Using and Porting GNU CC

Richard M. Stallman

last updated 16 Dec 1992

for version 2.3

(preliminary draft, which will change)

Copyright © 1988, 1989, 1992 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 sections entitled “GNU General Public License” and “Boycott” are included exactly as in the original, and provided 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, except that the sections entitled “GNU General Public License” and “Boycott”, and this permission notice, may be included in translations approved by the Free Software Foundation instead of in the original English.

GNU GENERAL PUBLIC LICENSE

Version 2, June 1991

Copyright © 1989, 1991 Free Software Foundation, Inc.
675 Mass Ave, Cambridge, MA 02139, USA

Everyone is permitted to copy and distribute verbatim copies
of this license document, but changing it is not allowed.

Preamble

The licenses for most software are designed to take away your freedom to share and change it. By contrast, the GNU General Public License is intended to guarantee your freedom to share and change free software—to make sure the software is free for all its users. This General Public License applies to most of the Free Software Foundation’s software and to any other program whose authors commit to using it. (Some other Free Software Foundation software is covered by the GNU Library General Public License instead.) You can apply it to your programs, too.

When we speak of free software, we are referring to freedom, not price. Our General Public Licenses are designed to make sure that you have the freedom to distribute copies of free software (and charge for this service if you wish), that you receive source code or can get it if you want it, that you can change the software or use pieces of it in new free programs; and that you know you can do these things.

To protect your rights, we need to make restrictions that forbid anyone to deny you these rights or to ask you to surrender the rights. These restrictions translate to certain responsibilities for you if you distribute copies of the software, or if you modify it.

For example, if you distribute copies of such a program, whether gratis or for a fee, you must give the recipients all the rights that you have. You must make sure that they, too, receive or can get the source code. And you must show them these terms so they know their rights.

We protect your rights with two steps: (1) copyright the software, and (2) offer you this license which gives you legal permission to copy, distribute and/or modify the software.

Also, for each author’s protection and ours, we want to make certain that everyone understands that there is no warranty for this free software. If the software is modified by someone else and passed on, we want its recipients to know that what they have is not the original, so that any problems introduced by others will not reflect on the original authors’ reputations.

Finally, any free program is threatened constantly by software patents. We wish to avoid the danger that redistributors of a free program will individually obtain patent licenses, in effect making the program proprietary. To prevent this, we have made it clear that any patent must be licensed for everyone’s free use or not licensed at all.

The precise terms and conditions for copying, distribution and modification follow.

1. This License applies to any program or other work which contains a notice placed by the copyright holder saying it may be distributed under the terms of this General Public License. The “Program”, below, refers to any such program or work, and a “work based on the Program” means either the Program or any derivative work under copyright law: that is to say, a work containing the Program or a portion of it, either verbatim or with modifications and/or translated into another language. (Hereinafter, translation is included without limitation in the term “modification”.) Each licensee is addressed as “you”.

   Activities other than copying, distribution and modification are not covered by this License; they are outside its scope. The act of running the Program is not restricted, and the output from the Program is covered only if its contents constitute a work based on the Program (independent of having been made by running the Program). Whether that is true depends on what the Program does.

2. You may copy and distribute verbatim copies of the Program’s source code as you receive it, in any medium, provided that you conspicuously and appropriately publish on each copy an appropriate copyright notice and disclaimer of warranty; keep intact all the notices that refer to this License and to the absence of any warranty; and give any other recipients of the Program a copy of this License along with the Program.

   You may charge a fee for the physical act of transferring a copy, and you may at your option offer warranty protection in exchange for a fee.

3. You may modify your copy or copies of the Program or any portion of it, thus forming a work based on the Program, and copy and distribute such modifications or work under the terms of Section 1 above, provided that you also meet all of these conditions:
   1. You must cause the modified files to carry prominent notices stating that you changed the files and the date of any change.
   2. You must cause any work that you distribute or publish, that in whole or in part contains or is derived from the Program or any part thereof, to be licensed as a whole at no charge to all third parties under the terms of this License.
   3. If the modified program normally reads commands interactively when run, you must cause it, when started running for such interactive use in the most ordinary way, to print or display an announcement including an appropriate copyright notice and a notice that there is no warranty (or else, saying that you provide a warranty) and that users may redistribute the program under these conditions, and telling the user how to view a copy of this License. (Exception: if the Program itself is interactive but does not normally print such an announcement, your work based on the Program is not required to print an announcement.)

   These requirements apply to the modified work as a whole. If identifiable sections of that work are not derived from the Program, and can be reasonably considered independent and separate works in themselves, then this License, and its terms, do not apply to those sections when you distribute them as separate works. But when you distribute the same sections as part of a whole which is a work based on the Program, the distribution of the whole must be on the terms of this License, whose permissions for other licensees extend to the entire whole, and thus to each and every part regardless of who wrote it.

   Thus, it is not the intent of this section to claim rights or contest your rights to work written entirely by you; rather, the intent is to exercise the right to control the distribution of derivative or collective works based on the Program.

   In addition, mere aggregation of another work not based on the Program with the Program (or with a work based on the Program) on a volume of a storage or distribution medium does not bring the other work under the scope of this License.

4. You may copy and distribute the Program (or a work based on it, under Section 2) in object code or executable form under the terms of Sections 1 and 2 above provided that you also do one of the following:
   1. Accompany it with the complete corresponding machine-readable source code, which must be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange; or,
   2. Accompany it with a written offer, valid for at least three years, to give any third party, for a charge no more than your cost of physically performing source distribution, a complete machine-readable copy of the corresponding source code, to be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange; or,
   3. Accompany it with the information you received as to the offer to distribute corresponding source code. (This alternative is allowed only for noncommercial distribution and only if you received the program in object code or executable form with such an offer, in accord with Subsection b above.)

   The source code for a work means the preferred form of the work for making modifications to it. For an executable work, complete source code means all the source code for all modules it contains, plus any associated interface definition files, plus the scripts used to control compilation and installation of the executable. However, as a special exception, the source code distributed need not include anything that is normally distributed (in either source or binary form) with the major components (compiler, kernel, and so on) of the operating system on which the executable runs, unless that component itself accompanies the executable.

   If distribution of executable or object code is made by offering access to copy from a designated place, then offering equivalent access to copy the source code from the same place counts as distribution of the source code, even though third parties are not compelled to copy the source along with the object code.

5. You may not copy, modify, sublicense, or distribute the Program except as expressly provided under this License. Any attempt otherwise to copy, modify, sublicense or distribute the Program is void, and will automatically terminate your rights under this License. However, parties who have received copies, or rights, from you under this License will not have their licenses terminated so long as such parties remain in full compliance.
6. You are not required to accept this License, since you have not signed it. However, nothing else grants you permission to modify or distribute the Program or its derivative works. These actions are prohibited by law if you do not accept this License. Therefore, by modifying or distributing the Program (or any work based on the Program), you indicate your acceptance of this License to do so, and all its terms and conditions for copying, distributing or modifying the Program or works based on it.
7. Each time you redistribute the Program (or any work based on the Program), the recipient automatically receives a license from the original licensor to copy, distribute or modify the Program subject to these terms and conditions. You may not impose any further restrictions on the recipients’ exercise of the rights granted herein. You are not responsible for enforcing compliance by third parties to this License.
8. If, as a consequence of a court judgment or allegation of patent infringement or for any other reason (not limited to patent issues), conditions are imposed on you (whether by court order, agreement or otherwise) that contradict the conditions of this License, they do not excuse you from the conditions of this License. If you cannot distribute so as to satisfy simultaneously your obligations under this License and any other pertinent obligations, then as a consequence you may not distribute the Program at all. For example, if a patent license would not permit royalty-free redistribution of the Program by all those who receive copies directly or indirectly through you, then the only way you could satisfy both it and this License would be to refrain entirely from distribution of the Program.

   If any portion of this section is held invalid or unenforceable under any particular circumstance, the balance of the section is intended to apply and the section as a whole is intended to apply in other circumstances.

   It is not the purpose of this section to induce you to infringe any patents or other property right claims or to contest validity of any such claims; this section has the sole purpose of protecting the integrity of the free software distribution system, which is implemented by public license practices. Many people have made generous contributions to the wide range of software distributed through that system in reliance on consistent application of that system; it is up to the author/donor to decide if he or she is willing to distribute software through any other system and a licensee cannot impose that choice.

   This section is intended to make thoroughly clear what is believed to be a consequence of the rest of this License.

9. If the distribution and/or use of the Program is restricted in certain countries either by patents or by copyrighted interfaces, the original copyright holder who places the Program under this License may add an explicit geographical distribution limitation excluding those countries, so that distribution is permitted only in or among countries not thus excluded. In such case, this License incorporates the limitation as if written in the body of this License.
10. The Free Software Foundation may publish revised and/or new versions of the General Public License from time to time. Such new versions will be similar in spirit to the present version, but may differ in detail to address new problems or concerns.

    Each version is given a distinguishing version number. If the Program specifies a version number of this License which applies to it and “any later version”, you have the option of following the terms and conditions either of that version or of any later version published by the Free Software Foundation. If the Program does not specify a version number of this License, you may choose any version ever published by the Free Software Foundation.

11. If you wish to incorporate parts of the Program into other free programs whose distribution conditions are different, write to the author to ask for permission. For software which is copyrighted by the Free Software Foundation, write to the Free Software Foundation; we sometimes make exceptions for this. Our decision will be guided by the two goals of preserving the free status of all derivatives of our free software and of promoting the sharing and reuse of software generally.
12. BECAUSE THE PROGRAM IS LICENSED FREE OF CHARGE, THERE IS NO WARRANTY FOR THE PROGRAM, TO THE EXTENT PERMITTED BY APPLICABLE LAW. EXCEPT WHEN OTHERWISE STATED IN WRITING THE COPYRIGHT HOLDERS AND/OR OTHER PARTIES PROVIDE THE PROGRAM “AS IS” WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE ENTIRE RISK AS TO THE QUALITY AND PERFORMANCE OF THE PROGRAM IS WITH YOU. SHOULD THE PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF ALL NECESSARY SERVICING, REPAIR OR CORRECTION.
13. IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MAY MODIFY AND/OR REDISTRIBUTE THE PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES, INCLUDING ANY GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING OUT OF THE USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT LIMITED TO LOSS OF DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY YOU OR THIRD PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER PROGRAMS), EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES.

How to Apply These Terms to Your New Programs

If you develop a new program, and you want it to be of the greatest possible use to the public, the best way to achieve this is to make it free software which everyone can redistribute and change under these terms.

To do so, attach the following notices to the program. It is safest to attach them to the start of each source file to most effectively convey the exclusion of warranty; and each file should have at least the “copyright” line and a pointer to where the full notice is found.

one line to give the program's name and an idea of what it does.
Copyright (C) 19yy  name of author

This program is free software; you can redistribute it and/or
modify it under the terms of the GNU General Public License
as published by the Free Software Foundation; either version 2
of the License, or (at your option) any later version.

This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
GNU General Public License for more details.

You should have received a copy of the GNU General Public License
along with this program; if not, write to the Free Software
Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.

Also add information on how to contact you by electronic and paper mail.

If the program is interactive, make it output a short notice like this when it starts in an interactive mode:

Gnomovision version 69, Copyright (C) 19yy name of author
Gnomovision comes with ABSOLUTELY NO WARRANTY; for details
type `show w'.  This is free software, and you are welcome
to redistribute it under certain conditions; type `show c'
for details.

The hypothetical commands ‘show w’ and ‘show c’ should show the appropriate parts of the General Public License. Of course, the commands you use may be called something other than ‘show w’ and ‘show c’; they could even be mouse-clicks or menu items—whatever suits your program.

You should also get your employer (if you work as a programmer) or your school, if any, to sign a “copyright disclaimer” for the program, if necessary. Here is a sample; alter the names:

Yoyodyne, Inc., hereby disclaims all copyright
interest in the program `Gnomovision'
(which makes passes at compilers) written 
by James Hacker.

signature of Ty Coon, 1 April 1989
Ty Coon, President of Vice

This General Public License does not permit incorporating your program into proprietary programs. If your program is a subroutine library, you may consider it more useful to permit linking proprietary applications with the library. If this is what you want to do, use the GNU Library General Public License instead of this License.

