(Aqt use the\n \\&\\fB\\-m\\fR option. The emulation can affect various aspects of linker\n behaviour, particularly the default linker script. You can list the\n available emulations with the \\fB\\-\\-verbose\\fR or \\fB\\-V\\fR options. If\n the \\fB\\-m\\fR option is not used, and the \\f(CW\\*(C`LDEMULATION\\*(C'\\fR environment\n variable is not defined, the default emulation depends upon how the\n linker was configured.\n .PP\n@@ -3711,10 +3714,10 @@\n .SH COPYRIGHT\n .IX Header \"COPYRIGHT\"\n Copyright (c) 1991\\-2025 Free Software Foundation, Inc.\n .PP\n Permission is granted to copy, distribute and/or modify this document\n under the terms of the GNU Free Documentation License, Version 1.3\n or any later version published by the Free Software Foundation;\n-with no Invariant Sections, with no Front-Cover Texts, and with no\n-Back-Cover Texts. A copy of the license is included in the\n+with no Invariant Sections, with no Front\\-Cover Texts, and with no\n+Back\\-Cover Texts. A copy of the license is included in the\n section entitled \"GNU Free Documentation License\".\n"}]}]}]}]}]}