This is Info file ld.info, produced by Makeinfo-1.64 from the input file ./ld.texinfo. START-INFO-DIR-ENTRY * Ld: (ld). The GNU linker. END-INFO-DIR-ENTRY This file documents the GNU linker LD. Copyright (C) 1991, 92, 93, 94, 95, 96, 1997 Free Software Foundation, Inc. Permission is granted to make and distribute verbatim copies of this manual provided the copyright notice and this permission notice are preserved on all copies. Permission is granted to copy and distribute modified versions of this manual under the conditions for verbatim copying, provided also that the entire resulting derived work is distributed under the terms of a permission notice identical to this one. Permission is granted to copy and distribute translations of this manual into another language, under the above conditions for modified versions.  File: ld.info, Node: Option Commands, Prev: Version Script, Up: Commands Option Commands =============== The command language includes a number of other commands that you can use for specialized purposes. They are similar in purpose to command-line options. `CONSTRUCTORS' When linking using the `a.out' object file format, the linker uses an unusual set construct to support C++ global constructors and destructors. When linking object file formats which do not support arbitrary sections, such as `ECOFF' and `XCOFF', the linker will automatically recognize C++ global constructors and destructors by name. For these object file formats, the `CONSTRUCTORS' command tells the linker where this information should be placed. The `CONSTRUCTORS' command is ignored for other object file formats. The symbol `__CTOR_LIST__' marks the start of the global constructors, and the symbol `__DTOR_LIST' marks the end. The first word in the list is the number of entries, followed by the address of each constructor or destructor, followed by a zero word. The compiler must arrange to actually run the code. For these object file formats GNU C++ calls constructors from a subroutine `__main'; a call to `__main' is automatically inserted into the startup code for `main'. GNU C++ runs destructors either by using `atexit', or directly from the function `exit'. For object file formats such as `COFF' or `ELF' which support multiple sections, GNU C++ will normally arrange to put the addresses of global constructors and destructors into the `.ctors' and `.dtors' sections. Placing the following sequence into your linker script will build the sort of table which the GNU C++ runtime code expects to see. __CTOR_LIST__ = .; LONG((__CTOR_END__ - __CTOR_LIST__) / 4 - 2) *(.ctors) LONG(0) __CTOR_END__ = .; __DTOR_LIST__ = .; LONG((__DTOR_END__ - __DTOR_LIST__) / 4 - 2) *(.dtors) LONG(0) __DTOR_END__ = .; Normally the compiler and linker will handle these issues automatically, and you will not need to concern yourself with them. However, you may need to consider this if you are using C++ and writing your own linker scripts. `FLOAT' `NOFLOAT' These keywords were used in some older linkers to request a particular math subroutine library. `ld' doesn't use the keywords, assuming instead that any necessary subroutines are in libraries specified using the general mechanisms for linking to archives; but to permit the use of scripts that were written for the older linkers, the keywords `FLOAT' and `NOFLOAT' are accepted and ignored. `FORCE_COMMON_ALLOCATION' This command has the same effect as the `-d' command-line option: to make `ld' assign space to common symbols even if a relocatable output file is specified (`-r'). `INCLUDE FILENAME' Include the linker script FILENAME at this point. The file will be searched for in the current directory, and in any directory specified with the `-L' option. You can nest calls to `INCLUDE' up to 10 levels deep. `INPUT ( FILE, FILE, ... )' `INPUT ( FILE FILE ... )' Use this command to include binary input files in the link, without including them in a particular section definition. Specify the full name for each FILE, including `.a' if required. `ld' searches for each FILE through the archive-library search path, just as for files you specify on the command line. See the description of `-L' in *Note Command Line Options: Options. If you use `-lFILE', `ld' will transform the name to `libFILE.a' as with the command line argument `-l'. `GROUP ( FILE, FILE, ... )' `GROUP ( FILE FILE ... )' This command is like `INPUT', except that the named files should all be archives, and they are searched repeatedly until no new undefined references are created. See the description of `-(' in *Note Command Line Options: Options. `OUTPUT ( FILENAME )' Use this command to name the link output file FILENAME. The effect of `OUTPUT(FILENAME)' is identical to the effect of `-o FILENAME', which overrides it. You can use this command to supply a default output-file name other than `a.out'. `OUTPUT_ARCH ( BFDNAME )' Specify a particular output machine architecture, with one of the names used by the BFD back-end routines (*note BFD::.). This command is often unnecessary; the architecture is most often set implicitly by either the system BFD configuration or as a side effect of the `OUTPUT_FORMAT' command. `OUTPUT_FORMAT ( BFDNAME )' When `ld' is configured to support multiple object code formats, you can use this command to specify a particular output format. BFDNAME is one of the names used by the BFD back-end routines (*note BFD::.). The effect is identical to the effect of the `--oformat' command-line option. This selection affects only the output file; the related command `TARGET' affects primarily input files. `SEARCH_DIR ( PATH )' Add PATH to the list of paths where `ld' looks for archive libraries. `SEARCH_DIR(PATH)' has the same effect as `-LPATH' on the command line. `STARTUP ( FILENAME )' Ensure that FILENAME is the first input file used in the link process. `TARGET ( FORMAT )' When `ld' is configured to support multiple object code formats, you can use this command to change the input-file object code format (like the command-line option `-b' or its synonym `--format'). The argument FORMAT is one of the strings used by BFD to name binary formats. If `TARGET' is specified but `OUTPUT_FORMAT' is not, the last `TARGET' argument is also used as the default format for the `ld' output file. *Note BFD::. If you don't use the `TARGET' command, `ld' uses the value of the environment variable `GNUTARGET', if available, to select the output file format. If that variable is also absent, `ld' uses the default format configured for your machine in the BFD libraries. `NOCROSSREFS ( SECTION SECTION ... )' This command may be used to tell `ld' to issue an error about any references among certain sections. In certain types of programs, particularly on embedded systems, when one section is loaded into memory, another section will not be. Any direct references between the two sections would be errors. For example, it would be an error if code in one section called a function defined in the other section. The `NOCROSSREFS' command takes a list of section names. If `ld' detects any cross references between the sections, it reports an error and returns a non-zero exit status. The `NOCROSSREFS' command uses output section names, defined in the `SECTIONS' command. It does not use the names of input sections.  File: ld.info, Node: Machine Dependent, Next: BFD, Prev: Commands, Up: Top Machine Dependent Features ************************** `ld' has additional features on some platforms; the following sections describe them. Machines where `ld' has no additional functionality are not listed. * Menu: * H8/300:: `ld' and the H8/300 * i960:: `ld' and the Intel 960 family  File: ld.info, Node: H8/300, Next: i960, Up: Machine Dependent `ld' and the H8/300 =================== For the H8/300, `ld' can perform these global optimizations when you specify the `--relax' command-line option. *relaxing address modes* `ld' finds all `jsr' and `jmp' instructions whose targets are within eight bits, and turns them into eight-bit program-counter relative `bsr' and `bra' instructions, respectively. *synthesizing instructions* `ld' finds all `mov.b' instructions which use the sixteen-bit absolute address form, but refer to the top page of memory, and changes them to use the eight-bit address form. (That is: the linker turns `mov.b `@'AA:16' into `mov.b `@'AA:8' whenever the address AA is in the top page of memory).  File: ld.info, Node: i960, Prev: H8/300, Up: Machine Dependent `ld' and the Intel 960 family ============================= You can use the `-AARCHITECTURE' command line option to specify one of the two-letter names identifying members of the 960 family; the option specifies the desired output target, and warns of any incompatible instructions in the input files. It also modifies the linker's search strategy for archive libraries, to support the use of libraries specific to each particular architecture, by including in the search loop names suffixed with the string identifying the architecture. For example, if your `ld' command line included `-ACA' as well as `-ltry', the linker would look (in its built-in search paths, and in any paths you specify with `-L') for a library with the names try libtry.a tryca libtryca.a The first two possibilities would be considered in any event; the last two are due to the use of `-ACA'. You can meaningfully use `-A' more than once on a command line, since the 960 architecture family allows combination of target architectures; each use will add another pair of name variants to search for when `-l' specifies a library. `ld' supports the `--relax' option for the i960 family. If you specify `--relax', `ld' finds all `balx' and `calx' instructions whose targets are within 24 bits, and turns them into 24-bit program-counter relative `bal' and `cal' instructions, respectively. `ld' also turns `cal' instructions into `bal' instructions when it determines that the target subroutine is a leaf routine (that is, the target subroutine does not itself call any subroutines).  