Contributors to GNU CC

In addition to Richard Stallman, several people have written parts of GNU CC.

• The idea of using RTL and some of the optimization ideas came from the program PO written at the University of Arizona by Jack Davidson and Christopher Fraser. See “Register Allocation and Exhaustive Peephole Optimization”, Software Practice and Experience 14 (9), Sept. 1984, 857-866.
• Paul Rubin wrote most of the preprocessor.
• Leonard Tower wrote parts of the parser, RTL generator, and RTL definitions, and of the Vax machine description.
• Ted Lemon wrote parts of the RTL reader and printer.
• Jim Wilson implemented loop strength reduction and some other loop optimizations.
• Nobuyuki Hikichi of Software Research Associates, Tokyo, contributed the support for the Sony NEWS machine.
• Charles LaBrec contributed the support for the Integrated Solutions 68020 system.
• Michael Tiemann of Cygnus Support wrote the front end for C++, as well as the support for inline functions and instruction scheduling. Also the descriptions of the National Semiconductor 32000 series cpu, the SPARC cpu and part of the Motorola 88000 cpu.
• Jan Stein of the Chalmers Computer Society provided support for Genix, as well as part of the 32000 machine description.
• Randy Smith finished the Sun FPA support.
• Robert Brown implemented the support for Encore 32000 systems.
• David Kashtan of SRI adapted GNU CC to the Vomit-Making System (VMS).
• Alex Crain provided changes for the 3b1.
• Greg Satz and Chris Hanson assisted in making GNU CC work on HP-UX for the 9000 series 300.
• William Schelter did most of the work on the Intel 80386 support.
• Christopher Smith did the port for Convex machines.
• Paul Petersen wrote the machine description for the Alliant FX/8.
• Alain Lichnewsky ported GNU CC to the Mips cpu.
• Devon Bowen, Dale Wiles and Kevin Zachmann ported GNU CC to the Tahoe.
• Jonathan Stone wrote the machine description for the Pyramid computer.
• Gary Miller ported GNU CC to Charles River Data Systems machines.
• Richard Kenner of the New York University Ultracomputer Research Laboratory wrote the machine descriptions for the AMD 29000, the DEC Alpha, the IBM RT PC, and the IBM RS/6000 as well as the support for instruction attributes. He also made changes to better support RISC processors including changes to common subexpression elimination, strength reduction, function calling sequence handling, and condition code support, in addition to generalizing the code for frame pointer elimination.
• Richard Kenner and Michael Tiemann jointly developed reorg.c, the delay slot scheduler.
• Mike Meissner and Tom Wood of Data General finished the port to the Motorola 88000.
• Masanobu Yuhara of Fujitsu Laboratories implemented the machine description for the Tron architecture (specifically, the Gmicro).
• NeXT, Inc. donated the front end that supports the Objective C language.
• James van Artsdalen wrote the code that makes efficient use of the Intel 80387 register stack.
• Mike Meissner at the Open Software Foundation finished the port to the MIPS cpu, including adding ECOFF debug support.
• Ron Guilmette implemented the protoize and unprotoize tools, the support for Dwarf symbolic debugging information, and much of the support for System V Release 4. He has also worked heavily on the Intel 386 and 860 support.
• Torbjorn Granlund of the Swedish Institute of Computer Science implemented multiply-by-constant optimization and better long long support, and improved leaf function register allocation.
• Mike Stump implemented the support for Elxsi 64 bit CPU.

1 Protect Your Freedom—Fight “Look And Feel”

This section is a political message from the League for Programming Freedom to the users of GNU CC. It is included here as an expression of support for the League on the part of the Free Software Foundation.

Apple and Lotus are trying to create a new form of legal monopoly: a copyright on a class of user interfaces. These monopolies would cause serious problems for users and developers of computer software and systems. Xerox, too, has tried to make a monopoly for itself on window systems; their suit against Apple was thrown out on a technicality, but Xerox has not said anything to indicate it wouldn’t try again.

Until a few years ago, the law seemed clear: no one could restrict others from using a user interface; programmers were free to implement any interface they chose. Imitating interfaces, sometimes with changes, was standard practice in the computer field. The interfaces we know evolved gradually in this way; for example, the Macintosh user interface drew ideas from the Xerox interface, which in turn drew on work done at Stanford and SRI. 1-2-3 imitated VisiCalc, and dBase imitated a database program from JPL.

Most computer companies, and nearly all computer users, were happy with this state of affairs. The companies that are suing say it does not offer “enough incentive” to develop their products, but they must have considered it “enough” when they made their decision to do so. It seems they are not satisfied with the opportunity to continue to compete in the marketplace—not even with a head start.

If companies like Xerox, Lotus, and Apple are permitted to make law through the courts, the precedent will hobble the software industry:

• Gratuitous incompatibilities will burden users. Imagine if each car manufacturer had to arrange the pedals in a different order.
• Software will become and remain more expensive. Users will be “locked in” to proprietary interfaces, for which there is no real competition.
• Large companies have an unfair advantage wherever lawsuits become commonplace. Since they can easily afford to sue, they can intimidate small companies with threats even when they don’t really have a case.
• User interface improvements will come slower, since incremental evolution through creative imitation will no longer be permitted.
• Even Apple, etc., will find it harder to make improvements if they can no longer adapt the good ideas that others introduce, for fear of weakening their own legal positions. Some users suggest that this stagnation may already have started.
• If you use GNU software, you might find it of some concern that user interface copyright will make it hard for the Free Software Foundation to develop programs compatible with the interfaces that you already know.

To protect our freedom from lawsuits like these, a group of programmers and users have formed a new grass-roots political organization, the League for Programming Freedom.

The purpose of the League is to oppose new monopolistic practices such as user-interface copyright and software patents; it calls for a return to the legal policies of the recent past, in which these practices were not allowed. The League is not concerned with free software as an issue, and not affiliated with the Free Software Foundation.

The League’s membership rolls include John McCarthy, inventor of Lisp, Marvin Minsky, founder of the Artificial Intelligence lab, Guy L. Steele, Jr., author of well-known books on Lisp and C, as well as Richard Stallman, the developer of GNU CC. Please join and add your name to the list. Membership dues in the League are $42 per year for programmers, managers and professionals; $10.50 for students; $21 for others.

The League needs both activist members and members who only pay their dues.

To join, or for more information, phone (617) 243-4091 or write to:

League for Programming Freedom
1 Kendall Square #143
P.O. Box 9171
Cambridge, MA 02139

You can also send electronic mail to league@prep.ai.mit.edu.

Here are some suggestions from the League for things you can do to protect your freedom to write programs:

• Don’t buy from Xerox, Lotus or Apple. Buy from their competitors or from the defendants they are suing.
• Don’t develop software to work with the systems made by these companies.
• Port your existing software to competing systems, so that you encourage users to switch.
• Write letters to company presidents to let them know their conduct is unacceptable.
• Tell your friends and colleagues about this issue and how it threatens to ruin the computer industry.
• Above all, don’t work for the look-and-feel plaintiffs, and don’t accept contracts from them.
• Write to Congress to explain the importance of this issue.

  House Subcommittee on Intellectual Property
  2137 Rayburn Bldg
  Washington, DC 20515
  
  Senate Subcommittee on Patents, Trademarks and Copyrights
  United States Senate
  Washington, DC 20510
  

  (These committees have received lots of mail already; let’s give them even more.)

Express your opinion! You can make a difference.

2 Known Causes of Trouble with GNU CC

This section describes known problems that affect users of GNU CC. Most of these are not GNU CC bugs per se—if they were, we would fix them. But the result for a user may be like the result of a bug.

Some of these problems are due to bugs in other software, some are missing features that are too much work to add, and some are places where people’s opinions differ as to what is best.

2.1 Actual Bugs We Haven’t Fixed Yet

• Loop unrolling doesn’t work properly for certain C++ programs. This is because of difficulty in updating the debugging information within the loop being unrolled. We plan to revamp the representation of debugging information so that this will work properly, but we have not done this in version 2.3 because we don’t want to delay it any further.

2.2 Installation Problems

This is a list of problems (and some apparent problems which don’t really mean anything is wrong) that show up during installation of GNU CC.

• On certain systems, defining certain environment variables such as CC can interfere with the functioning of make.
• If you encounter seemingly strange errors when trying to build the compiler in a directory other than the source directory, it could be because you have previously configured the compiler in the source directory. Make sure you have done all the necessary preparations. @xref{Other Dir}.
• In previous versions of GNU CC, the gcc driver program looked for as and ld in various places such as files beginning with ‘/usr/local/lib/gcc-’. GNU CC version 2 looks for them in the directory ‘/usr/local/lib/gcc-lib/target/version’.

  Thus, to use a version of as or ld that is not the system default, for example gas or GNU ld, you must put them in that directory (or make links to them from that directory).

• Some commands executed when making the compiler may fail (return a non-zero status) and be ignored by make. These failures, which are often due to files that were not found, are expected, and can safely be ignored.
• It is normal to have warnings in compiling certain files about unreachable code and about enumeration type clashes. These files’ names begin with ‘insn-’.
• Sometimes make recompiles parts of the compiler when installing the compiler. In one case, this was traced down to a bug in make. Either ignore the problem or switch to GNU Make.
• On some 386 systems, building the compiler never finishes because enquire hangs due to a hardware problem in the motherboard—it reports floating point exceptions to the kernel incorrectly. You can install GNU CC except for ‘float.h’ by patching out the command to run enquire. You may also be able to fix the problem for real by getting a replacement motherboard. This problem was observed in Revision E of the Micronics motherboard, and is fixed in Revision F.
• On some 386 systems, GNU CC crashes trying to compile ‘enquire.c’. This happens on machines that don’t have a 387 FPU chip. On 386 machines, the system kernel is supposed to emulate the 387 when you don’t have one. The crash is due to a bug in the emulator.

  One of these systems is the Unix from Interactive Systems: 386/ix. On this system, an alternate emulator is provided, and it does work. To use it, execute this command as super-user:

  ln /etc/emulator.rel1 /etc/emulator
  

  and then reboot the system. (The default emulator file remains present under the name ‘emulator.dflt’.)

  If you have such a problem on the SCO system, try using ‘/etc/emulator.att’.

  Another system which has this problem is Esix. We don’t know whether it has an alternate emulator that works.

• Sometimes on a Sun 4 you may observe a crash in the program genflags or genoutput while building GNU CC. This is said to be due to a bug in sh. You can probably get around it by running genflags or genoutput manually and then retrying the make.
• If you use the 1.31 version of the MIPS assembler (such as was shipped with Ultrix 3.1), you will need to use the -fno-delayed-branch switch when optimizing floating point code. Otherwise, the assembler will complain when the GCC compiler fills a branch delay slot with a floating point instruction, such as add.d.
• Users have reported some problems with version 2.0 of the MIPS compiler tools that were shipped with Ultrix 4.1. Version 2.10 which came with Ultrix 4.2 seems to work fine.
• Some versions of the MIPS linker will issue an assertion failure when linking code that uses alloca against shared libraries on RISC-OS 5.0, and DEC’s OSF/1 systems. This is a bug in the linker, that is supposed to be fixed in future revisions. To protect against this, GCC passes ‘-non_shared’ to the linker unless you pass an explicit ‘-shared’ or ‘-call_shared’ switch.
• On System V release 3, you may get this error message while linking:

  ld fatal: failed to write symbol name something 
   in strings table for file whatever
  

  This indicates that the disk is full or your ULIMIT won’t allow the file to be as large as it needs to be.

• On HP 9000 series 300 or 400 running HP-UX release 8.0, there is a bug in the assembler that must be fixed before GNU CC can be built. This bug manifests itself during the first stage of compilation, while building ‘libgcc2.a’:

  _floatdisf
  cc1: warning: `-g' option not supported on this version of GCC
  cc1: warning: `-g1' option not supported on this version of GCC
  ./gcc: Internal compiler error: program as got fatal signal 11
  

  A patched version of the assembler is available by anonymous ftp from altdorf.ai.mit.edu as the file ‘archive/cph/hpux-8.0-assembler’. If you have HP software support, the patch can also be obtained directly from HP, as described in the following note:

  This is the patched assembler, to patch SR#1653-010439, where the assembler aborts on floating point constants.

  The bug is not really in the assembler, but in the shared library version of the function “cvtnum(3c)”. The bug on “cvtnum(3c)” is SR#4701-078451. Anyway, the attached assembler uses the archive library version of “cvtnum(3c)” and thus does not exhibit the bug.

  This patch is also known as PHCO_0800.

• Another assembler problem on the HP PA results in an error message like this while compiling part of ‘libgcc2.a’:

  as: /usr/tmp/cca08196.s @line#30 [err#1060]
    Argument 1 or 3 in FARG upper
           - lookahead = RTNVAL=GR
  

  This happens because HP changed the assembler syntax after system release 8.02. GNU CC assumes the newer syntax; if your assembler wants the older syntax, comment out this line in the file ‘pa1-hpux.h’:

  #define HP_FP_ARG_DESCRIPTOR_REVERSED
  

• Some versions of the Pyramid C compiler are reported to be unable to compile GNU CC. You must use an older version of GNU CC for bootstrapping. One indication of this problem is if you get a crash when GNU CC compiles the function muldi3 in file ‘libgcc2.c’.

  You may be able to succeed by getting GNU CC version 1, installing it, and using it to compile GNU CC version 2. The bug in the Pyramid C compiler does not seem to affect GNU CC version 1.

• On the Tower models 4n0 and 6n0, by default a process is not allowed to have more than one megabyte of memory. GNU CC cannot compile itself (or many other programs) with ‘-O’ in that much memory.

  To solve this problem, reconfigure the kernel adding the following line to the configuration file:

  MAXUMEM = 4096
  

• On the Altos 3068, programs compiled with GNU CC won’t work unless you fix a kernel bug. This happens using system versions V.2.2 1.0gT1 and V.2.2 1.0e and perhaps later versions as well. See the file ‘README.ALTOS’.
• You will get several sorts of compilation and linking errors on the we32k if you don’t follow the special instructions. @xref{WE32K Install}.

2.3 Cross-Compiler Problems

• Cross compilation can run into trouble for certain machines because some target machines’ assemblers require floating point numbers to be written as integer constants in certain contexts.

  The compiler writes these integer constants by examining the floating point value as an integer and printing that integer, because this is simple to write and independent of the details of the floating point representation. But this does not work if the compiler is running on a different machine with an incompatible floating point format, or even a different byte-ordering.

  In addition, correct constant folding of floating point values requires representing them in the target machine’s format. (The C standard does not quite require this, but in practice it is the only way to win.)

  It is now possible to overcome these problems by defining macros such as REAL_VALUE_TYPE. But doing so is a substantial amount of work for each target machine. @xref{Cross-compilation}.

• At present, the program ‘mips-tfile’ which adds debug support to object files on MIPS systems does not work in a cross compile environment.

2.4 Interoperation

This section lists various difficulties encountered in using GNU C or GNU C++ together with other compilers or with the assemblers, linkers, libraries and debuggers on certain systems.

• GNU C normally compiles functions to return small structures and unions in registers. Most other compilers arrange to return them just like larger structures and unions. This can lead to trouble when you link together code compiled by different compilers. To avoid the problem, you can use the option ‘-fpcc-struct-return’ when compiling with GNU CC.
• GNU C++ does not do name mangling in the same way as other C++ compilers. This means that object files compiled with one compiler cannot be used with another.

  GNU C++ also uses different techniques for arranging virtual function tables and the layout of class instances. In general, therefore, linking code compiled with different C++ compilers does not work.

• Older GDB versions sometimes fail to read the output of GNU CC version 2. If you have trouble, get GDB version 4.4 or later.
• DBX rejects some files produced by GNU CC, though it accepts similar constructs in output from PCC. Until someone can supply a coherent description of what is valid DBX input and what is not, there is nothing I can do about these problems. You are on your own.
• The GNU assembler (GAS) does not support PIC. To generate PIC code, you must use some other assembler, such as ‘/bin/as’.
• On some BSD systems including some versions of Ultrix, use of profiling causes static variable destructors (currently used only in C++) not to be run.
• Use of ‘-I/usr/include’ may cause trouble.

  Many systems come with header files that won’t work with GNU CC unless corrected by fixincludes. The corrected header files go in a new directory; GNU CC searches this directory before ‘/usr/include’. If you use ‘-I/usr/include’, this tells GNU CC to search ‘/usr/include’ earlier on, before the corrected headers. The result is that you get the uncorrected header files.

  Instead, you should use these options:

  -I/usr/local/lib/gcc-lib/target/version/include -I/usr/include
  

• On a Sparc, GNU CC aligns all values of type double on an 8-byte boundary, and it expects every double to be so aligned. The Sun compiler usually gives double values 8-byte alignment, with one exception: function arguments of type double may not be aligned.

  As a result, if a function compiled with Sun CC takes the address of an argument of type double and passes this pointer of type double * to a function compiled with GNU CC, dereferencing the pointer may cause a fatal signal.

  One way to solve this problem is to compile your entire program with GNU CC. Another solution is to modify the function that is compiled with Sun CC to copy the argument into a local variable; local variables are always properly aligned. A third solution is to modify the function that uses the pointer to dereference it via the following function access_double instead of directly with ‘*’:

  inline double
  access_double (double *unaligned_ptr)
  {
    union d2i { double d; int i[2]; };
  
    union d2i *p = (union d2i *) unaligned_ptr;
    union d2i u;
  
    u.i[0] = p->i[0];
    u.i[1] = p->i[1];
  
    return u.d;
  }
  

  Storing into the pointer can be done likewise with the same union.

• On a Sun, linking using GNU CC fails to find a shared library and reports that the library doesn’t exist at all.

  This happens if you are using the GNU linker, because it does only static linking and looks only for unshared libraries. If you have a shared library with no unshared counterpart, the GNU linker won’t find anything.

  We hope to make a linker which supports Sun shared libraries, but please don’t ask when it will be finished—we don’t know.

• Sun forgot to include a static version of ‘libdl.a’ with some versions of SunOS (mainly 4.1). This results in undefined symbols when linking static binaries (that is, if you use ‘-static’). If you see undefined symbols _dlclose, _dlsym or _dlopen when linking, compile and link against the file ‘mit/util/misc/dlsym.c’ from the MIT version of X windows.
• On the HP PA machine, ADB sometimes fails to work on functions compiled with GNU CC. Specifically, it fails to work on functions that use alloca or variable-size arrays. This is because GNU CC doesn’t generate HP-UX unwind descriptors for such functions. It may even be impossible to generate them.
• Debugging (‘-g’) is not supported on the HP PA machine, unless you use the preliminary GNU tools (@pxref{Installation}).
• The HP-UX linker has a bug which can cause programs which make use of const variables to fail in unusual ways. If your program makes use of global const variables, we suggest you compile with the following additional options:

  -Dconst="" -D__const="" -D__const__="" -fwritable-strings
  

  This will force the const variables into the DATA subspace which will avoid the linker bug.

  Another option one might use to work around this problem is ‘-mkernel’. ‘-mkernel’ changes how the address of variables is computed to a sequence less likely to tickle the HP-UX linker bug.

  We hope to work around this problem in GNU CC 2.4, if HP does not fix it.

• Taking the address of a label may generate errors from the HP-UX PA assembler. GAS for the PA does not have this problem.
• GNU CC produced code will not yet link against HP-UX 8.0 shared libraries. We expect to fix this problem in GNU CC 2.4.
• The current version of the assembler (‘/bin/as’) for the RS/6000 has certain problems that prevent the ‘-g’ option in GCC from working.

  IBM has produced a fixed version of the assembler. The replacement assembler is not a standard component of either AIX 3.1.5 or AIX 3.2, but is expected to become standard in a future distribution. This assembler is available from IBM as APAR IX22829. Yet more bugs have been fixed in a newer assembler, which will shortly be available as APAR IX26107. See the file ‘README.RS6000’ for more details on these assemblers.

• On the IBM RS/6000, compiling code of the form

  extern int foo;
  
  … foo …
  
  static int foo;
  

  will cause the linker to report an undefined symbol foo. Although this behavior differs from most other systems, it is not a bug because redefining an extern variable as static is undefined in ANSI C.

• On VMS, GAS versions 1.38.1 and earlier may cause spurious warning messages from the linker. These warning messages complain of mismatched psect attributes. You can ignore them. @xref{VMS Install}.
• On NewsOS version 3, if you include both ‘stddef.h’ and ‘sys/types.h’, you get an error because there are two typedefs of size_t. You should change ‘sys/types.h’ by adding these lines around the definition of size_t:

  #ifndef _SIZE_T
  #define _SIZE_T
  actual typedef here
  #endif
  

• On the Alliant, the system’s own convention for returning structures and unions is unusual, and is not compatible with GNU CC no matter what options are used.
• On the IBM RT PC, the MetaWare HighC compiler (hc) uses yet another convention for structure and union returning. Use ‘-mhc-struct-return’ to tell GNU CC to use a convention compatible with it.
• On Ultrix, the Fortran compiler expects registers 2 through 5 to be saved by function calls. However, the C compiler uses conventions compatible with BSD Unix: registers 2 through 5 may be clobbered by function calls.

  GNU CC uses the same convention as the Ultrix C compiler. You can use these options to produce code compatible with the Fortran compiler:

  -fcall-saved-r2 -fcall-saved-r3 -fcall-saved-r4 -fcall-saved-r5
  

• On the WE32k, you may find that programs compiled with GNU CC do not work with the standard shared C ilbrary. You may need to link with the ordinary C compiler. If you do so, you must specify the following options:

  -L/usr/local/lib/gcc-lib/we32k-att-sysv/2.3 -lgcc -lc_s
  

  The first specifies where to find the library ‘libgcc.a’ specified with the ‘-lgcc’ option.

  GNU CC does linking by invoking ld, just as cc does, and there is no reason why it should matter which compilation program you use to invoke ld. If someone tracks this problem down, it can probably be fixed easily.

2.5 Incompatibilities of GNU CC

There are several noteworthy incompatibilities between GNU C and most existing (non-ANSI) versions of C. The ‘-traditional’ option eliminates many of these incompatibilities, but not all, by telling GNU C to behave like the other C compilers.

• GNU CC normally makes string constants read-only. If several identical-looking string constants are used, GNU CC stores only one copy of the string.

  One consequence is that you cannot call mktemp with a string constant argument. The function mktemp always alters the string its argument points to.

  Another consequence is that sscanf does not work on some systems when passed a string constant as its format control string or input. This is because sscanf incorrectly tries to write into the string constant. Likewise fscanf and scanf.

  The best solution to these problems is to change the program to use char-array variables with initialization strings for these purposes instead of string constants. But if this is not possible, you can use the ‘-fwritable-strings’ flag, which directs GNU CC to handle string constants the same way most C compilers do. ‘-traditional’ also has this effect, among others.

• -2147483648 is positive.

  This is because 2147483648 cannot fit in the type int, so (following the ANSI C rules) its data type is unsigned long int. Negating this value yields 2147483648 again.

• GNU CC does not substitute macro arguments when they appear inside of string constants. For example, the following macro in GNU CC

  #define foo(a) "a"
  

  will produce output "a" regardless of what the argument a is.

  The ‘-traditional’ option directs GNU CC to handle such cases (among others) in the old-fashioned (non-ANSI) fashion.

• When you use setjmp and longjmp, the only automatic variables guaranteed to remain valid are those declared volatile. This is a consequence of automatic register allocation. Consider this function:

  jmp_buf j;
  
  foo ()
  {
    int a, b;
  
    a = fun1 ();
    if (setjmp (j))
      return a;
  
    a = fun2 ();
    /* longjmp (j) may occur in fun3. */
    return a + fun3 ();
  }
  

  Here a may or may not be restored to its first value when the longjmp occurs. If a is allocated in a register, then its first value is restored; otherwise, it keeps the last value stored in it.

  If you use the ‘-W’ option with the ‘-O’ option, you will get a warning when GNU CC thinks such a problem might be possible.

  The ‘-traditional’ option directs GNU C to put variables in the stack by default, rather than in registers, in functions that call setjmp. This results in the behavior found in traditional C compilers.