File: ld.info, Node: BFD, Next: Reporting Bugs, Prev: Machine Dependent, Up: Top BFD *** The linker accesses object and archive files using the BFD libraries. These libraries allow the linker to use the same routines to operate on object files whatever the object file format. A different object file format can be supported simply by creating a new BFD back end and adding it to the library. To conserve runtime memory, however, the linker and associated tools are usually configured to support only a subset of the object file formats available. You can use `objdump -i' (*note objdump: (binutils.info)objdump.) to list all the formats available for your configuration. As with most implementations, BFD is a compromise between several conflicting requirements. The major factor influencing BFD design was efficiency: any time used converting between formats is time which would not have been spent had BFD not been involved. This is partly offset by abstraction payback; since BFD simplifies applications and back ends, more time and care may be spent optimizing algorithms for a greater speed. One minor artifact of the BFD solution which you should bear in mind is the potential for information loss. There are two places where useful information can be lost using the BFD mechanism: during conversion and during output. *Note BFD information loss::. * Menu: * BFD outline:: How it works: an outline of BFD  File: ld.info, Node: BFD outline, Up: BFD How it works: an outline of BFD =============================== When an object file is opened, BFD subroutines automatically determine the format of the input object file. They then build a descriptor in memory with pointers to routines that will be used to access elements of the object file's data structures. As different information from the the object files is required, BFD reads from different sections of the file and processes them. For example, a very common operation for the linker is processing symbol tables. Each BFD back end provides a routine for converting between the object file's representation of symbols and an internal canonical format. When the linker asks for the symbol table of an object file, it calls through a memory pointer to the routine from the relevant BFD back end which reads and converts the table into a canonical form. The linker then operates upon the canonical form. When the link is finished and the linker writes the output file's symbol table, another BFD back end routine is called to take the newly created symbol table and convert it into the chosen output format. * Menu: * BFD information loss:: Information Loss * Canonical format:: The BFD canonical object-file format  File: ld.info, Node: BFD information loss, Next: Canonical format, Up: BFD outline Information Loss ---------------- *Information can be lost during output.* The output formats supported by BFD do not provide identical facilities, and information which can be described in one form has nowhere to go in another format. One example of this is alignment information in `b.out'. There is nowhere in an `a.out' format file to store alignment information on the contained data, so when a file is linked from `b.out' and an `a.out' image is produced, alignment information will not propagate to the output file. (The linker will still use the alignment information internally, so the link is performed correctly). Another example is COFF section names. COFF files may contain an unlimited number of sections, each one with a textual section name. If the target of the link is a format which does not have many sections (e.g., `a.out') or has sections without names (e.g., the Oasys format), the link cannot be done simply. You can circumvent this problem by describing the desired input-to-output section mapping with the linker command language. *Information can be lost during canonicalization.* The BFD internal canonical form of the external formats is not exhaustive; there are structures in input formats for which there is no direct representation internally. This means that the BFD back ends cannot maintain all possible data richness through the transformation between external to internal and back to external formats. This limitation is only a problem when an application reads one format and writes another. Each BFD back end is responsible for maintaining as much data as possible, and the internal BFD canonical form has structures which are opaque to the BFD core, and exported only to the back ends. When a file is read in one format, the canonical form is generated for BFD and the application. At the same time, the back end saves away any information which may otherwise be lost. If the data is then written back in the same format, the back end routine will be able to use the canonical form provided by the BFD core as well as the information it prepared earlier. Since there is a great deal of commonality between back ends, there is no information lost when linking or copying big endian COFF to little endian COFF, or `a.out' to `b.out'. When a mixture of formats is linked, the information is only lost from the files whose format differs from the destination.  