• Programs that use preprocessor directives in the middle of macro arguments do not work with GNU CC. For example, a program like this will not work:

  foobar (
  #define luser
          hack)
  

  ANSI C does not permit such a construct. It would make sense to support it when ‘-traditional’ is used, but it is too much work to implement.

• Declarations of external variables and functions within a block apply only to the block containing the declaration. In other words, they have the same scope as any other declaration in the same place.

  In some other C compilers, a extern declaration affects all the rest of the file even if it happens within a block.

  The ‘-traditional’ option directs GNU C to treat all extern declarations as global, like traditional compilers.

• In traditional C, you can combine long, etc., with a typedef name, as shown here:

  typedef int foo;
  typedef long foo bar;
  

  In ANSI C, this is not allowed: long and other type modifiers require an explicit int. Because this criterion is expressed by Bison grammar rules rather than C code, the ‘-traditional’ flag cannot alter it.

• PCC allows typedef names to be used as function parameters. The difficulty described immediately above applies here too.
• PCC allows whitespace in the middle of compound assignment operators such as ‘+=’. GNU CC, following the ANSI standard, does not allow this. The difficulty described immediately above applies here too.
• GNU CC complains about unterminated character constants inside of preprocessor conditionals that fail. Some programs have English comments enclosed in conditionals that are guaranteed to fail; if these comments contain apostrophes, GNU CC will probably report an error. For example, this code would produce an error:

  #if 0
  You can't expect this to work.
  #endif
  

  The best solution to such a problem is to put the text into an actual C comment delimited by ‘/*…*/’. However, ‘-traditional’ suppresses these error messages.

• Many user programs contain the declaration ‘long time ();’. In the past, the system header files on many systems did not actually declare time, so it did not matter what type your program declared it to return. But in systems with ANSI C headers, time is declared to return time_t, and if that is not the same as long, then ‘long time ();’ is erroneous.

  The solution is to change your program to use time_t as the return type of time.

• When compiling functions that return float, PCC converts it to a double. GNU CC actually returns a float. If you are concerned with PCC compatibility, you should declare your functions to return double; you might as well say what you mean.
• When compiling functions that return structures or unions, GNU CC output code normally uses a method different from that used on most versions of Unix. As a result, code compiled with GNU CC cannot call a structure-returning function compiled with PCC, and vice versa.

  The method used by GNU CC is as follows: a structure or union which is 1, 2, 4 or 8 bytes long is returned like a scalar. A structure or union with any other size is stored into an address supplied by the caller (usually in a special, fixed register, but on some machines it is passed on the stack). The machine-description macros STRUCT_VALUE and STRUCT_INCOMING_VALUE tell GNU CC where to pass this address.

  By contrast, PCC on most target machines returns structures and unions of any size by copying the data into an area of static storage, and then returning the address of that storage as if it were a pointer value. The caller must copy the data from that memory area to the place where the value is wanted. GNU CC does not use this method because it is slower and nonreentrant.

  On some newer machines, PCC uses a reentrant convention for all structure and union returning. GNU CC on most of these machines uses a compatible convention when returning structures and unions in memory, but still returns small structures and unions in registers.

  You can tell GNU CC to use a compatible convention for all structure and union returning with the option ‘-fpcc-struct-return’.

2.6 Disappointments and Misunderstandings

These problems are perhaps regrettable, but we don’t know any practical way around them.

• Certain local variables aren’t recognized by debuggers when you compile with optimization.

  This occurs because sometimes GNU CC optimizes the variable out of existence. There is no way to tell the debugger how to compute the value such a variable “would have had”, and it is not clear that would be desirable anyway. So GNU CC simply does not mention the eliminated variable when it writes debugging information.

  You have to expect a certain amount of disagreement between the executable and your source code, when you use optimization.

• Users often think it is a bug when GNU CC reports an error for code like this:

  int foo (struct mumble *);
  
  struct mumble { … };
  
  int foo (struct mumble *x)
  { … }
  

  This code really is erroneous, because the scope of struct mumble in the prototype is limited to the argument list containing it. It does not refer to the struct mumble defined with file scope immediately below—they are two unrelated types with similar names in different scopes.

  But in the definition of foo, the file-scope type is used because that is available to be inherited. Thus, the definition and the prototype do not match, and you get an error.

  This behavior may seem silly, but it’s what the ANSI standard specifies. It is easy enough for you to make your code work by moving the definition of struct mumble above the prototype. It’s not worth being incompatible with ANSI C just to avoid an error for the example shown above.

• Accesses to bitfields even in volatile objects works by accessing larger objects, such as a byte or a word. You cannot rely on what size of object is accessed in order to read or write the bitfield; it may even vary for a given bitfield according to the precise usage.

  If you care about controlling the amount of memory that is accessed, use volatile but do not use bitfields.

• On 68000 systems, you can get paradoxical results if you test the precise values of floating point numbers. For example, you can find that a floating point value which is not a NaN is not equal to itself. This results from the fact that the the floating point registers hold a few more bits of precision than fit in a double in memory. Compiled code moves values between memory and floating point registers at its convenience, and moving them into memory truncates them.

  You can partially avoid this problem by using the option ‘-ffloat-store’ (@pxref{Optimize Options}).

• On the MIPS, variable argument functions using ‘varargs.h’ cannot have a floating point value for the first argument. The reason for this is that in the absence of a prototype in scope, if the first argument is a floating point, it is passed in a floating point register, rather than an integer resgister.

  If the code is rewritten to use the ANSI standard ‘stdarg.h’ method of variable arguments, and the prototype is in scope at the time of the call, everything will work fine.

2.7 Caveats of using protoize

The conversion programs protoize and unprotoize can sometimes change a source file in a way that won’t work unless you rearrange it.

• protoize can insert references to a type name or type tag before the definition, or in a file where they are not defined.

  If this happens, compiler error messages should show you where the new references are, so fixing the file by hand is straightforward.

• There are some C constructs which protoize cannot figure out. For example, it can’t determine argument types for declaring a pointer-to-function variable; this you must do by hand. protoize inserts a comment containing ‘???’ each time it finds such a variable; so you can find all such variables by searching for this string. ANSI C does not require declaring the argument types of pointer-to-function types.
• Using unprotoize can easily introduce bugs. If the program relied on prototypes to bring about conversion of arguments, these conversions will not take place in the program without prototypes. One case in which you can be sure unprotoize is safe is when you are removing prototypes that were made with protoize; if the program worked before without any prototypes, it will work again without them.

  You can find all the places where this problem might occur by compiling the program with the ‘-Wconversion’ option. It prints a warning whenever an argument is converted.

• Both conversion programs can be confused if there are macro calls in and around the text to be converted. In other words, the standard syntax for a declaration or definition must not result from expanding a macro. This problem is inherent in the design of C and cannot be fixed. If only a few functions have confusing macro calls, you can easily convert them manually.
• protoize cannot get the argument types for a function whose definition was not actually compiled due to preprocessor conditionals. When this happens, protoize changes nothing in regard to such a function. protoize tries to detect such instances and warn about them.

  You can generally work around this problem by using protoize step by step, each time specifying a different set of ‘-D’ options for compilation, until all of the functions have been converted. There is no automatic way to verify that you have got them all, however.

• Confusion may result if there is an occasion to convert a function declaration or definition in a region of source code where there is more than one formal parameter list present. Thus, attempts to convert code containing multiple (conditionally compiled) versions of a single function header (in the same vicinity) may not produce the desired (or expected) results.

  If you plan on converting source files which contain such code, it is recommended that you first make sure that each conditionally compiled region of source code which contains an alternative function header also contains at least one additional follower token (past the final right parenthesis of the function header). This should circumvent the problem.

• unprotoize can become confused when trying to convert a function definition or declaration which contains a declaration for a pointer-to-function formal argument which has the same name as the function being defined or declared. We recommand you avoid such choices of formal parameter names.
• You might also want to correct some of the indentation by hand and break long lines. (The conversion programs don’t write lines longer than eighty characters in any case.)

2.8 Certain Changes We Don’t Want to Make

This section lists changes that people frequently request, but which we do not make because we think GNU CC is better without them.

• Checking the number and type of arguments to a function which has an old-fashioned definition and no prototype.

  Such a feature would work only occasionally—only for calls that appear in the same file as the called function, following the definition. The only way to check all calls reliably is to add a prototype for the function. But adding a prototype eliminates the motivation for this feature. So the feature is not worthwhile.

• Warning about using an expression whose type is signed as a shift count.

  Shift count operands are probably signed more often than unsigned. Warning about this would cause far more annoyance than good.

• Warning about assigning a signed value to an unsigned variable.

  Such assignments must be very common; warning about them would cause more annoyance than good.

• Warning about unreachable code.

  It’s very common to have unreachable code in machine-generated programs. For example, this happens normally in some files of GNU C itself.

• Warning when a non-void function value is ignored.

  Coming as I do from a Lisp background, I balk at the idea that there is something dangerous about discarding a value. There are functions that return values which some callers may find useful; it makes no sense to clutter the program with a cast to void whenever the value isn’t useful.

• Assuming (for optimization) that the address of an external symbol is never zero.

  This assumption is false on certain systems when ‘#pragma weak’ is used.

• Making ‘-fshort-enums’ the default.

  This would cause storage layout to be incompatible with most other C compilers. And it doesn’t seem very important, given that you can get the same result in other ways. The case where it matters most is when the enumeration-valued object is inside a structure, and in that case you can specify a field width explicitly.

• Making bitfields unsigned by default on particular machines where “the ABI standard” says to do so.

  The ANSI C standard leaves it up to the implementation whether a bitfield declared plain int is signed or not. This in effect creates two alternative dialects of C.

  The GNU C compiler supports both dialects; you can specify the dialect you want with the option ‘-fsigned-bitfields’ or ‘-funsigned-bitfields’. However, this leaves open the question of which dialect to use by default.

  Currently, the preferred dialect makes plain bitfields signed, because this is simplest. Since int is the same as signed int in every other context, it is cleanest for them to be the same in bitfields as well.

  Some computer manufacturers have published Application Binary Interface standards which specify that plain bitfields should be unsigned. It is a mistake, however, to say anything about this issue in an ABI. This is because the handling of plain bitfields distinguishes two dialects of C. Both dialects are meaningful on every type of machine. Whether a particular object file was compiled using signed bitfields or unsigned is of no concern to other object files, even if they access the same bitfields in the same data structures.

  A given program is written in one or the other of these two dialects. The program stands a chance to work on most any machine if it is compiled with the proper dialect. It is unlikely to work at all if compiled with the wrong dialect.

  Many users appreciate the GNU C compiler because it provides an environment that is uniform across machines. These users would be inconvenienced if the compiler treated plain bitfields differently on certain machines.

  Occasionally users write programs intended only for a particular machine type. On these occasions, the users would benefit if the GNU C compiler were to support by default the same dialect as the other compilers on that machine. But such applications are rare. And users writing a program to run on more than one type of machine cannot possibly benefit from this kind of compatibility.

  This is why GNU CC does and will treat plain bitfields in the same fashion on all types of machines (by default).

  There are some arguments for making bitfields unsigned by default on all machines. If, for example, this becomes a universal de facto standard, it would make sense for GNU CC to go along with it. This is something to be considered in the future.

  (Of course, users strongly concerned about portability should indicate explicitly in each bitfield whether it is signed or not. In this way, they write programs which have the same meaning in both C dialects.)

• Undefining __STDC__ when ‘-ansi’ is not used.

  Currently, GNU CC defines __STDC__ as long as you don’t use ‘-traditional’. This provides good results in practice.

  Programmers normally use conditionals on __STDC__ to ask whether it is safe to use certain features of ANSI C, such as function prototypes or ANSI token concatenation. Since plain ‘gcc’ supports all the features of ANSI C, the correct answer to these questions is “yes”.

  Some users try to use __STDC__ to check for the availability of certain library facilities. This is actually incorrect usage in an ANSI C program, because the ANSI C standard says that a conforming freestanding implementation should define __STDC__ even though it does not have the library facilities. ‘gcc -ansi -pedantic’ is a conforming freestanding implementation, and it is therefore required to define __STDC__, even though it does not come with an ANSI C library.

  Sometimes people say that defining __STDC__ in a compiler that does not completely conform to the ANSI C standard somehow violates the standard. This is illogical. The standard is a standard for compilers that claim to support ANSI C, such as ‘gcc -ansi’—not for other compilers such as plain ‘gcc’. Whatever the ANSI C standard says is relevant to the design of plain ‘gcc’ without ‘-ansi’ only for pragmatic reasons, not as a requirement.

• Undefining __STDC__ in C++.

  Programs written to compile with C++-to-C translators get the value of __STDC__ that goes with the C compiler that is subsequently used. These programs must test __STDC__ to determine what kind of C preprocessor that compiler uses: whether they should concatenate tokens in the ANSI C fashion or in the traditional fashion.

  These programs work properly with GNU C++ if __STDC__ is defined. They would not work otherwise.

  In addition, many header files are written to provide prototypes in ANSI C but not in traditional C. Many of these header files can work without change in C++ provided __STDC__ is defined. If __STDC__ is not defined, they will all fail, and will all need to be changed to test explicitly for C++ as well.

• Deleting “empty” loops.

  GNU CC does not delete “empty” loops because the most likely reason you would put one in a program is to have a delay. Deleting them will not make real programs run any faster, so it would be pointless.

  It would be different if optimization of a nonempty loop could produce an empty one. But this generally can’t happen.

3 Reporting Bugs

Your bug reports play an essential role in making GNU CC reliable.

When you encounter a problem, the first thing to do is to see if it is already known. See section Known Causes of Trouble with GNU CC. If it isn’t known, then you should report the problem.

Reporting a bug may help you by bringing a solution to your problem, or it may not. (If it does not, look in the service directory; see How To Get Help with GNU CC.) In any case, the principal function of a bug report is to help the entire community by making the next version of GNU CC work better. Bug reports are your contribution to the maintenance of GNU CC.

In order for a bug report to serve its purpose, you must include the information that makes for fixing the bug.

3.1 Have You Found a Bug?

If you are not sure whether you have found a bug, here are some guidelines:

• If the compiler gets a fatal signal, for any input whatever, that is a compiler bug. Reliable compilers never crash.
• If the compiler produces invalid assembly code, for any input whatever (except an asm statement), that is a compiler bug, unless the compiler reports errors (not just warnings) which would ordinarily prevent the assembler from being run.
• If the compiler produces valid assembly code that does not correctly execute the input source code, that is a compiler bug.

  However, you must double-check to make sure, because you may have run into an incompatibility between GNU C and traditional C (see section Incompatibilities of GNU CC). These incompatibilities might be considered bugs, but they are inescapable consequences of valuable features.

  Or you may have a program whose behavior is undefined, which happened by chance to give the desired results with another C or C++ compiler.

  For example, in many nonoptimizing compilers, you can write ‘x;’ at the end of a function instead of ‘return x;’, with the same results. But the value of the function is undefined if return is omitted; it is not a bug when GNU CC produces different results.

  Problems often result from expressions with two increment operators, as in f (*p++, *p++). Your previous compiler might have interpreted that expression the way you intended; GNU CC might interpret it another way. Neither compiler is wrong. The bug is in your code.

  After you have localized the error to a single source line, it should be easy to check for these things. If your program is correct and well defined, you have found a compiler bug.

• If the compiler produces an error message for valid input, that is a compiler bug.
• If the compiler does not produce an error message for invalid input, that is a compiler bug. However, you should note that your idea of “invalid input” might be my idea of “an extension” or “support for traditional practice”.
• If you are an experienced user of C or C++ compilers, your suggestions for improvement of GNU CC or GNU C++ are welcome in any case.

3.2 Where to Report Bugs

Send bug reports for GNU C to one of these addresses:

bug-gcc@prep.ai.mit.edu
{ucbvax|mit-eddie|uunet}!prep.ai.mit.edu!bug-gcc

Send bug reports for GNU C++ to one of these addresses:

bug-g++@prep.ai.mit.edu
{ucbvax|mit-eddie|uunet}!prep.ai.mit.edu!bug-g++

Do not send bug reports to ‘help-gcc’, or to the newsgroup ‘gnu.gcc.help’. Most users of GNU CC do not want to receive bug reports. Those that do, have asked to be on ‘bug-gcc’ and/or ‘bug-g++’.

The mailing lists ‘bug-gcc’ and ‘bug-g++’ both have newsgroups which serve as repeaters: ‘gnu.gcc.bug’ and ‘gnu.g++.bug’. Each mailing list and its newsgroup carry exactly the same messages.

Often people think of posting bug reports to the newsgroup instead of mailing them. This appears to work, but it has one problem which can be crucial: a newsgroup posting does not contain a mail path back to the sender. Thus, if maintainers need more information, they may be unable to reach you. For this reason, you should always send bug reports by mail to the proper mailing list.

As a last resort, send bug reports on paper to:

GNU Compiler Bugs
Free Software Foundation
675 Mass Ave
Cambridge, MA 02139

3.3 How to Report Bugs

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 they conclude that some details don’t matter. Thus, you might assume that the name of the variable you use in an example does not matter. Well, probably it doesn’t, 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 compiler 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 someone to fix the bug if it is not known. It isn’t very important what happens if the bug is already known. Therefore, always write your bug reports on the assumption that the bug is not known.

Sometimes people give a few sketchy facts and ask, “Does this ring a bell?” This cannot help us fix a bug, so it is basically useless. We respond by asking for enough details to enable us to investigate. You might as well expedite matters by sending them to begin with.

Try to make your bug report self-contained. If we have to ask you for more information, it is best if you include all the previous information in your response, as well as the information that was missing.

To enable someone to investigate the bug, you should include all these things:

• The version of GNU CC. You can get this by running it with the ‘-v’ option.

  Without this, we won’t know whether there is any point in looking for the bug in the current version of GNU CC.

• A complete input file that will reproduce the bug. If the bug is in the C preprocessor, send a source file and any header files that it requires. If the bug is in the compiler proper (‘cc1’), run your source file through the C preprocessor by doing ‘gcc -E sourcefile > outfile’, then include the contents of outfile in the bug report. (When you do this, use the same ‘-I’, ‘-D’ or ‘-U’ options that you used in actual compilation.)

  A single statement is not enough of an example. In order to compile it, it must be embedded in a complete file of compiler input; and the bug might depend on the details of how this is done.

  Without a real example one can compile, all anyone can do about your bug report is wish you luck. It would be futile to try to guess how to provoke the bug. For example, bugs in register allocation and reloading frequently depend on every little detail of the function they happen in.

  Even if the input file that fails comes from a GNU program, you should still send the complete test case. Don’t ask the GNU CC maintainers to do the extra work of obtaining the program in question—they are all overworked as it is. Also, the problem may depend on what is in the header files on your system; it is unreliable for the GNU CC maintainers to try the problem with the header files available to them. By sending CPP output, you can eliminate this source of uncertainty.

• The command arguments you gave GNU CC or GNU C++ to compile that example and observe the bug. For example, did you use ‘-O’? To guarantee you won’t omit something important, list all the options.

  If we were to try to guess the arguments, we would probably guess wrong and then we would not encounter the bug.

• The type of machine you are using, and the operating system name and version number.
• The operands you gave to the configure command when you installed the compiler.
• A complete list of any modifications you have made to the compiler source. (We don’t promise to investigate the bug unless it happens in an unmodified compiler. But if you’ve made modifications and don’t tell us, then you are sending us on a wild goose chase.)

  Be precise about these changes. A description in English is not enough—send a context diff for them.

  Adding files of your own (such as a machine description for a machine we don’t support) is a modification of the compiler source.

• Details of any other deviations from the standard procedure for installing GNU CC.
• A description of what behavior you observe that you believe is incorrect. For example, “The compiler gets a fatal signal,” or, “The assembler instruction at line 208 in the output is incorrect.”

  Of course, if the bug is that the compiler gets a fatal signal, then one can’t miss it. But if the bug is incorrect output, the maintainer might not notice unless it is glaringly wrong. None of us has time to study all the assembler code from a 50-line C program just on the chance that one instruction might be wrong. We need you to do this part!

  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 the compiler 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 the copy here would not. If you said to expect a crash, then when the compiler here fails to crash, we would know that the bug was not happening. If you don’t say to expect a crash, then we would not know whether the bug was happening. We would not be able to draw any conclusion from our observations.

  If the problem is a diagnostic when compiling GNU CC with some other compiler, say whether it is a warning or an error.

  Often the observed symptom is incorrect output when your program is run. Sad to say, this is not enough information unless the program is short and simple. None of us has time to study a large program to figure out how it would work if compiled correctly, much less which line of it was compiled wrong. So you will have to do that. Tell us which source line it is, and what incorrect result happens when that line is executed. A person who understands the program can find this as easily as finding a bug in the program itself.

• If you send examples of assembler code output from GNU CC or GNU C++, please use ‘-g’ when you make them. The debugging information includes source line numbers which are essential for correlating the output with the input.
• If you wish to mention something in the GNU CC source, refer to it by context, not by line number.

  The line numbers in the development sources don’t match those in your sources. Your line numbers would convey no useful information to the maintainers.

• Additional information from a debugger might enable someone to find a problem on a machine which he does not have available. However, you need to think when you collect this information if you want it to have any chance of being useful.

  For example, many people send just a backtrace, but that is never useful by itself. A simple backtrace with arguments conveys little about GNU CC because the compiler is largely data-driven; the same functions are called over and over for different RTL insns, doing different things depending on the details of the insn.

  Most of the arguments listed in the backtrace are useless because they are pointers to RTL list structure. The numeric values of the pointers, which the debugger prints in the backtrace, have no significance whatever; all that matters is the contents of the objects they point to (and most of the contents are other such pointers).

  In addition, most compiler passes consist of one or more loops that scan the RTL insn sequence. The most vital piece of information about such a loop—which insn it has reached—is usually in a local variable, not in an argument.

  What you need to provide in addition to a backtrace are the values of the local variables for several stack frames up. When a local variable or an argument is an RTX, first print its value and then use the GDB command pr to print the RTL expression that it points to. (If GDB doesn’t run on your machine, use your debugger to call the function debug_rtx with the RTX as an argument.) In general, whenever a variable is a pointer, its value is no use without the data it points to.

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. You might as well 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. Errors in the output will be easier to spot, running under the debugger will take less time, etc. Most GNU CC bugs involve just one function, so the most straightforward way to simplify an example is to delete all the function definitions except the one where the bug occurs. Those earlier in the file may be replaced by external declarations if the crucial function depends on them. (Exception: inline functions may affect compilation of functions defined later in the file.)

  However, simplification is not vital; if you don’t want to do this, report the bug anyway and send the entire test case you used.

• In particular, some people insert conditionals ‘#ifdef BUG’ around a statement which, if removed, makes the bug not happen. These are just clutter; we won’t pay any attention to them anyway. Besides, you should send us cpp output, and that can’t have conditionals.
• A patch for the bug.

  A patch for the bug is useful if it is a good one. But don’t 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 GNU CC it is very hard to construct an example that will make the program follow a certain path through the code. If you don’t send the example, we won’t be able to construct one, so we won’t be able to verify that the bug is fixed.

  And if we can’t understand what bug you are trying to fix, or why your patch should be an improvement, we won’t install it. A test case will help us to understand.

  See section Sending Patches for GNU CC, for guidelines on how to make it easy for us to understand and install your patches.

• A guess about what the bug is or what it depends on.

  Such guesses are usually wrong. Even I can’t guess right about such things without first using the debugger to find the facts.

• A core dump file.

  We have no way of examining a core dump for your type of machine unless we have an identical system—and if we do have one, we should be able to reproduce the crash ourselves.

3.4 Sending Patches for GNU CC

If you would like to write bug fixes or improvements for the GNU C compiler, that is very helpful. When you send your changes, please follow these guidelines to avoid causing extra work for us in studying the patches.

If you don’t follow these guidelines, your information might still be useful, but using it will take extra work. Maintaining GNU C is a lot of work in the best of circumstances, and we can’t keep up unless you do your best to help.

• Send an explanation with your changes of what problem they fix or what improvement they bring about. For a bug fix, just include a copy of the bug report, and explain why the change fixes the bug.