File: ld.info, Node: Canonical format, Prev: BFD information loss, Up: BFD outline The BFD canonical object-file format ------------------------------------ The greatest potential for loss of information occurs when there is the least overlap between the information provided by the source format, that stored by the canonical format, and that needed by the destination format. A brief description of the canonical form may help you understand which kinds of data you can count on preserving across conversions. *files* Information stored on a per-file basis includes target machine architecture, particular implementation format type, a demand pageable bit, and a write protected bit. Information like Unix magic numbers is not stored here--only the magic numbers' meaning, so a `ZMAGIC' file would have both the demand pageable bit and the write protected text bit set. The byte order of the target is stored on a per-file basis, so that big- and little-endian object files may be used with one another. *sections* Each section in the input file contains the name of the section, the section's original address in the object file, size and alignment information, various flags, and pointers into other BFD data structures. *symbols* Each symbol contains a pointer to the information for the object file which originally defined it, its name, its value, and various flag bits. When a BFD back end reads in a symbol table, it relocates all symbols to make them relative to the base of the section where they were defined. Doing this ensures that each symbol points to its containing section. Each symbol also has a varying amount of hidden private data for the BFD back end. Since the symbol points to the original file, the private data format for that symbol is accessible. `ld' can operate on a collection of symbols of wildly different formats without problems. Normal global and simple local symbols are maintained on output, so an output file (no matter its format) will retain symbols pointing to functions and to global, static, and common variables. Some symbol information is not worth retaining; in `a.out', type information is stored in the symbol table as long symbol names. This information would be useless to most COFF debuggers; the linker has command line switches to allow users to throw it away. There is one word of type information within the symbol, so if the format supports symbol type information within symbols (for example, COFF, IEEE, Oasys) and the type is simple enough to fit within one word (nearly everything but aggregates), the information will be preserved. *relocation level* Each canonical BFD relocation record contains a pointer to the symbol to relocate to, the offset of the data to relocate, the section the data is in, and a pointer to a relocation type descriptor. Relocation is performed by passing messages through the relocation type descriptor and the symbol pointer. Therefore, relocations can be performed on output data using a relocation method that is only available in one of the input formats. For instance, Oasys provides a byte relocation format. A relocation record requesting this relocation type would point indirectly to a routine to perform this, so the relocation may be performed on a byte being written to a 68k COFF file, even though 68k COFF has no such relocation type. *line numbers* Object formats can contain, for debugging purposes, some form of mapping between symbols, source line numbers, and addresses in the output file. These addresses have to be relocated along with the symbol information. Each symbol with an associated list of line number records points to the first record of the list. The head of a line number list consists of a pointer to the symbol, which allows finding out the address of the function whose line number is being described. The rest of the list is made up of pairs: offsets into the section and line numbers. Any format which can simply derive this information can pass it successfully between formats (COFF, IEEE and Oasys).  File: ld.info, Node: Reporting Bugs, Next: MRI, Prev: BFD, Up: Top Reporting Bugs ************** Your bug reports play an essential role in making `ld' reliable. Reporting a bug may help you by bringing a solution to your problem, or it may not. But in any case the principal function of a bug report is to help the entire community by making the next version of `ld' work better. Bug reports are your contribution to the maintenance of `ld'. In order for a bug report to serve its purpose, you must include the information that enables us to fix the bug. * Menu: * Bug Criteria:: Have you found a bug? * Bug Reporting:: How to report bugs  File: ld.info, Node: Bug Criteria, Next: Bug Reporting, Up: Reporting Bugs Have you found a bug? ===================== If you are not sure whether you have found a bug, here are some guidelines: * If the linker gets a fatal signal, for any input whatever, that is a `ld' bug. Reliable linkers never crash. * If `ld' produces an error message for valid input, that is a bug. * If `ld' does not produce an error message for invalid input, that may be a bug. In the general case, the linker can not verify that object files are correct. * If you are an experienced user of linkers, your suggestions for improvement of `ld' are welcome in any case.  