  (Referring to a bug report is not as good as including it, because then we will have to look it up, and we have probably already deleted it if we’ve already fixed the bug.)

• Always include a proper bug report for the problem you think you have fixed. We need to convince ourselves that the change is right before installing it. Even if it is right, we might have trouble judging it if we don’t have a way to reproduce the problem.
• Include all the comments that are appropriate to help people reading the source in the future understand why this change was needed.
• Don’t mix together changes made for different reasons. Send them individually.

  If you make two changes for separate reasons, then we might not want to install them both. We might want to install just one. If you send them all jumbled together in a single set of diffs, we have to do extra work to disentangle them—to figure out which parts of the change serve which purpose. If we don’t have time for this, we might have to ignore your changes entirely.

  If you send each change as soon as you have written it, with its own explanation, then the two changes never get tangled up, and we can consider each one properly without any extra work to disentangle them.

  Ideally, each change you send should be impossible to subdivide into parts that we might want to consider separately, because each of its parts gets its motivation from the other parts.

• Send each change as soon as that change is finished. Sometimes people think they are helping us by accumulating many changes to send them all together. As explained above, this is absolutely the worst thing you could do.

  Since you should send each change separately, you might as well send it right away. That gives us the option of installing it immediately if it is important.

• Use ‘diff -c’ to make your diffs. Diffs without context are hard for us to install reliably. More than that, they make it hard for us to study the diffs to decide whether we want to install them. Unidiff format is better than contextless diffs, but not as easy to read as ‘-c’ format.

  If you have GNU diff, use ‘diff -cp’, which shows the name of the function that each change occurs in.

• Write the change log entries for your changes. We get lots of changes, and we don’t have time to do all the change log writing ourselves.

  Read the ‘ChangeLog’ file to see what sorts of information to put in, and to learn the style that we use. The purpose of the change log is to show people where to find what was changed. So you need to be specific about what functions you changed; in large functions, it’s often helpful to indicate where within the function the change was.

  On the other hand, once you have shown people where to find the change, you need not explain its purpose. Thus, if you add a new function, all you need to say about it is that it is new. If you feel that the purpose needs explaining, it probably does—but the explanation will be much more useful if you put it in comments in the code.

  If you would like your name to appear in the header line for who made the change, send us the header line.

• When you write the fix, keep in mind that we can’t install a change that would break other systems.

  People often suggest fixing a problem by changing machine-independent files such as ‘toplev.c’ to do something special that a particular system needs. Sometimes it is totally obvious that such changes would break GNU CC for almost all users. We can’t possibly make a change like that. At best it might tell us how to write another patch that would solve the problem acceptably.

  Sometimes people send fixes that might be an improvement in general—but it is hard to be sure of this. It’s hard to install such changes because we have to study them very carefully. Of course, a good explanation of the reasoning by which you concluded the change was correct can help convince us.

  The safest changes are changes to the configuration files for a particular machine. These are safe because they can’t create new bugs on other machines.

  Please help us keep up with the workload by designing the patch in a form that is good to install.

4 How To Get Help with GNU CC

If you need help installing, using or changing GNU CC, there are two ways to find it:

• Send a message to a suitable network mailing list. First try bug-gcc@prep.ai.mit.edu, and if that brings no response, try help-gcc@prep.ai.mit.edu.
• Look in the service directory for someone who might help you for a fee. The service directory is found in the file named ‘SERVICE’ in the GNU CC distribution.

5 Using GNU CC on VMS

5.1 Include Files and VMS

Due to the differences between the filesystems of Unix and VMS, GNU CC attempts to translate file names in ‘#include’ into names that VMS will understand. The basic strategy is to prepend a prefix to the specification of the include file, convert the whole filename to a VMS filename, and then try to open the file. GNU CC tries various prefixes one by one until one of them succeeds:

1. The first prefix is the ‘GNU_CC_INCLUDE:’ logical name: this is where GNU C header files are traditionally stored. If you wish to store header files in non-standard locations, then you can assign the logical ‘GNU_CC_INCLUDE’ to be a search list, where each element of the list is suitable for use with a rooted logical.
2. The next prefix tried is ‘SYS$SYSROOT:[SYSLIB.]’. This is where VAX-C header files are traditionally stored.
3. If the include file specification by itself is a valid VMS filename, the preprocessor then uses this name with no prefix in an attempt to open the include file.
4. If the file specification is not a valid VMS filename (i.e. does not contain a device or a directory specifier, and contains a ‘/’ character), the preprocessor tries to convert it from Unix syntax to VMS syntax.

   Conversion works like this: the first directory name becomes a device, and the rest of the directories are converted into VMS-format directory names. For example, ‘X11/foobar.h’ is translated to ‘X11:[000000]foobar.h’ or ‘X11:foobar.h’, whichever one can be opened. This strategy allows you to assign a logical name to point to the actual location of the header files.

5. If none of these strategies succeeds, the ‘#include’ fails.

Include directives of the form:

#include foobar

are a common source of incompatibility between VAX-C and GNU CC. VAX-C treats this much like a standard #include <foobar.h> directive. That is incompatible with the ANSI C behavior implemented by GNU CC: to expand the name foobar as a macro. Macro expansion should eventually yield one of the two standard formats for #include:

#include "file"
#include <file>

If you have this problem, the best solution is to modify the source to convert the #include directives to one of the two standard forms. That will work with either compiler. If you want a quick and dirty fix, define the file names as macros with the proper expansion, like this:

#define stdio <stdio.h>

This will work, as long as the name doesn’t conflict with anything else in the program.

Another source of incompatibility is that VAX-C assumes that:

#include "foobar"

is actually asking for the file ‘foobar.h’. GNU CC does not make this assumption, and instead takes what you ask for literally; it tries to read the file ‘foobar’. The best way to avoid this problem is to always specify the desired file extension in your include directives.

GNU CC for VMS is distributed with a set of include files that is sufficient to compile most general purpose programs. Even though the GNU CC distribution does not contain header files to define constants and structures for some VMS system-specific functions, there is no reason why you cannot use GNU CC with any of these functions. You first may have to generate or create header files, either by using the public domain utility UNSDL (which can be found on a DECUS tape), or by extracting the relevant modules from one of the system macro libraries, and using an editor to construct a C header file.

A #include file name cannot contain a DECNET node name. The preprocessor reports an I/O error if you attempt to use a node name, whether explicitly, or implicitly via a logical name.

5.2 Global Declarations and VMS

GNU CC does not provide the globalref, globaldef and globalvalue keywords of VAX-C. You can get the same effect with an obscure feature of GAS, the GNU assembler. (This requires GAS version 1.39 or later.) The following macros allow you to use this feature in a fairly natural way:

#ifdef __GNUC__
#define GLOBALREF(TYPE,NAME)                      \
  TYPE NAME                                       \
  asm ("_$$PsectAttributes_GLOBALSYMBOL$$" #NAME)
#define GLOBALDEF(TYPE,NAME,VALUE)                \
  TYPE NAME                                       \
  asm ("_$$PsectAttributes_GLOBALSYMBOL$$" #NAME) \
    = VALUE
#define GLOBALVALUEREF(TYPE,NAME)                 \
  const TYPE NAME[1]                              \     
  asm ("_$$PsectAttributes_GLOBALVALUE$$" #NAME)
#define GLOBALVALUEDEF(TYPE,NAME,VALUE)           \
  const TYPE NAME[1]                              \
  asm ("_$$PsectAttributes_GLOBALVALUE$$" #NAME)  \
    = {VALUE}
#else
#define GLOBALREF(TYPE,NAME) \
  globalref TYPE NAME
#define GLOBALDEF(TYPE,NAME,VALUE) \
  globaldef TYPE NAME = VALUE
#define GLOBALVALUEDEF(TYPE,NAME,VALUE) \
  globalvalue TYPE NAME = VALUE
#define GLOBALVALUEREF(TYPE,NAME) \
  globalvalue TYPE NAME
#endif

(The _$$PsectAttributes_GLOBALSYMBOL prefix at the start of the name is removed by the assembler, after it has modified the attributes of the symbol). These macros are provided in the VMS binaries distribution in a header file ‘GNU_HACKS.H’. An example of the usage is:

GLOBALREF (int, ijk);
GLOBALDEF (int, jkl, 0);

The macros GLOBALREF and GLOBALDEF cannot be used straightforwardly for arrays, since there is no way to insert the array dimension into the declaration at the right place. However, you can declare an array with these macros if you first define a typedef for the array type, like this:

typedef int intvector[10];
GLOBALREF (intvector, foo);

Array and structure initializers will also break the macros; you can define the initializer to be a macro of its own, or you can expand the GLOBALDEF macro by hand. You may find a case where you wish to use the GLOBALDEF macro with a large array, but you are not interested in explicitly initializing each element of the array. In such cases you can use an initializer like: {0,}, which will initialize the entire array to 0.

A shortcoming of this implementation is that a variable declared with GLOBALVALUEREF or GLOBALVALUEDEF is always an array. For example, the declaration:

GLOBALVALUEREF(int, ijk);

declares the variable ijk as an array of type int [1]. This is done because a globalvalue is actually a constant; its “value” is what the linker would normally consider an address. That is not how an integer value works in C, but it is how an array works. So treating the symbol as an array name gives consistent results—with the exception that the value seems to have the wrong type. Don’t try to access an element of the array. It doesn’t have any elements. The array “address” may not be the address of actual storage.

The fact that the symbol is an array may lead to warnings where the variable is used. Insert type casts to avoid the warnings. Here is an example; it takes advantage of the ANSI C feature allowing macros that expand to use the same name as the macro itself.

GLOBALVALUEREF (int, ss$_normal);
GLOBALVALUEDEF (int, xyzzy,123);
#ifdef __GNUC__
#define ss$_normal ((int) ss$_normal)
#define xyzzy ((int) xyzzy)
#endif

Don’t use globaldef or globalref with a variable whose type is an enumeration type; this is not implemented. Instead, make the variable an integer, and use a globalvaluedef for each of the enumeration values. An example of this would be:

#ifdef __GNUC__
GLOBALDEF (int, color, 0);
GLOBALVALUEDEF (int, RED, 0);
GLOBALVALUEDEF (int, BLUE, 1);
GLOBALVALUEDEF (int, GREEN, 3);
#else
enum globaldef color {RED, BLUE, GREEN = 3};
#endif

5.3 Other VMS Issues

GNU CC automatically arranges for main to return 1 by default if you fail to specify an explicit return value. This will be interpreted by VMS as a status code indicating a normal successful completion. Version 1 of GNU CC did not provide this default.

GNU CC on VMS works only with the GNU assembler, GAS. You need version 1.37 or later of GAS in order to produce value debugging information for the VMS debugger. Use the ordinary VMS linker with the object files produced by GAS.

Under previous versions of GNU CC, the generated code would occasionally give strange results when linked to the sharable ‘VAXCRTL’ library. Now this should work.

A caveat for use of const global variables: the const modifier must be specified in every external declaration of the variable in all of the source files that use that variable. Otherwise the linker will issue warnings about conflicting attributes for the variable. Your program will still work despite the warnings, but the variable will be placed in writable storage.

Although the VMS linker does distinguish between upper and lower case letters in global symbols, most VMS compilers convert all such symbols into upper case and most run-time library routines also have upper case names. To be able to reliably call such routines, GNU CC (by means of the assembler GAS) converts global symbols into upper case like other VMS compilers. However, since the usual practice in C is to distinguish case, GNU CC (via GAS) tries to preserve usual C behavior by augmenting each name that is not all lower case. This means truncating the name to at most 23 characters and then adding more characters at the end which encode the case pattern of those 23. Names which contain at least one dollar sign are an exception; they are converted directly into upper case without augmentation.

Name augmentation yields bad results for programs that use precompiled libraries (such as Xlib) which were generated by another compiler. You can use the compiler option ‘/NOCASE_HACK’ to inhibit augmentation; it makes external C functions and variables case-independent as is usual on VMS. Alternatively, you could write all references to the functions and variables in such libraries using lower case; this will work on VMS, but is not portable to other systems. The compiler option ‘/NAMES’ also provides control over global name handling.