File: ld.info, Node: Bug Reporting, Prev: Bug Criteria, Up: Reporting Bugs How to report bugs ================== A number of companies and individuals offer support for GNU products. If you obtained `ld' from a support organization, we recommend you contact that organization first. You can find contact information for many support companies and individuals in the file `etc/SERVICE' in the GNU Emacs distribution. In any event, we also recommend that you send bug reports for `ld' to `bug-gnu-utils@prep.ai.mit.edu'. The fundamental principle of reporting bugs usefully is this: *report all the facts*. If you are not sure whether to state a fact or leave it out, state it! Often people omit facts because they think they know what causes the problem and assume that some details do not matter. Thus, you might assume that the name of a symbol you use in an example does not matter. Well, probably it does not, but one cannot be sure. Perhaps the bug is a stray memory reference which happens to fetch from the location where that name is stored in memory; perhaps, if the name were different, the contents of that location would fool the linker into doing the right thing despite the bug. Play it safe and give a specific, complete example. That is the easiest thing for you to do, and the most helpful. Keep in mind that the purpose of a bug report is to enable us to fix the bug if it is new to us. Therefore, always write your bug reports on the assumption that the bug has not been reported previously. Sometimes people give a few sketchy facts and ask, "Does this ring a bell?" Those bug reports are useless, and we urge everyone to *refuse to respond to them* except to chide the sender to report bugs properly. To enable us to fix the bug, you should include all these things: * The version of `ld'. `ld' announces it if you start it with the `--version' argument. Without this, we will not know whether there is any point in looking for the bug in the current version of `ld'. * Any patches you may have applied to the `ld' source, including any patches made to the `BFD' library. * The type of machine you are using, and the operating system name and version number. * What compiler (and its version) was used to compile `ld'--e.g. "`gcc-2.7'". * The command arguments you gave the linker to link your example and observe the bug. To guarantee you will not omit something important, list them all. A copy of the Makefile (or the output from make) is sufficient. If we were to try to guess the arguments, we would probably guess wrong and then we might not encounter the bug. * A complete input file, or set of input files, that will reproduce the bug. It is generally most helpful to send the actual object files, uuencoded if necessary to get them through the mail system. Making them available for anonymous FTP is not as good, but may be the only reasonable choice for large object files. If the source files were assembled using `gas' or compiled using `gcc', then it may be OK to send the source files rather than the object files. In this case, be sure to say exactly what version of `gas' or `gcc' was used to produce the object files. Also say how `gas' or `gcc' were configured. * A description of what behavior you observe that you believe is incorrect. For example, "It gets a fatal signal." Of course, if the bug is that `ld' gets a fatal signal, then we will certainly notice it. But if the bug is incorrect output, we might not notice unless it is glaringly wrong. You might as well not give us a chance to make a mistake. Even if the problem you experience is a fatal signal, you should still say so explicitly. Suppose something strange is going on, such as, your copy of `ld' is out of synch, or you have encountered a bug in the C library on your system. (This has happened!) Your copy might crash and ours would not. If you told us to expect a crash, then when ours fails to crash, we would know that the bug was not happening for us. If you had not told us to expect a crash, then we would not be able to draw any conclusion from our observations. * If you wish to suggest changes to the `ld' source, send us context diffs, as generated by `diff' with the `-u', `-c', or `-p' option. Always send diffs from the old file to the new file. If you even discuss something in the `ld' source, refer to it by context, not by line number. The line numbers in our development sources will not match those in your sources. Your line numbers would convey no useful information to us. Here are some things that are not necessary: * A description of the envelope of the bug. Often people who encounter a bug spend a lot of time investigating which changes to the input file will make the bug go away and which changes will not affect it. This is often time consuming and not very useful, because the way we will find the bug is by running a single example under the debugger with breakpoints, not by pure deduction from a series of examples. We recommend that you save your time for something else. Of course, if you can find a simpler example to report *instead* of the original one, that is a convenience for us. Errors in the output will be easier to spot, running under the debugger will take less time, and so on. However, simplification is not vital; if you do not want to do this, report the bug anyway and send us the entire test case you used. * A patch for the bug. A patch for the bug does help us if it is a good