Function and variable names are handled somewhat differently with GNU C++. The GNU C++ compiler performs name mangling on function names, which means that it adds information to the function name to describe the data types of the arguments that the function takes. One result of this is that the name of a function can become very long. Since the VMS linker only recognizes the first 31 characters in a name, special action is taken to ensure that each function and variable has a unique name that can be represented in 31 characters.

If the name (plus a name augmentation, if required) is less than 32 characters in length, then no special action is performed. If the name is longer than 31 characters, the assembler (GAS) will generate a hash string based upon the function name, truncate the function name to 23 characters, and append the hash string to the truncated name. If the ‘/VERBOSE’ compiler option is used, the assembler will print both the full and truncated names of each symbol that is truncated.

The ‘/NOCASE_HACK’ compiler option should not be used when you are compiling programs that use libg++. libg++ has several instances of objects (i.e. Filebuf and filebuf) which become indistinguishable in a case-insensitive environment. This leads to cases where you need to inhibit augmentation selectively (if you were using libg++ and Xlib in the same program, for example). There is no special feature for doing this, but you can get the result by defining a macro for each mixed case symbol for which you wish to inhibit augmentation. The macro should expand into the lower case equivalent of itself. For example:

#define StuDlyCapS studlycaps

These macro definitions can be placed in a header file to minimize the number of changes to your source code.

6 GNU CC and Portability

The main goal of GNU CC was to make a good, fast compiler for machines in the class that the GNU system aims to run on: 32-bit machines that address 8-bit bytes and have several general registers. Elegance, theoretical power and simplicity are only secondary.

GNU CC gets most of the information about the target machine from a machine description which gives an algebraic formula for each of the machine’s instructions. This is a very clean way to describe the target. But when the compiler needs information that is difficult to express in this fashion, I have not hesitated to define an ad-hoc parameter to the machine description. The purpose of portability is to reduce the total work needed on the compiler; it was not of interest for its own sake.

GNU CC does not contain machine dependent code, but it does contain code that depends on machine parameters such as endianness (whether the most significant byte has the highest or lowest address of the bytes in a word) and the availability of autoincrement addressing. In the RTL-generation pass, it is often necessary to have multiple strategies for generating code for a particular kind of syntax tree, strategies that are usable for different combinations of parameters. Often I have not tried to address all possible cases, but only the common ones or only the ones that I have encountered. As a result, a new target may require additional strategies. You will know if this happens because the compiler will call abort. Fortunately, the new strategies can be added in a machine-independent fashion, and will affect only the target machines that need them.

7 Interfacing to GNU CC Output

GNU CC is normally configured to use the same function calling convention normally in use on the target system. This is done with the machine-description macros described (@pxref{Target Macros}).

However, returning of structure and union values is done differently on some target machines. As a result, functions compiled with PCC returning such types cannot be called from code compiled with GNU CC, and vice versa. This does not cause trouble often because few Unix library routines return structures or unions.

GNU CC code returns structures and unions that are 1, 2, 4 or 8 bytes long in the same registers used for int or double return values. (GNU CC typically allocates variables of such types in registers also.) Structures and unions of other sizes are returned by storing them into an address passed by the caller (usually in a register). The machine-description macros STRUCT_VALUE and STRUCT_INCOMING_VALUE tell GNU CC where to pass this address.

By contrast, PCC on most target machines returns structures and unions of any size by copying the data into an area of static storage, and then returning the address of that storage as if it were a pointer value. The caller must copy the data from that memory area to the place where the value is wanted. This is slower than the method used by GNU CC, and fails to be reentrant.

On some target machines, such as RISC machines and the 80386, the standard system convention is to pass to the subroutine the address of where to return the value. On these machines, GNU CC has been configured to be compatible with the standard compiler, when this method is used. It may not be compatible for structures of 1, 2, 4 or 8 bytes.

GNU CC uses the system’s standard convention for passing arguments. On some machines, the first few arguments are passed in registers; in others, all are passed on the stack. It would be possible to use registers for argument passing on any machine, and this would probably result in a significant speedup. But the result would be complete incompatibility with code that follows the standard convention. So this change is practical only if you are switching to GNU CC as the sole C compiler for the system. We may implement register argument passing on certain machines once we have a complete GNU system so that we can compile the libraries with GNU CC.

On some machines (particularly the Sparc), certain types of arguments are passed “by invisible reference”. This means that the value is stored in memory, and the address of the memory location is passed to the subroutine.

If you use longjmp, beware of automatic variables. ANSI C says that automatic variables that are not declared volatile have undefined values after a longjmp. And this is all GNU CC promises to do, because it is very difficult to restore register variables correctly, and one of GNU CC’s features is that it can put variables in registers without your asking it to.

If you want a variable to be unaltered by longjmp, and you don’t want to write volatile because old C compilers don’t accept it, just take the address of the variable. If a variable’s address is ever taken, even if just to compute it and ignore it, then the variable cannot go in a register:

{
  int careful;
  &careful;
  …
}

Code compiled with GNU CC may call certain library routines. Most of them handle arithmetic for which there are no instructions. This includes multiply and divide on some machines, and floating point operations on any machine for which floating point support is disabled with ‘-msoft-float’. Some standard parts of the C library, such as bcopy or memcpy, are also called automatically. The usual function call interface is used for calling the library routines.

These library routines should be defined in the library ‘libgcc.a’, which GNU CC automatically searches whenever it links a program. On machines that have multiply and divide instructions, if hardware floating point is in use, normally ‘libgcc.a’ is not needed, but it is searched just in case.

Each arithmetic function is defined in ‘libgcc1.c’ to use the corresponding C arithmetic operator. As long as the file is compiled with another C compiler, which supports all the C arithmetic operators, this file will work portably. However, ‘libgcc1.c’ does not work if compiled with GNU CC, because each arithmetic function would compile into a call to itself!

8 Passes and Files of the Compiler

The overall control structure of the compiler is in ‘toplev.c’. This file is responsible for initialization, decoding arguments, opening and closing files, and sequencing the passes.

The parsing pass is invoked only once, to parse the entire input. The RTL intermediate code for a function is generated as the function is parsed, a statement at a time. Each statement is read in as a syntax tree and then converted to RTL; then the storage for the tree for the statement is reclaimed. Storage for types (and the expressions for their sizes), declarations, and a representation of the binding contours and how they nest, remain until the function is finished being compiled; these are all needed to output the debugging information.

Each time the parsing pass reads a complete function definition or top-level declaration, it calls the function rest_of_compilation or rest_of_decl_compilation in ‘toplev.c’, which are responsible for all further processing necessary, ending with output of the assembler language. All other compiler passes run, in sequence, within rest_of_compilation. When that function returns from compiling a function definition, the storage used for that function definition’s compilation is entirely freed, unless it is an inline function (@pxref{Inline}).

Here is a list of all the passes of the compiler and their source files. Also included is a description of where debugging dumps can be requested with ‘-d’ options.

• Parsing. This pass reads the entire text of a function definition, constructing partial syntax trees. This and RTL generation are no longer truly separate passes (formerly they were), but it is easier to think of them as separate.

  The tree representation does not entirely follow C syntax, because it is intended to support other languages as well.

  Language-specific data type analysis is also done in this pass, and every tree node that represents an expression has a data type attached. Variables are represented as declaration nodes.

  Constant folding and some arithmetic simplifications are also done during this pass.

  The language-independent source files for parsing are ‘stor-layout.c’, ‘fold-const.c’, and ‘tree.c’. There are also header files ‘tree.h’ and ‘tree.def’ which define the format of the tree representation.

  The source files for parsing C are ‘c-parse.y’, ‘c-decl.c’, ‘c-typeck.c’, ‘c-convert.c’, ‘c-lang.c’, and ‘c-aux-info.c’ along with header files ‘c-lex.h’, and ‘c-tree.h’.

  The source files for parsing C++ are ‘cp-parse.y’, ‘cp-class.c’,
  ‘cp-cvt.c’, ‘cp-decl.c’, ‘cp-decl2.c’, ‘cp-dem.c’, ‘cp-except.c’,
  ‘cp-expr.c’, ‘cp-init.c’, ‘cp-lex.c’, ‘cp-method.c’, ‘cp-ptree.c’,
  ‘cp-search.c’, ‘cp-tree.c’, ‘cp-type2.c’, and ‘cp-typeck.c’, along with header files ‘cp-tree.def’, ‘cp-tree.h’, and ‘cp-decl.h’.

  The special source files for parsing Objective C are ‘objc-parse.y’, ‘objc-actions.c’, ‘objc-tree.def’, and ‘objc-actions.h’. Certain C-specific files are used for this as well.

  The file ‘c-common.c’ is also used for all of the above languages.

• RTL generation. This is the conversion of syntax tree into RTL code. It is actually done statement-by-statement during parsing, but for most purposes it can be thought of as a separate pass.

  This is where the bulk of target-parameter-dependent code is found, since often it is necessary for strategies to apply only when certain standard kinds of instructions are available. The purpose of named instruction patterns is to provide this information to the RTL generation pass.

  Optimization is done in this pass for if-conditions that are comparisons, boolean operations or conditional expressions. Tail recursion is detected at this time also. Decisions are made about how best to arrange loops and how to output switch statements.

  The source files for RTL generation include ‘stmt.c’, ‘function.c’, ‘expr.c’, ‘calls.c’, ‘explow.c’, ‘expmed.c’, ‘optabs.c’ and ‘emit-rtl.c’. Also, the file ‘insn-emit.c’, generated from the machine description by the program genemit, is used in this pass. The header file ‘expr.h’ is used for communication within this pass.

  The header files ‘insn-flags.h’ and ‘insn-codes.h’, generated from the machine description by the programs genflags and gencodes, tell this pass which standard names are available for use and which patterns correspond to them.

  Aside from debugging information output, none of the following passes refers to the tree structure representation of the function (only part of which is saved).

  The decision of whether the function can and should be expanded inline in its subsequent callers is made at the end of rtl generation. The function must meet certain criteria, currently related to the size of the function and the types and number of parameters it has. Note that this function may contain loops, recursive calls to itself (tail-recursive functions can be inlined!), gotos, in short, all constructs supported by GNU CC. The file ‘integrate.c’ contains the code to save a function’s rtl for later inlining and to inline that rtl when the function is called. The header file ‘integrate.h’ is also used for this purpose.

  The option ‘-dr’ causes a debugging dump of the RTL code after this pass. This dump file’s name is made by appending ‘.rtl’ to the input file name.

• Jump optimization. This pass simplifies jumps to the following instruction, jumps across jumps, and jumps to jumps. It deletes unreferenced labels and unreachable code, except that unreachable code that contains a loop is not recognized as unreachable in this pass. (Such loops are deleted later in the basic block analysis.) It also converts some code originally written with jumps into sequences of instructions that directly set values from the results of comparisons, if the machine has such instructions.

  Jump optimization is performed two or three times. The first time is immediately following RTL generation. The second time is after CSE, but only if CSE says repeated jump optimization is needed. The last time is right before the final pass. That time, cross-jumping and deletion of no-op move instructions are done together with the optimizations described above.

  The source file of this pass is ‘jump.c’.

  The option ‘-dj’ causes a debugging dump of the RTL code after this pass is run for the first time. This dump file’s name is made by appending ‘.jump’ to the input file name.

• Register scan. This pass finds the first and last use of each register, as a guide for common subexpression elimination. Its source is in ‘regclass.c’.
• Jump threading. This pass detects a condition jump that branches to an identical or inverse test. Such jumps can be ‘threaded’ through the second conditional test. The source code for this pass is in ‘jump.c’. This optimization is only performed if ‘-fthread-jumps’ is enabled.
• Common subexpression elimination. This pass also does constant propagation. Its source file is ‘cse.c’. If constant propagation causes conditional jumps to become unconditional or to become no-ops, jump optimization is run again when CSE is finished.

  The option ‘-ds’ causes a debugging dump of the RTL code after this pass. This dump file’s name is made by appending ‘.cse’ to the input file name.

• Loop optimization. This pass moves constant expressions out of loops, and optionally does strength-reduction and loop unrolling as well. Its source files are ‘loop.c’ and ‘unroll.c’, plus the header ‘loop.h’ used for communication between them. Loop unrolling uses some functions in ‘integrate.c’ and the header ‘integrate.h’.