one. But do not omit the necessary information, such as the test case, on the assumption that a patch is all we need. We might see problems with your patch and decide to fix the problem another way, or we might not understand it at all. Sometimes with a program as complicated as `ld' it is very hard to construct an example that will make the program follow a certain path through the code. If you do not send us the example, we will not be able to construct one, so we will not be able to verify that the bug is fixed. And if we cannot understand what bug you are trying to fix, or why your patch should be an improvement, we will not install it. A test case will help us to understand. * A guess about what the bug is or what it depends on. Such guesses are usually wrong. Even we cannot guess right about such things without first using the debugger to find the facts.  File: ld.info, Node: MRI, Next: Index, Prev: Reporting Bugs, Up: Top MRI Compatible Script Files *************************** To aid users making the transition to GNU `ld' from the MRI linker, `ld' can use MRI compatible linker scripts as an alternative to the more general-purpose linker scripting language described in *Note Command Language: Commands. MRI compatible linker scripts have a much simpler command set than the scripting language otherwise used with `ld'. GNU `ld' supports the most commonly used MRI linker commands; these commands are described here. In general, MRI scripts aren't of much use with the `a.out' object file format, since it only has three sections and MRI scripts lack some features to make use of them. You can specify a file containing an MRI-compatible script using the `-c' command-line option. Each command in an MRI-compatible script occupies its own line; each command line starts with the keyword that identifies the command (though blank lines are also allowed for punctuation). If a line of an MRI-compatible script begins with an unrecognized keyword, `ld' issues a warning message, but continues processing the script. Lines beginning with `*' are comments. You can write these commands using all upper-case letters, or all lower case; for example, `chip' is the same as `CHIP'. The following list shows only the upper-case form of each command. `ABSOLUTE SECNAME' `ABSOLUTE SECNAME, SECNAME, ... SECNAME' Normally, `ld' includes in the output file all sections from all the input files. However, in an MRI-compatible script, you can use the `ABSOLUTE' command to restrict the sections that will be present in your output program. If the `ABSOLUTE' command is used at all in a script, then only the sections named explicitly in `ABSOLUTE' commands will appear in the linker output. You can still use other input sections (whatever you select on the command line, or using `LOAD') to resolve addresses in the output file. `ALIAS OUT-SECNAME, IN-SECNAME' Use this command to place the data from input section IN-SECNAME in a section called OUT-SECNAME in the linker output file. IN-SECNAME may be an integer. `ALIGN SECNAME = EXPRESSION' Align the section called SECNAME to EXPRESSION. The EXPRESSION should be a power of two. `BASE EXPRESSION' Use the value of EXPRESSION as the lowest address (other than absolute addresses) in the output file. `CHIP EXPRESSION' `CHIP EXPRESSION, EXPRESSION' This command does nothing; it is accepted only for compatibility. `END' This command does nothing whatever; it's only accepted for compatibility. `FORMAT OUTPUT-FORMAT' Similar to the `OUTPUT_FORMAT' command in the more general linker language, but restricted to one of these output formats: 1. S-records, if OUTPUT-FORMAT is `S' 2. IEEE, if OUTPUT-FORMAT is `IEEE' 3. COFF (the `coff-m68k' variant in BFD), if OUTPUT-FORMAT is `COFF' `LIST ANYTHING...' Print (to the standard output file) a link map, as produced by the `ld' command-line option `-M'. The keyword `LIST' may be followed by anything on the same line, with no change in its effect. `LOAD FILENAME' `LOAD FILENAME, FILENAME, ... FILENAME' Include one or more object file FILENAME in the link; this has the same effect as specifying FILENAME directly on the `ld' command line. `NAME OUTPUT-NAME' OUTPUT-NAME is the name for the program produced by `ld'; the MRI-compatible command `NAME' is equivalent to the command-line option `-o' or the general script language command `OUTPUT'. `ORDER SECNAME, SECNAME, ... SECNAME' `ORDER SECNAME SECNAME SECNAME' Normally, `ld' orders the sections in its output file in the order in which they first appear in the input files. In an MRI-compatible script, you can override this ordering with the `ORDER' command. The sections you list with `ORDER' will appear first in your output file, in the order specified. `PUBLIC NAME=EXPRESSION' `PUBLIC NAME,EXPRESSION' `PUBLIC NAME EXPRESSION' Supply a value (EXPRESSION) for external symbol NAME used in the linker input files. `SECT SECNAME, EXPRESSION' `SECT SECNAME=EXPRESSION' `SECT SECNAME EXPRESSION' You can use any of these three forms of the `SECT' command to specify the start address (EXPRESSION) for section SECNAME. If you have more than one `SECT' statement for the same SECNAME, only the *first* sets the start address.