  The option ‘-dL’ causes a debugging dump of the RTL code after this pass. This dump file’s name is made by appending ‘.loop’ to the input file name.

• If ‘-frerun-cse-after-loop’ was enabled, a second common subexpression elimination pass is performed after the loop optimization pass. Jump threading is also done again at this time if it was specified.

  The option ‘-dt’ causes a debugging dump of the RTL code after this pass. This dump file’s name is made by appending ‘.cse2’ to the input file name.

• Stupid register allocation is performed at this point in a nonoptimizing compilation. It does a little data flow analysis as well. When stupid register allocation is in use, the next pass executed is the reloading pass; the others in between are skipped. The source file is ‘stupid.c’.
• Data flow analysis (‘flow.c’). This pass divides the program into basic blocks (and in the process deletes unreachable loops); then it computes which pseudo-registers are live at each point in the program, and makes the first instruction that uses a value point at the instruction that computed the value.

  This pass also deletes computations whose results are never used, and combines memory references with add or subtract instructions to make autoincrement or autodecrement addressing.

  The option ‘-df’ causes a debugging dump of the RTL code after this pass. This dump file’s name is made by appending ‘.flow’ to the input file name. If stupid register allocation is in use, this dump file reflects the full results of such allocation.

• Instruction combination (‘combine.c’). This pass attempts to combine groups of two or three instructions that are related by data flow into single instructions. It combines the RTL expressions for the instructions by substitution, simplifies the result using algebra, and then attempts to match the result against the machine description.

  The option ‘-dc’ causes a debugging dump of the RTL code after this pass. This dump file’s name is made by appending ‘.combine’ to the input file name.

• Instruction scheduling (‘sched.c’). This pass looks for instructions whose output will not be available by the time that it is used in subsequent instructions. (Memory loads and floating point instructions often have this behavior on RISC machines). It re-orders instructions within a basic block to try to separate the definition and use of items that otherwise would cause pipeline stalls.

  Instruction scheduling is performed twice. The first time is immediately after instruction combination and the second is immediately after reload.

  The option ‘-dS’ causes a debugging dump of the RTL code after this pass is run for the first time. The dump file’s name is made by appending ‘.sched’ to the input file name.

• Register class preferencing. The RTL code is scanned to find out which register class is best for each pseudo register. The source file is ‘regclass.c’.
• Local register allocation (‘local-alloc.c’). This pass allocates hard registers to pseudo registers that are used only within one basic block. Because the basic block is linear, it can use fast and powerful techniques to do a very good job.

  The option ‘-dl’ causes a debugging dump of the RTL code after this pass. This dump file’s name is made by appending ‘.lreg’ to the input file name.

• Global register allocation (‘global.c’). This pass allocates hard registers for the remaining pseudo registers (those whose life spans are not contained in one basic block).
• Reloading. This pass renumbers pseudo registers with the hardware registers numbers they were allocated. Pseudo registers that did not get hard registers are replaced with stack slots. Then it finds instructions that are invalid because a value has failed to end up in a register, or has ended up in a register of the wrong kind. It fixes up these instructions by reloading the problematical values temporarily into registers. Additional instructions are generated to do the copying.

  The reload pass also optionally eliminates the frame pointer and inserts instructions to save and restore call-clobbered registers around calls.

  Source files are ‘reload.c’ and ‘reload1.c’, plus the header ‘reload.h’ used for communication between them.

  The option ‘-dg’ causes a debugging dump of the RTL code after this pass. This dump file’s name is made by appending ‘.greg’ to the input file name.

• Instruction scheduling is repeated here to try to avoid pipeline stalls due to memory loads generated for spilled pseudo registers.

  The option ‘-dR’ causes a debugging dump of the RTL code after this pass. This dump file’s name is made by appending ‘.sched2’ to the input file name.

• Jump optimization is repeated, this time including cross-jumping and deletion of no-op move instructions.

  The option ‘-dJ’ causes a debugging dump of the RTL code after this pass. This dump file’s name is made by appending ‘.jump2’ to the input file name.

• Delayed branch scheduling. This optional pass attempts to find instructions that can go into the delay slots of other instructions, usually jumps and calls. The source file name is ‘reorg.c’.

  The option ‘-dd’ causes a debugging dump of the RTL code after this pass. This dump file’s name is made by appending ‘.dbr’ to the input file name.

• Conversion from usage of some hard registers to usage of a register stack may be done at this point. Currently, this is supported only for the floating-point registers of the Intel 80387 coprocessor. The source file name is ‘reg-stack.c’.

  The options ‘-dk’ causes a debugging dump of the RTL code after this pass. This dump file’s name is made by appending ‘.stack’ to the input file name.

• Final. This pass outputs the assembler code for the function. It is also responsible for identifying spurious test and compare instructions. Machine-specific peephole optimizations are performed at the same time. The function entry and exit sequences are generated directly as assembler code in this pass; they never exist as RTL.

  The source files are ‘final.c’ plus ‘insn-output.c’; the latter is generated automatically from the machine description by the tool ‘genoutput’. The header file ‘conditions.h’ is used for communication between these files.

• Debugging information output. This is run after final because it must output the stack slot offsets for pseudo registers that did not get hard registers. Source files are ‘dbxout.c’ for DBX symbol table format, ‘sdbout.c’ for SDB symbol table format, and ‘dwarfout.c’ for DWARF symbol table format.

Some additional files are used by all or many passes:

• Every pass uses ‘machmode.def’ and ‘machmode.h’ which define the machine modes.
• Several passes use ‘real.h’, which defines the default representation of floating point constants and how to operate on them.
• All the passes that work with RTL use the header files ‘rtl.h’ and ‘rtl.def’, and subroutines in file ‘rtl.c’. The tools gen* also use these files to read and work with the machine description RTL.
• Several passes refer to the header file ‘insn-config.h’ which contains a few parameters (C macro definitions) generated automatically from the machine description RTL by the tool genconfig.
• Several passes use the instruction recognizer, which consists of ‘recog.c’ and ‘recog.h’, plus the files ‘insn-recog.c’ and ‘insn-extract.c’ that are generated automatically from the machine description by the tools ‘genrecog’ and ‘genextract’.
• Several passes use the header files ‘regs.h’ which defines the information recorded about pseudo register usage, and ‘basic-block.h’ which defines the information recorded about basic blocks.
• ‘hard-reg-set.h’ defines the type HARD_REG_SET, a bit-vector with a bit for each hard register, and some macros to manipulate it. This type is just int if the machine has few enough hard registers; otherwise it is an array of int and some of the macros expand into loops.
• Several passes use instruction attributes. A definition of the attributes defined for a particular machine is in file ‘insn-attr.h’, which is generated from the machine description by the program ‘genattr’. The file ‘insn-attrtab.c’ contains subroutines to obtain the attribute values for insns. It is generated from the machine description by the program ‘genattrtab’.

9 The Configuration File

The configuration file ‘xm-machine.h’ contains macro definitions that describe the machine and system on which the compiler is running, unlike the definitions in ‘machine.h’, which describe the machine for which the compiler is producing output. Most of the values in ‘xm-machine.h’ are actually the same on all machines that GNU CC runs on, so large parts of all configuration files are identical. But there are some macros that vary:

USG

Define this macro if the host system is System V.

VMS

Define this macro if the host system is VMS.

FAILURE_EXIT_CODE

A C expression for the status code to be returned when the compiler exits after serious errors.

SUCCESS_EXIT_CODE

A C expression for the status code to be returned when the compiler exits without serious errors.

HOST_WORDS_BIG_ENDIAN

Defined if the host machine stores words of multi-word values in big-endian order. (GNU CC does not depend on the host byte ordering within a word.)

HOST_FLOAT_FORMAT

A numeric code distinguishing the floating point format for the host machine. See TARGET_FLOAT_FORMAT in @ref{Storage Layout} for the alternatives and default.

HOST_BITS_PER_CHAR

A C expression for the number of bits in char on the host machine.

HOST_BITS_PER_SHORT

A C expression for the number of bits in short on the host machine.

HOST_BITS_PER_INT

A C expression for the number of bits in int on the host machine.

HOST_BITS_PER_LONG

A C expression for the number of bits in long on the host machine.

ONLY_INT_FIELDS

Define this macro to indicate that the host compiler only supports int bit fields, rather than other integral types, including enum, as do most C compilers.

EXECUTABLE_SUFFIX

Define this macro if the host system uses a naming convention for executable files that involves a common suffix (such as, in some systems, ‘.exe’) that must be mentioned explicitly when you run the program.

OBSTACK_CHUNK_SIZE

A C expression for the size of ordinary obstack chunks. If you don’t define this, a usually-reasonable default is used.

OBSTACK_CHUNK_ALLOC

The function used to allocate obstack chunks. If you don’t define this, xmalloc is used.

OBSTACK_CHUNK_FREE

The function used to free obstack chunks. If you don’t define this, free is used.

USE_C_ALLOCA

Define this macro to indicate that the compiler is running with the alloca implemented in C. This version of alloca can be found in the file ‘alloca.c’; to use it, you must also alter the ‘Makefile’ variable ALLOCA. (This is done automatically for the systems on which we know it is needed.)

If you do define this macro, you should probably do it as follows:

#ifndef __GNUC__
#define USE_C_ALLOCA
#else
#define alloca __builtin_alloca
#endif

so that when the compiler is compiled with GNU CC it uses the more efficient built-in alloca function.

FUNCTION_CONVERSION_BUG

Define this macro to indicate that the host compiler does not properly handle converting a function value to a pointer-to-function when it is used in an expression.

HAVE_VPRINTF

Define this if the library function vprintf is available on your system.

MULTIBYTE_CHARS

Define this macro to enable support for multibyte characters in the input to GNU CC. This requires that the host system support the ANSI C library functions for converting multibyte characters to wide characters.

HAVE_PUTENV

Define this if the library function putenv is available on your system.

NO_SYS_SIGLIST

Define this if your system does not provide the variable sys_siglist.

USE_PROTOTYPES

Define this to be 1 if you know that the host compiler supports prototypes, even if it doesn’t define __STDC__, or define it to be 0 if you do not want any prototypes used in compiling GNU CC. If ‘USE_PROTOTYPES’ is not defined, it will be determined automatically whether your compiler supports prototypes by checking if ‘__STDC__’ is defined.

NO_MD_PROTOTYPES

Define this if you wish suppression of prototypes generated from the machine description file, but to use other prototypes within GNU CC. If ‘USE_PROTOTYPES’ is defined to be 0, or the host compiler does not support prototypes, this macro has no effect.

MD_CALL_PROTOTYPES

Define this if you wish to generate prototypes for the gen_call or gen_call_value functions generated from the machine description file. If ‘USE_PROTOTYPES’ is defined to be 0, or the host compiler does not support prototypes, or ‘NO_MD_PROTOTYPES’ is defined, this macro has no effect. As soon as all of the machine descriptions are modified to have the appropriate number of arguments, this macro will be removed.

Some systems do provide this variable, but with a different name such as _sys_siglist. On these systems, you can define sys_siglist as a macro which expands into the name actually provided.

NO_STAB_H

Define this if your system does not have the include file ‘stab.h’. If ‘USG’ is defined, ‘NO_STAB_H’ is assumed.

In addition, configuration files for system V define bcopy, bzero and bcmp as aliases. Some files define alloca as a macro when compiled with GNU CC, in order to take advantage of the benefit of GNU CC’s built-in alloca.

Index

Table of Contents

About This Document

This document was generated on September 19, 2022 using texi2html 5.0.

The buttons in the navigation panels have the following meaning:

where the Example assumes that the current position is at Subsubsection One-Two-Three of a document of the following structure:

• 1. Section One
  • 1.1 Subsection One-One
    • ...
  • 1.2 Subsection One-Two
    • 1.2.1 Subsubsection One-Two-One
    • 1.2.2 Subsubsection One-Two-Two
    • 1.2.3 Subsubsection One-Two-Three     <== Current Position
    • 1.2.4 Subsubsection One-Two-Four
  • 1.3 Subsection One-Three
    • ...
  • 1.4 Subsection One-Four

This document was generated on September 19, 2022 using texi2html 5.0.